<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Home on Blogito, ergo sum.</title><link>https://sebastiansemper.de/</link><description>Recent content in Home on Blogito, ergo sum.</description><generator>Hugo</generator><language>de-de</language><copyright>Sebastian R. Semper</copyright><atom:link href="https://sebastiansemper.de/index.xml" rel="self" type="application/rss+xml"/><item><title>Die Dusche</title><link>https://sebastiansemper.de/posts/260128_dusche/</link><pubDate>Wed, 28 Jan 2026 13:46:48 +0100</pubDate><guid>https://sebastiansemper.de/posts/260128_dusche/</guid><description>&lt;p&gt;Es fing damit an, dass Er nach der Gartenarbeit und nach der schlussendlichen Reinigung der elektrischen Kettensäge eine Menge Öl an Seinen Fingern kleben hatte.
Auch unter der Dusche hielten sich einige wenige dieser Flecken noch hartnäckig, selbst nach einigem unnachgiebigem Schrubben.
Eigentlich war Er nie vom übermäßig reinlichen Typ gewesen.
Das hier war aber etwas, das Seine Aufmerksamkeit verlangte.
Das Öl musste weg.&lt;/p&gt;</description></item><item><title>Mein Brot</title><link>https://sebastiansemper.de/posts/brot/</link><pubDate>Thu, 11 Dec 2025 20:31:16 +0100</pubDate><guid>https://sebastiansemper.de/posts/brot/</guid><description>&lt;p&gt;Ich habe über die Jahre &amp;ldquo;mein&amp;rdquo; Brotrezept aus anderen Büchern, von Freunden, dem Internet und meinen auch eigenen Erfahrungen zusammen gefrankensteint.
Dabei ist mir zwar einerseits der Geschmack wichtig, aber auch, dass es sich unkompliziert vorbereiten lässt.
Auch am Backtag muss man das neben der Arbeit im Homeoffice hinbekommen.
Es gibt nichts schlimmeres, als sich vom Brot Backen das Wochenende vermiesen zu lassen.
Es sein denn, man will es.
Ich bin kein Bäcker, ich will keiner werden, ich werd nie einer sein.
Aber ich habe die Mehlschwämme aus dem Supermarkt satt und backe quasi nur, weil ich es muss.&lt;/p&gt;</description></item><item><title>Stockholm: Nordisches Museum</title><link>https://sebastiansemper.de/museums/2025_04_04_stockholm_nordic/</link><pubDate>Fri, 04 Apr 2025 12:39:11 +0200</pubDate><guid>https://sebastiansemper.de/museums/2025_04_04_stockholm_nordic/</guid><description>&lt;ul&gt;
&lt;li&gt;500 Jahre Leben im Norden&lt;/li&gt;
&lt;li&gt;Grönland, Sami, Island, Schweden, Norwegen&lt;/li&gt;
&lt;li&gt;Wie funktioniert Missioniernug von Wikingern?&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Stockholm: Nationalmuseum</title><link>https://sebastiansemper.de/museums/2025_04_03_stockholm_national/</link><pubDate>Thu, 03 Apr 2025 12:39:11 +0200</pubDate><guid>https://sebastiansemper.de/museums/2025_04_03_stockholm_national/</guid><description>&lt;ul&gt;
&lt;li&gt;Kunst vom 16. Jhd an&lt;/li&gt;
&lt;li&gt;Original Bild von Martin Luther&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Estimating Gender Equality</title><link>https://sebastiansemper.de/posts/equality/</link><pubDate>Sun, 23 Mar 2025 19:23:45 +0100</pubDate><guid>https://sebastiansemper.de/posts/equality/</guid><description>&lt;h2 id="modeling"&gt;Modeling&lt;/h2&gt;
&lt;p&gt;Assume we have data about job application procedures \(P_i\), with \(i \leqslant N \in \mathbb{N}\), where we have data about the number of male \(M_i\) and \(F_i\) applicants and if a male applicant was offered a job \(P_i = 0\), or a female contestant, i.e., \(P_i = 1\).
Now, we wish to model the influence of a gender bias \(\rho \in [0,1]\), where \(\rho = 0\) means absolute bias towards men, so \(P_i=0\) for all \(i\), and \(\rho=1\) would mean absolute bias in favor of female applicants.
They way \(\rho\) works is that if we have exactly two applicants, one male, one female, we select the male applicant with probability \((1-\rho)\) and the female applicant with probability \(\rho\).
Based on this and the assumption that all considered job applicants are absolutely equally qualified, we can model the density \(f_i\) of \(P_i\) as
$$
f_i(n, \rho) =
\delta_0(n) \cdot \frac{(1-\rho) M_i}{(1-\rho)M_i + \rho F_i} +
\delta_1(n) \cdot \frac{\rho F_i}{(1-\rho)M_i + \rho F_i},
$$
which essentially generalizes above two-applicant scenario to arbitrary number of applicants.
Now, since we have real worl data of the past, so \(P_i = x_i\) together with concrete values for \(M_i\) and \(F_i\), we can turn things around and instead estimate \(\rho\) from these cases.
Since, we have a precise statistical model for our observations, we can use the maximum likelihood approach over the joint density of all observations.
Under the assumption, that the job application procedures are mutually independent, we have that
$$
f(\rho \vert x_1, \dots, x_N) = \prod\limits_{i=1}^N f_i(\rho \vert x_i),
$$
which after applying a \(\log\) on both sides and negating gives the negative log-likelihood
$$
\lambda(\rho \vert x_1, \dots, x_N) = -\sum\limits_{i=1}^N \log f_i(\rho \vert x_i).
$$
The value \(\rho^\ast\) for which \( \lambda(\cdot \vert x_1, \dots, x_N) \) is maximized is called the maximum likelihood estimate.&lt;/p&gt;</description></item><item><title>Make it stawp, plz!</title><link>https://sebastiansemper.de/memes/8wso95/</link><pubDate>Sat, 13 Jul 2024 09:17:56 +0200</pubDate><guid>https://sebastiansemper.de/memes/8wso95/</guid><description>&lt;img src="https://sebastiansemper.de/memes/8wso95/8wso95.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Did you take the last One?</title><link>https://sebastiansemper.de/memes/effort/</link><pubDate>Sun, 07 Jul 2024 15:03:25 +0200</pubDate><guid>https://sebastiansemper.de/memes/effort/</guid><description>&lt;img src="https://sebastiansemper.de/memes/effort/effort.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Highway to Convergence</title><link>https://sebastiansemper.de/memes/newtons/</link><pubDate>Sun, 07 Jul 2024 15:03:25 +0200</pubDate><guid>https://sebastiansemper.de/memes/newtons/</guid><description>&lt;img src="https://sebastiansemper.de/memes/newtons/Newtons.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>I still need to know</title><link>https://sebastiansemper.de/memes/in_bed_hyper/</link><pubDate>Sun, 07 Jul 2024 15:03:25 +0200</pubDate><guid>https://sebastiansemper.de/memes/in_bed_hyper/</guid><description>&lt;img src="https://sebastiansemper.de/memes/in_bed_hyper/in_bed_hyper.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>My Tears are your Lube</title><link>https://sebastiansemper.de/memes/5ig9h7/</link><pubDate>Sun, 07 Jul 2024 15:03:25 +0200</pubDate><guid>https://sebastiansemper.de/memes/5ig9h7/</guid><description>&lt;img src="https://sebastiansemper.de/memes/5ig9h7/5ig9h7.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Numpy, I am talking to you!</title><link>https://sebastiansemper.de/memes/neat_python/</link><pubDate>Sun, 07 Jul 2024 15:03:25 +0200</pubDate><guid>https://sebastiansemper.de/memes/neat_python/</guid><description>&lt;img src="https://sebastiansemper.de/memes/neat_python/neat_python.png" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Why even bother?</title><link>https://sebastiansemper.de/memes/yes_no_mse/</link><pubDate>Sun, 07 Jul 2024 15:03:25 +0200</pubDate><guid>https://sebastiansemper.de/memes/yes_no_mse/</guid><description>&lt;img src="https://sebastiansemper.de/memes/yes_no_mse/yes_no_mse.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Don't Shoot the Messenger</title><link>https://sebastiansemper.de/memes/53zhl9/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/53zhl9/</guid><description>&lt;img src="https://sebastiansemper.de/memes/53zhl9/53zhl9.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Editors, beware!</title><link>https://sebastiansemper.de/memes/4b2qfa/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/4b2qfa/</guid><description>&lt;img src="https://sebastiansemper.de/memes/4b2qfa/4b2qfa.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Everyone's Favorite Coworker</title><link>https://sebastiansemper.de/memes/4jecjc/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/4jecjc/</guid><description>&lt;img src="https://sebastiansemper.de/memes/4jecjc/4jecjc.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Fucking Up in just 5 LOC</title><link>https://sebastiansemper.de/memes/43o8cc/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/43o8cc/</guid><description>&lt;img src="https://sebastiansemper.de/memes/43o8cc/43o8cc.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>I am Employee of the Month</title><link>https://sebastiansemper.de/memes/586j10/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/586j10/</guid><description>&lt;img src="https://sebastiansemper.de/memes/586j10/586j10.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Ingorance is Bliss</title><link>https://sebastiansemper.de/memes/5azbby/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/5azbby/</guid><description>&lt;img src="https://sebastiansemper.de/memes/5azbby/5azbby.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>It's 0.4. Really. Shut up.</title><link>https://sebastiansemper.de/memes/52cri8/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/52cri8/</guid><description>&lt;img src="https://sebastiansemper.de/memes/52cri8/52cri8.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>It's just very robust!</title><link>https://sebastiansemper.de/memes/16220587213022745718255791134849/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/16220587213022745718255791134849/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16220587213022745718255791134849/16220587213022745718255791134849.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Just use a Crystal Ball</title><link>https://sebastiansemper.de/memes/43o47n/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/43o47n/</guid><description>&lt;img src="https://sebastiansemper.de/memes/43o47n/43o47n.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Kalman, is it you?</title><link>https://sebastiansemper.de/memes/4t208t/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/4t208t/</guid><description>&lt;img src="https://sebastiansemper.de/memes/4t208t/4t208t.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Me 2022 Colorized</title><link>https://sebastiansemper.de/memes/43sw8g/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/43sw8g/</guid><description>&lt;img src="https://sebastiansemper.de/memes/43sw8g/43sw8g.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>My Secret Strategy</title><link>https://sebastiansemper.de/memes/5f4uvl/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/5f4uvl/</guid><description>&lt;img src="https://sebastiansemper.de/memes/5f4uvl/5f4uvl.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>No raise for you!</title><link>https://sebastiansemper.de/memes/469xuj/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/469xuj/</guid><description>&lt;img src="https://sebastiansemper.de/memes/469xuj/469xuj.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>NVIDIA Stock goes BRRRRT</title><link>https://sebastiansemper.de/memes/4quko1/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/4quko1/</guid><description>&lt;img src="https://sebastiansemper.de/memes/4quko1/4quko1.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Priorities</title><link>https://sebastiansemper.de/memes/16247044796071816141627717206188/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/16247044796071816141627717206188/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16247044796071816141627717206188/16247044796071816141627717206188.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Rinse and Repeat</title><link>https://sebastiansemper.de/memes/16269818967545413844649764743323/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/16269818967545413844649764743323/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16269818967545413844649764743323/16269818967545413844649764743323.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Spongebob understands</title><link>https://sebastiansemper.de/memes/16599032494755683490148210608945/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/16599032494755683490148210608945/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16599032494755683490148210608945/16599032494755683490148210608945.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Stick with the First</title><link>https://sebastiansemper.de/memes/495d7h/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/495d7h/</guid><description>&lt;img src="https://sebastiansemper.de/memes/495d7h/495d7h.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Thank God for Giving Birth to Gauß</title><link>https://sebastiansemper.de/memes/4cn0tm/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/4cn0tm/</guid><description>&lt;img src="https://sebastiansemper.de/memes/4cn0tm/4cn0tm.png" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>The one about MIMO?</title><link>https://sebastiansemper.de/memes/16223606538892030214506715423982/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/16223606538892030214506715423982/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16223606538892030214506715423982/16223606538892030214506715423982.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Watch me</title><link>https://sebastiansemper.de/memes/44j5vm/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/44j5vm/</guid><description>&lt;img src="https://sebastiansemper.de/memes/44j5vm/44j5vm.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>We call it BACKPROP</title><link>https://sebastiansemper.de/memes/53zhwz/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/53zhwz/</guid><description>&lt;img src="https://sebastiansemper.de/memes/53zhwz/53zhwz.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>We need M0re</title><link>https://sebastiansemper.de/memes/52oc0s/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/52oc0s/</guid><description>&lt;img src="https://sebastiansemper.de/memes/52oc0s/52oc0s.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Whole lotta Money for such a Sissy</title><link>https://sebastiansemper.de/memes/16234781425256285653690571710877/</link><pubDate>Sun, 07 Jul 2024 15:03:24 +0200</pubDate><guid>https://sebastiansemper.de/memes/16234781425256285653690571710877/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16234781425256285653690571710877/16234781425256285653690571710877.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>2022a might also work</title><link>https://sebastiansemper.de/memes/16156203373241613262171500966365/</link><pubDate>Sun, 07 Jul 2024 15:03:23 +0200</pubDate><guid>https://sebastiansemper.de/memes/16156203373241613262171500966365/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16156203373241613262171500966365/16156203373241613262171500966365.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>I'm Not complaining</title><link>https://sebastiansemper.de/memes/16217751431956458329165991068047/</link><pubDate>Sun, 07 Jul 2024 15:03:23 +0200</pubDate><guid>https://sebastiansemper.de/memes/16217751431956458329165991068047/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16217751431956458329165991068047/16217751431956458329165991068047.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Still Waiting on the Last One</title><link>https://sebastiansemper.de/memes/16180344128313858441760937097437/</link><pubDate>Sun, 07 Jul 2024 15:03:23 +0200</pubDate><guid>https://sebastiansemper.de/memes/16180344128313858441760937097437/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16180344128313858441760937097437/16180344128313858441760937097437.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>The Paper Grind is Real</title><link>https://sebastiansemper.de/memes/16127311863372551430021292530518/</link><pubDate>Sun, 07 Jul 2024 15:03:23 +0200</pubDate><guid>https://sebastiansemper.de/memes/16127311863372551430021292530518/</guid><description>&lt;img src="https://sebastiansemper.de/memes/16127311863372551430021292530518/16127311863372551430021292530518.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Why be Bothered with Details?</title><link>https://sebastiansemper.de/memes/4cn0zr/</link><pubDate>Sun, 07 Jul 2024 14:55:11 +0200</pubDate><guid>https://sebastiansemper.de/memes/4cn0zr/</guid><description>&lt;img src="https://sebastiansemper.de/memes/4cn0zr/4cn0zr.png" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Wirtinger Calculus FTW</title><link>https://sebastiansemper.de/memes/3xy4lp/</link><pubDate>Sun, 07 Jul 2024 14:55:11 +0200</pubDate><guid>https://sebastiansemper.de/memes/3xy4lp/</guid><description>&lt;img src="https://sebastiansemper.de/memes/3xy4lp/3xy4lp.jpg" alt="&amp;amp;nbsp;"&gt;</description></item><item><title>Testing Derivatives</title><link>https://sebastiansemper.de/posts/testing-derivatives/</link><pubDate>Mon, 06 May 2024 10:08:26 +0100</pubDate><guid>https://sebastiansemper.de/posts/testing-derivatives/</guid><description>&lt;h2 id="introduction"&gt;Introduction&lt;/h2&gt;
&lt;p&gt;There still might be people out there who are actually so insane to implement functions and their derivatives by hand.
By hand I mean without refraining to some &lt;code&gt;autodiff&lt;/code&gt; functionality.
These features are offered by virtually every serious machine learning framework out there and for obvious reasons.
Well performing Backprop is essential for fast learning procedures.
However, and this might come as a shock to you, not everyone is working in modern statistical learning.
So some of us might, for instance, use functions that are not very common in deep learning, like complex valued functions, or transcendental ones.
As such, we might still do some implementations in &lt;code&gt;numpy&lt;/code&gt; or even &lt;code&gt;C&lt;/code&gt;, but we still wish to see, if our function and its supposed derivative actually match.&lt;/p&gt;</description></item><item><title>Seoul: Hangeul Museum</title><link>https://sebastiansemper.de/museums/2024_04_15_seoul_hangeul/</link><pubDate>Mon, 15 Apr 2024 19:51:36 +0200</pubDate><guid>https://sebastiansemper.de/museums/2024_04_15_seoul_hangeul/</guid><description>&lt;ul&gt;
&lt;li&gt;Eigentlich wollten wir da nicht rein&lt;/li&gt;
&lt;li&gt;Der König hat die Schrift selber in die Hand genommen&lt;/li&gt;
&lt;li&gt;Scheint wirklich leicht zu lernen sein.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Seoul: National Museum</title><link>https://sebastiansemper.de/museums/2024_04_15_seoul_national/</link><pubDate>Mon, 15 Apr 2024 19:51:03 +0200</pubDate><guid>https://sebastiansemper.de/museums/2024_04_15_seoul_national/</guid><description>&lt;ul&gt;
&lt;li&gt;Die haben echt nen Problem mit Japan.&lt;/li&gt;
&lt;li&gt;Unklar, wie lange es auch da schon Menschen gibt.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Evaluating Likelihoods</title><link>https://sebastiansemper.de/posts/likelihoods/</link><pubDate>Sat, 09 Mar 2024 10:08:26 +0100</pubDate><guid>https://sebastiansemper.de/posts/likelihoods/</guid><description>&lt;p&gt;I do a lot of work that revolves around model based parameter estimation. In order to fit models to data one can invoke the maximum likelihood principle, as it has nice statistical properties in terms of bias and variance of the resulting estimators. Due to the fact that I mostly deal with measurement noise and other distributions where it is safe to assume that the central limit theorem holds, I am usually concerned with Gaussian random variables. One of the things we have to do during optimization is actually calculating the likelihood&amp;rsquo;s current value and possibly (read: certainly) some partial first- and second-order derivatives. However, in cases when the data itself is high-dimensional, the model for the mean being algebraically more involved or the covariance matrix non-identity, one needs to invest some kind of effort into an efficient implementation.&lt;/p&gt;</description></item><item><title>Maximaler Konsum</title><link>https://sebastiansemper.de/posts/maximaler_konsum/</link><pubDate>Wed, 19 Jul 2023 02:47:00 +0200</pubDate><guid>https://sebastiansemper.de/posts/maximaler_konsum/</guid><description>&lt;p&gt;Im selben Moment als die Ware und deren Konsumption auf der Bühne der Welt entstanden ist, begann auch ihr Fluch uns in ihren Bann zu ziehen. Es scheint, als habe in diesem Augenblick die Menschheit die Kontrolle über einen kleinen Teil von sich selbst verloren. Blickt man in der Geschichte zurück scheint es als hätte uns die Industrialisierung die Lust am Kaufen gebracht. Die freie Produktion von Waren und deren Konsum haben einer nie versiegenden Kreativität Bahn gebrochen.&lt;/p&gt;</description></item><item><title>New Orleans: Museum of Art</title><link>https://sebastiansemper.de/museums/2023_06_17_neworleans_modern/</link><pubDate>Sat, 17 Jun 2023 19:55:27 +0200</pubDate><guid>https://sebastiansemper.de/museums/2023_06_17_neworleans_modern/</guid><description>&lt;ul&gt;
&lt;li&gt;Es war schön kühl.&lt;/li&gt;
&lt;li&gt;Direkt neben einem Park mit Eisenskulpturen&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>New Orleans: Sydney and Walda Besthoff Sculpture Garden</title><link>https://sebastiansemper.de/museums/2023_06_17_neworleans_sculptures/</link><pubDate>Sat, 17 Jun 2023 19:55:18 +0200</pubDate><guid>https://sebastiansemper.de/museums/2023_06_17_neworleans_sculptures/</guid><description>&lt;ul&gt;
&lt;li&gt;Predator is real.&lt;/li&gt;
&lt;li&gt;Kein Plan, wie man bei der Hitze arbeiten kann.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>New Orleans: Jazz Museum</title><link>https://sebastiansemper.de/museums/2023_06_13_neworleans_jazz/</link><pubDate>Tue, 13 Jun 2023 19:53:41 +0200</pubDate><guid>https://sebastiansemper.de/museums/2023_06_13_neworleans_jazz/</guid><description>&lt;ul&gt;
&lt;li&gt;Ich hab echt kein Plan von Jazz.&lt;/li&gt;
&lt;li&gt;Snare auf 2 und 4 kommt aus New Orleans.&lt;/li&gt;
&lt;li&gt;Ich hab gecheckt, dass Schlagzeug, wirklich früher aus ner Marchingband zusammengebaut wurde.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Datenschutz</title><link>https://sebastiansemper.de/datenschutz/</link><pubDate>Sat, 03 Jun 2023 20:20:28 +0200</pubDate><guid>https://sebastiansemper.de/datenschutz/</guid><description>&lt;p&gt;Diese Webseite sammelt keine Daten.&lt;/p&gt;</description></item><item><title>Impressum</title><link>https://sebastiansemper.de/impressum/</link><pubDate>Sat, 03 Jun 2023 20:20:28 +0200</pubDate><guid>https://sebastiansemper.de/impressum/</guid><description>&lt;p&gt;Dr.-Ing. Sebastian Semper&lt;br&gt;
Ehrenbergstr. 24, 98693 Ilmenau, Germany&lt;br&gt;
&lt;a href="mailto:post@sebastiansemper.de"&gt;post@sebastiansemper.de&lt;/a&gt;&lt;/p&gt;</description></item><item><title>New York: Museum of Natural History</title><link>https://sebastiansemper.de/museums/2023_05_14_nyc_natural_history/</link><pubDate>Sun, 14 May 2023 19:59:02 +0200</pubDate><guid>https://sebastiansemper.de/museums/2023_05_14_nyc_natural_history/</guid><description>&lt;ul&gt;
&lt;li&gt;Nachts im Museum&lt;/li&gt;
&lt;li&gt;Ich hab noch nie was größeres gesehen&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>New York: Solomon R. Guggenheim Museum</title><link>https://sebastiansemper.de/museums/2023_05_13_nyc_guggenheim/</link><pubDate>Sat, 13 May 2023 19:59:48 +0200</pubDate><guid>https://sebastiansemper.de/museums/2023_05_13_nyc_guggenheim/</guid><description>&lt;ul&gt;
&lt;li&gt;Gego: Gertrud Louise Goldschmidt. Linien.&lt;/li&gt;
&lt;li&gt;Ich hab zum ersten Mal verstanden, was abstrakte Kunst tun kann.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Philadelphia: Museum of Art</title><link>https://sebastiansemper.de/museums/2023_04_23_philly_art/</link><pubDate>Sun, 23 Apr 2023 20:23:34 +0200</pubDate><guid>https://sebastiansemper.de/museums/2023_04_23_philly_art/</guid><description>&lt;ul&gt;
&lt;li&gt;Direkt vor den Rocky Steps&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Karl Marx: Das Kapital</title><link>https://sebastiansemper.de/posts/mew23/</link><pubDate>Tue, 22 Nov 2022 10:55:48 +0100</pubDate><guid>https://sebastiansemper.de/posts/mew23/</guid><description>&lt;p&gt;Am 18. April 2017 beginnen wir die Lektüre von &lt;a href="https://sebastiansemper.de/literatur/#MEW23"&gt;[MEW23]&lt;/a&gt; mit dem ersten Kapitel. Die Einleitung muss ihres Gegenstandes wegen so vage und empirisch bezogen sein, wie wir es weiter unten vorfinden werden, da der Gegenstand noch nicht entwickelt worden ist, zu welchem das Buch einen Beitrag leisten will.&lt;/p&gt;</description></item><item><title>Friedrich Nietzsche: Jenseits von Gut und Böse</title><link>https://sebastiansemper.de/posts/jvgub/</link><pubDate>Tue, 22 Nov 2022 10:44:15 +0100</pubDate><guid>https://sebastiansemper.de/posts/jvgub/</guid><description>&lt;h3 id="wille-zur-wahrheit"&gt;Wille zur Wahrheit&lt;/h3&gt;
&lt;p&gt;&lt;em&gt;&amp;ldquo;Der Wille zur Wahrheit&amp;rdquo;&lt;/em&gt; ist nach Nietzsche eine bloße Verführung, der die Menschen beziehungsweise die Menschheit verfallen zu sein scheinen. Insbesondere die Philosophen scheinen von der Wahrheit &lt;em&gt;&amp;ldquo;mit Ehrerbietung&amp;rdquo;&lt;/em&gt; zu sprechen. Aristoteles beginnt beispielsweise mit dem Satz &lt;em&gt;&amp;ldquo;Die Menschen streben von &lt;strong&gt;Natur&lt;/strong&gt; aus nach Wissen&amp;rdquo;&lt;/em&gt;, welcher gut veranschaulicht, was N. aussagen möchte.&lt;/p&gt;</description></item><item><title>Wien: Tutanchamun</title><link>https://sebastiansemper.de/museums/2022_11_19_wien_tutanchamun/</link><pubDate>Sat, 19 Nov 2022 20:12:00 +0200</pubDate><guid>https://sebastiansemper.de/museums/2022_11_19_wien_tutanchamun/</guid><description>&lt;ul&gt;
&lt;li&gt;So viel Gold.&lt;/li&gt;
&lt;li&gt;Ich muss nach Ägypten.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Hamburg: Kunsthalle</title><link>https://sebastiansemper.de/museums/2022_10_29_hamburg_kunsthalle/</link><pubDate>Sat, 29 Oct 2022 20:09:01 +0200</pubDate><guid>https://sebastiansemper.de/museums/2022_10_29_hamburg_kunsthalle/</guid><description>&lt;ul&gt;
&lt;li&gt;Männer, die auf Bilder starren.&lt;/li&gt;
&lt;li&gt;Caspar David Friedrich. Das Bild.&lt;/li&gt;
&lt;/ul&gt;</description></item><item><title>Der Wert in 'Das Kapital'</title><link>https://sebastiansemper.de/posts/der-wert-in-das-kapital/</link><pubDate>Mon, 29 Nov 2021 08:31:02 +0000</pubDate><guid>https://sebastiansemper.de/posts/der-wert-in-das-kapital/</guid><description>&lt;h3 id="wert-als-einheit"&gt;Wert als Einheit&lt;/h3&gt;
&lt;p&gt;Wir lesen in der Sitzung weiter in [@Arndt2015][Seite 171f], worin es darum geht die formale Bestimmung der Einheit &amp;ldquo;Wert&amp;rdquo; voranzutreiben.&lt;/p&gt;
&lt;p&gt;&lt;em&gt;&amp;ldquo;Diese Einheit bezeichnet der &amp;lsquo;Wert&amp;rsquo;, ohne daß der Wert als normativ vorausgesetzte Einheitskategorie zu verstehen ist.&amp;rdquo;&lt;/em&gt;&lt;/p&gt;
&lt;p&gt;Ganz im hegelschen Sinne als Einheit gefunden, ergibt sich der Wert als notwendig durch die Analyse der Ware. Der Begriff wird demnach von Marx dialektisch &lt;em&gt;entwickelt&lt;/em&gt; und ist somit kein Konstituiertes oder Vorausgesetztes, sondern Resultat geistiger Reflexion.&lt;/p&gt;</description></item><item><title>Getting Stuck</title><link>https://sebastiansemper.de/posts/getting-stuck/</link><pubDate>Wed, 12 May 2021 14:55:08 +0100</pubDate><guid>https://sebastiansemper.de/posts/getting-stuck/</guid><description>&lt;p&gt;We all know the situation when you are getting stuck at something. And you know what I mean. You just don&amp;rsquo;t make progress towards the goal you are having in spite of you putting your best effort into the situation at hand. And it is effing frustrating the sheeze out of us. And in these situations it is easy to conclude that you are too stupid, the task too hard and everything is rigged against you. And my conclusion to this is: exactly.&lt;/p&gt;</description></item><item><title>Die Natur des Menschen</title><link>https://sebastiansemper.de/posts/die-natur-des-menschen/</link><pubDate>Sun, 01 Nov 2020 10:01:56 +0000</pubDate><guid>https://sebastiansemper.de/posts/die-natur-des-menschen/</guid><description>&lt;p&gt;Im Philosophie Seminar lesen wir gerade &lt;a href="https://sebastiansemper.de/literatur/#Ru"&gt;[Ru]&lt;/a&gt;, worin es um das Verhältnis der Natur und dem Kapitalismus geht. In der Einleitung geht es um das stetig anwachsende Ausmaß der Ausbeutung der Natur durch den Menschen und &amp;ldquo;[d]ie Gründe für dieses Anwachsen einer existenziellen Bedrohung liegen in der Organisation der gesellschaftlichen Produktion&amp;rdquo;, also in der &amp;ldquo;kapitalistischen Produktionsweise&amp;rdquo;.&lt;/p&gt;</description></item><item><title>Die Entstehung der Ewigkeit</title><link>https://sebastiansemper.de/posts/die-entstehung-der-ewigkeit/</link><pubDate>Sun, 19 Jul 2020 14:25:55 +0000</pubDate><guid>https://sebastiansemper.de/posts/die-entstehung-der-ewigkeit/</guid><description>&lt;p&gt;Dass für uns Dinge existieren, ist nicht von der Hand zu weisen. Wir nehmen Dinge wahr und unser Bewusstsein sieht sich mit etwas konfrontiert. Wir haben keine stichhaltigen Gründe davon auszugehen, dass nichts existiert.&lt;/p&gt;
&lt;p&gt;Darüber hinaus haben wir in unseren Methoden im Umgang mit unserer Umwelt gelernt, dass Dinge entstehen und vergehen. Dies trifft auf alles zu, was wir als Phänomene wahrnehmen können. Jede Sache entsteht und vergeht. Jede Idee entsteht und vergeht, Materie und Energie sind in einem ewigen Wechselspiel.&lt;/p&gt;</description></item><item><title>e</title><link>https://sebastiansemper.de/posts/e/</link><pubDate>Mon, 13 Apr 2020 13:32:52 +0000</pubDate><guid>https://sebastiansemper.de/posts/e/</guid><description>&lt;p&gt;&lt;em&gt;Es geht im Mythos nicht so sehr um die Formulierung einer Wahrheit, sondern viel mehr um die erzwungene Ordnung von Gedanken, die sich sachlich noch nicht artikulieren lassen.&lt;/em&gt;&lt;/p&gt;
&lt;p&gt;&lt;em&gt;Wie kann sich Ordnung im Chaos bilden? Macht dieser Prozess den Anfang greifbar, vielleicht sogar fasslich?&lt;/em&gt;&lt;/p&gt;
&lt;h2 id="prolog"&gt;Prolog&lt;/h2&gt;
&lt;p&gt;Die Sterne funkeln dröhnend durch das All.
Es gibt nur ihr Licht, das Alles macht.
Eine Stille, die sich selber hört,
Macht alles laut und gleich auch stumm.
So kommt das Etwas zu dem Nichts und fällt
Für immer und doch auch nie dahin zurück.&lt;/p&gt;
&lt;p&gt;Das Universum war sich kurz ganz selbst bewusst,
Bevor es nicht mehr merken konnte, wie es stirbt.
Es gab nur flüchtig Alles und dann Nichts.&lt;/p&gt;
&lt;p&gt;Des Kelches Rand zu weit für unser&amp;rsquo;n Geist,
Wir tranken von unsrer eignen Seele
In mächtig Zügen, bis die Stille wieder
Um uns knallte, hallte, wallte.&lt;/p&gt;</description></item><item><title>Linux From Scratch</title><link>https://sebastiansemper.de/posts/linux-from-scratch/</link><pubDate>Tue, 31 Dec 2019 10:38:14 +0000</pubDate><guid>https://sebastiansemper.de/posts/linux-from-scratch/</guid><description>&lt;p&gt;Ich stieß schon vor einigen Jahren auf Linux From Scratch (LFS), und war von der Idee fasziniert, weil ich wusste, wie viel ich dabei über Linux lernen würde. Bei LFS zieht man sich quasi selbst an den eigenen Haaren aus dem Teich, weil man ein komplettes GNU/Linux Ökosystem von Grund auf selbst aufbaut. Man kompiliert jedes Paket, erstellt jedes Konfigurations-File und bügelt jeden Patch auf. Ich hatte aber nie so richtig Lust, Zeit und Hardware. Jetzt habe ich das alte Thinkpad der Holden hergenommen und werde sehen, wie weit ich damit kommen kann. Von außen betrachtet scheint es recht straightforward zu sein. Host-System aufsetzen, massig Quellcode runterladen und ab die Post!&lt;/p&gt;
&lt;p&gt;Da ich dem alten Thinkpad eh lange beim Kompilieren zugucken muss, kann ich auch nebenbei locker einige Zeilen hier in den Blog tippeln und ein meine Eindrücke dokumentieren.&lt;/p&gt;</description></item><item><title>Einheit Welt Und Anfang</title><link>https://sebastiansemper.de/posts/einheit-welt-und-anfang/</link><pubDate>Wed, 07 Aug 2019 14:53:09 +0100</pubDate><guid>https://sebastiansemper.de/posts/einheit-welt-und-anfang/</guid><description>&lt;h3 id="die-einheit-von-welt-und-anfang"&gt;Die Einheit von Welt und Anfang&lt;/h3&gt;
&lt;p&gt;Unser Geist, der die Welt erfasst und konzeptioniert, bildet hierzu Kategorien, Begriffe, schöpft Bedeutungen, zeigt Verbindungen auf, leitet Gesetze ab und vollzieht alle möglichen Handlungen um die Welt um sich herum zu beschreiben. Eine natürliche Frage ist hierbei, wie die Welt entstanden ist, wie sie ihren Anfang gemacht haben kann. Denn schließlich ist der Anfang auch Teil der Welt deren Anfang er ist. Der Geist geht also von seinen Begriffen aus und versucht diesen Anfang, genauso wie er es vorher mit der Welt getan hat, zu fassen. Die Konzepte und Begriffe, die hierbei zum Tragen kommen, sind aber durch das Welt-Bild des Geistes geprägt, gefärbt, ja notwendigerweise beeinflusst. Hierbei kann der Geist dies nicht zufällig getan haben, denn die Konzepte müssen sich immernoch an der Welt messen. Doch je nach Entwicklungsstufe des Weltbildes können diese verschieden ausgeklügelt und komplex sein. Weiterhin kann die Konzeption des Anfangs auch nicht am explorativen Anfang stehen, denn es ist noch nicht gefasst, wovon diese der Anfang sein soll.&lt;/p&gt;</description></item><item><title>Wert durch Arbeit</title><link>https://sebastiansemper.de/posts/wert-durch-arbeit/</link><pubDate>Tue, 30 Jul 2019 11:33:38 +0000</pubDate><guid>https://sebastiansemper.de/posts/wert-durch-arbeit/</guid><description>&lt;p&gt;Im ersten Abschnitt von MEW23 geht es um die Ware und welche Bestimmungen ihr zukommen. Zunächst ist eine Ware ein simples Ding, welches einen Nutzen hat. Dieser Nutzen bestimmt sich allein durch materielle Eigenschaften dieses Dinges. Aus dem Nutzen eines bestimmten Dinges folgt die Tatsache, dass es Gebrauchswert ist.&lt;/p&gt;</description></item><item><title>Kommentar – "Los, komm, wir sterben endlich aus!"</title><link>https://sebastiansemper.de/posts/kommentar-los-komm-wir-sterben-endlich-aus/</link><pubDate>Wed, 03 Jul 2019 12:54:07 +0000</pubDate><guid>https://sebastiansemper.de/posts/kommentar-los-komm-wir-sterben-endlich-aus/</guid><description>&lt;p&gt;In der Ausgabe 27/2019 der ZEIT findet sich ein &lt;a href="https://www.zeit.de/2019/27/klimaschutz-kinderlos-menschliche-fortpflanzung-klimawandel"&gt;Artikel&lt;/a&gt; über Antinatalisten. Diese Bewegung ist eine auf vielfache Weise motivierte, welche behauptet, dass es ein fundamentales Problem mit der Geburt von Menschen gibt. Die einen behaupten dadurch eine Verbesserung des Klimas und wieder andere sehen sich zu Unrecht mit der Grausamkeit des Lebens konfrontiert und geben daran ihren Eltern die Schuld. Im Folgenden habe ich versucht die grundsätzlichen Argumente herauszuarbeiten und diese dann zu kommentieren.&lt;/p&gt;</description></item><item><title>Vom den einen zu dem Einen</title><link>https://sebastiansemper.de/posts/vom-den-einen-zu-dem-einen/</link><pubDate>Fri, 31 May 2019 13:49:31 +0000</pubDate><guid>https://sebastiansemper.de/posts/vom-den-einen-zu-dem-einen/</guid><description>&lt;p&gt;Als Menschen nehmen wir die Welt zunächst als Individuen wahr. Jedes Individuum ist in einer Gruppe verhaftet, im kleinsten Fall in einer Familie, welche meist mit noch vielen anderen in Kontakt steht, sodass sich schlussendlich eine Gesellschaft ergibt. Diese kann alle möglichen Ausgestaltungen annehmen und die Interaktionen der Individuen in diesen können mehr oder minder angenehm erscheinen. Fakt ist, dass ein Zusammenspiel der Individuen stattfindet, welches ultimativ die Vorgänge innerhalb der Gesellschaft ausbildet und wir diese als solche wahrnehmen. Das heißt nicht nur, dass die Wahrnehmung im Individuum erfolgt, sondern dass auch die Motivation zu Handlungen und die Bewertung den Individuen entspringt.&lt;/p&gt;</description></item><item><title>Von Anfang An</title><link>https://sebastiansemper.de/posts/von-anfang-an/</link><pubDate>Wed, 08 May 2019 22:18:23 +0000</pubDate><guid>https://sebastiansemper.de/posts/von-anfang-an/</guid><description>&lt;p&gt;Dinge existieren. Das ist schwerlich von der Hand zu weisen. &lt;em&gt;Was&lt;/em&gt; die Dinge überhaupt &lt;em&gt;sind&lt;/em&gt; spielt erst einmal keine Rolle. Wir wollen auch nicht darüber diskutieren, ob diese Dinge auch &lt;em&gt;aufhören&lt;/em&gt; zu existieren, oder auch sogar &lt;em&gt;anfangen&lt;/em&gt; können zu existieren. Hierbei ist das Ding im allgemeinsten Sinn überhaupt gemeint. Es ist auch nicht klar, ob es wirklich &lt;em&gt;mehr&lt;/em&gt; als &lt;em&gt;ein&lt;/em&gt; Ding gibt, doch es gibt etwas, das existiert, denn schließlich ist die Welt nicht &lt;em&gt;Nichts&lt;/em&gt;. Ganz unumwunden formuliert gibt es mindestens ein Etwas, das ist. Es gibt ein Ding an dem sich das &lt;em&gt;Alles&lt;/em&gt; abspielt. Nennen wir dieses Ding ein wenig spezifischer &lt;em&gt;Welt&lt;/em&gt;.&lt;/p&gt;</description></item><item><title>Quotendiskussion</title><link>https://sebastiansemper.de/posts/quotendiskussion/</link><pubDate>Mon, 11 Feb 2019 11:59:18 +0000</pubDate><guid>https://sebastiansemper.de/posts/quotendiskussion/</guid><description>&lt;p&gt;Bei uns an der Uni soll es ein Gremium geben, welches die Interessen von Doktoranden vertritt. Dabei soll eine Quotenempfehlung für Frauen von 40% in die Satzung wandern. Als kleine meditative Übung hier eine Diskussion über Gründe, Nutzen und Implikationen. Dabei soll konkret auf die Argumente eines Mitwirkenden an der Satzung eingegangen werden, welche als Zitate eingebracht werden.&lt;/p&gt;</description></item><item><title>Sind wir in einer ETF Blase?</title><link>https://sebastiansemper.de/posts/sind-wir-in-einer-etf-blase/</link><pubDate>Sun, 20 Jan 2019 14:12:01 +0000</pubDate><guid>https://sebastiansemper.de/posts/sind-wir-in-einer-etf-blase/</guid><description>&lt;p&gt;Mir wurde ein &lt;a href="https://www.onvista.de/news/sind-wir-in-einer-etf-blase-178631959"&gt;Artikel&lt;/a&gt; gezeigt, bei dem es darum geht, ob wir in einer ETF Blase sind, und welche Auswirkungen das haben könnte.&lt;/p&gt;</description></item><item><title>Efficient Market Theory (EMT)</title><link>https://sebastiansemper.de/posts/efficient-market-theory-emt/</link><pubDate>Mon, 31 Dec 2018 10:33:13 +0000</pubDate><guid>https://sebastiansemper.de/posts/efficient-market-theory-emt/</guid><description>&lt;p&gt;Wozu gibt es eigentlichen diesen freien Markt? Scheinbar benötigt man den freien Markt, nicht nur zum Handeln, sondern auch um überhaupt erst einmal den Preis einer zu handelnden Ware zu ermitteln.&lt;/p&gt;</description></item><item><title>Die Membran</title><link>https://sebastiansemper.de/posts/die-membran/</link><pubDate>Sun, 30 Dec 2018 23:20:58 +0000</pubDate><guid>https://sebastiansemper.de/posts/die-membran/</guid><description>&lt;p&gt;Er sitzt in seinem Badezimmer auf dem Boden. Ein verzweifelter Versuch seine sterblichen Überreste leichter entsorgbar zu machen. Viel Planung war nicht in die Herbeiführung seiner derzeitigen Situation geflossen. Er hält einen rostigen Revolver in seinen Händen. Zumindest vermutet er dies. Sowohl die Feuerwaffe als auch seine Extremitäten kann er nicht mehr wahrnehmen. Die Membran wurde schon vor Wochen fertig gestellt. Er streicht über den rauen Stahl und verspürt nichts dabei. Die Membran hält ihn davon ab. Zunächst hatte er gedacht, er könne sie überkommen, durchleiden, ja vielleicht sogar überdauern. Wie naiv dieser Gedanke doch gewesen ist! Doch er ist gewachsen, genau wie der Schleier um seine Sinne. Hierbei hat er leidvoll erfahren müssen, dass es nur einen Ausweg gibt, die Flucht nach vorn. Er richtet sich auf, streckt die Beine nach vorn aus und hebt beide Hände ruhig und besonnen vor sein Gesicht, darin der Revolver.&lt;/p&gt;</description></item><item><title>Indexing und ETFs aka "Fat stacks, yo!"</title><link>https://sebastiansemper.de/posts/indexing-und-etfs-aka-fat-stacks-yo/</link><pubDate>Sun, 30 Dec 2018 23:07:45 +0000</pubDate><guid>https://sebastiansemper.de/posts/indexing-und-etfs-aka-fat-stacks-yo/</guid><description>&lt;p&gt;Jeder hat sie, diese auf magische Weise immer länger werdende Bank, auf der man die Dinge ablegt, die einem wichtig erscheinen. Leider hat diese Bank eben nicht die Eigenschaft gleich lang zu bleiben, dass die Möglichkeit besteht, die Dinge am anderen Ende einfach herunterpurzeln zu lassen. Stattdessen können sich einfach immer größere und schwerere Projekte auf der unendlichen Bank ansammeln.&lt;/p&gt;
&lt;p&gt;Neben einer überfälligen Deload-Woche im Training und einer Steuererklärung hatte sich da auch einiges zum Thema Sparen aufgetürmt. Wie es das Schicksal so wollte, hatten wir neulich eine kleine Reunion vom Studium, wobei einer meiner alten Kollegen, der mittlerweile in Frankfurt bei einer kleinen Finanzklitsche den Untergang des Kapitalismus vorantreibt, auch am Start war. Dabei habe ich ihn mal durch die grüne Brille hindurch gefragt, wie man denn nun sein Geld anlegen solle. Und da kam er mit dem Thema &amp;ldquo;Indexing&amp;rdquo; und &amp;ldquo;ETFs&amp;rdquo; um die Ecke. Die Idee leuchtete mir, zumindest halbwegs, ziemlich schnell ein. Anstatt auf gewisse Aktien zu setzen, bildet man &amp;ldquo;einfach&amp;rdquo; einen Aktienindex nach und wartet, dass der Markt es einfach regelt.&lt;/p&gt;</description></item><item><title>Monologe</title><link>https://sebastiansemper.de/posts/monologe/</link><pubDate>Sun, 03 May 2015 10:09:17 +0100</pubDate><guid>https://sebastiansemper.de/posts/monologe/</guid><description>&lt;p&gt;So trug es sich zu, dass ein Wanderer des Geistes sich eines Tages durch die Straßen einer großen Stadt zu stehlen suchte. Er war schon eine ganze Weile auf seinem Weg und plötzlich hatte er sein Ziel vor Augen. Er erkannte diese belebte Straße sofort. Das war sie, wo er seinen Geist ergießen wollte. Er wollte vor die Menschen in dieser belebten Gegend sein Wissen dahinschütten auf dass sie sich darin die Füße befeuchten sollten.&lt;/p&gt;</description></item><item><title>Blumenmeer</title><link>https://sebastiansemper.de/posts/blumenmeer/</link><pubDate>Sat, 28 Mar 2015 10:08:26 +0100</pubDate><guid>https://sebastiansemper.de/posts/blumenmeer/</guid><description>&lt;p&gt;Es ist ein Frühlingstag. Aber ein solcher, welchen man noch nicht als einen ausgemacht hat. Es ist noch ein wenig neblig, windig und deshalb noch frisch in der Welt. Ein Blick gen Himmel lässt einem schon mehr die Wärme spüren als die Nase wirklich entgegen der Sonne zu halten. Sie ist noch nicht Herrin über die Geister des Winters geworden. Doch ein mehr als minder erzwungener Spaziergang durch die Stadt führt an einem Park vorbei, dessen Wiese zu einem nicht geringen Teil von weißen und lila Krokussen bevölkert wird. Dicht an dicht formen die Blüten ein Meer, dessen Wogen den Frühling lauter proklamieren, als man es an jenem Tag zu erwarten wagt. Es ist noch Zeit bis zum eigentlichen Zeck des Spaziergangs und es wäre ein Frevel sich nicht zumindest kurz auf einer Bank im Park nieder zu lassen und das Meer und dessen Brise aufzusaugen. Währendessen passieren einige Fußgänger. Sie schießen ein Foto, welches dann in der Wolke für immer verschwindet. Vielleicht lernen wir mittlerweile auch die Schönheit von Dingen zu erahnen, welche sich uns nur auf Displays präsentieren. Wahrscheinlich lassen sich alte Erinnerungen und deren damit verbundenen Gefühle vortrefflich durch simple digitale visuelle Reize zurück gewinnen. Man sollte eigentlich immernoch die Blüten und ihr Wogen im Wind bestaunen. Man könnte sich immernoch über diese mutigen Vorboten der kommenden Jahreszeit freuen. Wie sie furchtlos, villeicht sogar ein wenig herausfordernd, ihre Köpfe nach der Sonne aurichten, als wollten sie ja keinen einzigen Strahl vergeuden und jedes kleinste Fünkchen Licht und Wärme in pralle Farben und betörende Düfte verwandeln. Ein Gruppe Urlauber schart sich um das Ufer und ein wildes Knipsen setzt ein. So als wollte die Meute um jeden Preis vermeiden den Moment wirklich mit allen Sinnen wahrnehmen zu müssen. Es scheint schöner und einfacher zu sein an solchen Eindrücken nach dem Drucken der Fotos den Moment selbst aus dem Gedächtnis zu rekonstruieren. Aber welche Erinnerung bleibt denn bestehen, wenn der einzige jemals empfangene Reiz durch ein Display oder Sucher gefiltert und verharmlost wurde? Die ganze Szene beginnt immer mehr zu schwingen und vibrieren je länger man sich ihr hingibt. Einige übereifrige Bienen und Hummeln versuchen ihre ersten Landeanflüge auf die Blütenblätter. Wie kleine Schiffchen schaukeln sie von Woge zu Woge und ergattern immer mehr Ladung für ihre ersten Nachkommen. Die Pause nähert sich nach einer knappen Stunde gemächlich ihrem Ende und die Reise muss weiter gehen. Fast schon Seekrankheit hat sich im Magen breit gemacht und es fällt umso schwerer wieder an Land Fuß zu fassen. Beim Aufstehen sollte einem noch einmal kurz einfallen ein Foto für die Liebste zu machen, auch ohne schlechtes Gewissen nur einen schlechten Abdruck festhalten zu können.&lt;/p&gt;</description></item><item><title>Vor der Tür</title><link>https://sebastiansemper.de/posts/vor-der-tuer/</link><pubDate>Fri, 27 Feb 2015 10:07:32 +0100</pubDate><guid>https://sebastiansemper.de/posts/vor-der-tuer/</guid><description>&lt;p&gt;Alles begann mit einer einfachen einschlägigen Tür, die täglich von vielen Passanten passiert und per paranoider Passion ignoriert wurde. Doch einmal ergab ein Ereignis, dass einer des Weges kommend wegbog und durch den Durchgang über die Schwelle schwungvoll schritt. Aber das wurd wahrlich wage wahrgenommen und niemand nahm Notiz. Schließlich schlich der Schleicher schläfrig aber seelig über die Schwelle und ward wahnsinnig wachsam wahrgenommen. Jeder um ihn herum huldigte häufig des Herren glänzendes Glück. Es schien schön und schmeichelnd aus dessen Gesicht zu scheinen.&lt;/p&gt;</description></item><item><title>Der Kastanienbaum</title><link>https://sebastiansemper.de/posts/kastanienbaum/</link><pubDate>Sat, 15 Nov 2014 10:05:37 +0100</pubDate><guid>https://sebastiansemper.de/posts/kastanienbaum/</guid><description>&lt;p&gt;Der Bauhof war mein Abenteuerspielplatz. Er war aufregend, ständig im Wandel, weitläufig und ein klein bisschen gefährlich. Neben dem Grundstückszaun, der das Gelände ringsum berahmte, wurde ein Teil des Randes von einer mächtigen Kastanie bewacht. Sie wirkte, als wäre sie vor Äonen aus eines Titanes Hand hier in unsere Welt gerammt worden. Zwar dachte ich mir das damals als kleiner Bub&amp;rsquo; nicht so, doch rückblickend muss ich mich so gefühlt haben. Jeden Sommer warf sie einen riesigen Schatten, in dessen milder Kühle vor allem die stacheligen Früchte den Anfang ihres Lebens fristeten.&lt;/p&gt;</description></item><item><title>Die Letzte Ehre</title><link>https://sebastiansemper.de/posts/die-letzte-ehre/</link><pubDate>Wed, 24 Sep 2014 10:04:29 +0100</pubDate><guid>https://sebastiansemper.de/posts/die-letzte-ehre/</guid><description>&lt;p&gt;Wir schreiben das Jahr 2016 und der Einsatz der deutschen Soldaten im mittleren und nahen Osten ist vorbei. Einige von ihnen sind zurück gekehrt und einige mehr sind es nicht. Der Tag der letzten Ehre für die Gefallenen ist gekommen. Die Zeremonie wird von einem sich schwermütig gebenden Bundespräsidenten in fast schwarz abgehalten. Zugegen sind Politiker, Generäle, Offiziere, Veteranen, Journalisten, das Fernsehen, Industrielle, Künstler, Prominente und auch ein paar Eltern von Gefallenen. In Erinnerung an diese bewegende Stunde, gilt es nun die Worte einiger Redner wieder zu geben:&lt;/p&gt;</description></item><item><title>Nihilia</title><link>https://sebastiansemper.de/posts/nihilia/</link><pubDate>Thu, 12 Dec 2013 20:52:04 +0100</pubDate><guid>https://sebastiansemper.de/posts/nihilia/</guid><description>&lt;p&gt;In Nihilia ist das Leben schön. Man wird in eine leere und ereignislose Welt geboren. Im Moment der Geburt wird dem Menschen noch für einen kurzen Moment alles gewährt. In diesem Augenblick ist man das Unvollkommenste. Zuerst wird man der Verbindung zur eigenen Mutter beraubt. Das für Gebährende und Geborene nicht sehr schmerzhafte Durchtrennen der Nabelschnur ist für die Frau Befreiung. Für das Kind der Beginn des Verzichts. Der Verzicht auf Nahrung, Sauerstoff und Lebensraum. Jedes menschliche und damit erbarmungswürdige Lebewesen trägt seit Geburt einen Mundschutz, welcher den Anteil des inhalierten lebenswichtigen Sauerstoffes auf ein extremal minimales Höchstmaß reduziert, um keine unverholen gierigen Einflüsse auf die umgebende und sehr fragile Welt zu nehmen. Durch die Zufuhr von speziellen Hormonen werden Appetit und Wachstum so weit reguliert, dass ein übermäßig unnötiges Wachstum und unverhältnismäßiger Konsum von kostbaren Nahrungsmitteln in Zaum gehalten werden können. Somit wird recht von Beginn an schon eine unnötige Vielfalt an körperlicher und geistiger Bewegungsfreiheit vermieden. Doch Kinder neigen letztlich teilweise immernoch dazu sich ihrem Drang nach Bewegungen durch Ausführen von nutzlosen Erkundungen und sinnbefreitem Ausprobieren hinzugeben.&lt;/p&gt;</description></item><item><title>Schubladen</title><link>https://sebastiansemper.de/posts/schubladen/</link><pubDate>Fri, 31 Aug 2012 09:22:03 +0100</pubDate><guid>https://sebastiansemper.de/posts/schubladen/</guid><description>&lt;h2 id="einteilen"&gt;Einteilen&lt;/h2&gt;
&lt;p&gt;Wie teilen wir uns unsere Welt so ein? Nunja, da gibt es die Männer und Frauen, die Weißen und die Schwarzen, die Armen und die Reichen, die Schlauen und die Dummen. Dann die Hipster, die Gangster, die Punks, die Nerds, die Nazis, die Ökos, die Goths, die Metalheads und die Chayas. Dann gibt es die Scientologen, die Christen, die Katholiken, die Protestanten, die Orthodoxen, die Juden, die Hindus, die Konfizionisten, die Buddhisten, die Atheisten. Dann gibt es die Heteros, der Transen, die Homos und die die selbst nicht wissen, was sie sind. Dann gibt es die Trainierten, die Bierbauchträger, die Flachbrüstigen, die mit den Rieseneutern, die mit den Pferdebeinen. Die Rocker, die Schranzer, die Hopper, die {Black, Death, Pagan, Viking, Mittelalter, Power, Heavy}-Metaller. Dann gibt es die Sportler, die Musiker, die Künstler, die Astronauten, die Prediger, die Handwerker, die Intelektuellen, die Fließbandarbeiter. Dann gibt es die Fleischesser, die Vegetarier, die Veganer und die Frutarier. Dann gibt es die Lieben, die Netten, die Sympathischen, die Arroganten, die Untreuen, die Verlässlichen, die Ergeizigen, die Faulen, die Gleichgültigen, die Antriebslosen, die Arschlöcher, die Träumer, die Waghalsigen, die Abenteuerlichen. Die Unschuldslämmer, die Ladendiebe, die Drugaddicts, die Dealer, die Kinderschänder, die Nekrophilen. Doch wie entstehen solche Kategorien? Ich kann mir vorstellen, dass Dinge wie die eigenen Erfahrungen, die Erziehung und das eigene Umfeld eine wichtige Rolle in diesem Prozess Kategorisierung spielen. Es stellt sich mir auch die Frage, wie statisch solche Kriterien sind. Wie leicht oder schwer lassen sich diese beeinflussen? Sind sie überhaupt von außen änderbar? Kann diese Veränderung nur von Innen heraus passieren?&lt;/p&gt;</description></item><item><title>Diskussion über das Gewicht des Wirklichen</title><link>https://sebastiansemper.de/posts/gewicht-des-wirklichen/</link><pubDate>Mon, 02 Jul 2012 09:24:16 +0100</pubDate><guid>https://sebastiansemper.de/posts/gewicht-des-wirklichen/</guid><description>&lt;p&gt;&amp;ldquo;Texte zur Theorie der Fotografie&amp;rdquo; - Reclam Verlag (2010); PHILIPPE DUBOIS - Die Fotografie als Spur eines Wirklichen (1990), Seite 102ff.&lt;/p&gt;</description></item><item><title>Zur Definition</title><link>https://sebastiansemper.de/posts/zur-definition/</link><pubDate>Thu, 01 Dec 2011 09:19:10 +0100</pubDate><guid>https://sebastiansemper.de/posts/zur-definition/</guid><description>&lt;p&gt;Ein Kommentar zu &amp;ldquo;Theodor Adorno - Philosophische Terminologie - Zur Einleitung, Suhrkamp 1973&amp;rdquo;&lt;/p&gt;
&lt;p&gt;In den ersten drei Kapiteln des genannten Buches zu einer Vorlesung, welche von Adorno zu eben diesem Thema gehalten wurde - ich denke für Erstsemester der Philosophie - dreht es sich darum, in wie weit es sinnvoll ist in einem philosophischen Definitionen zu fordern und sie deshalb auch zu suchen. Hierbei bezieht er sich einerseits auf Kant und erläutert auch die Ansichten, welche Nietzsche dazu geäußert hatte.&lt;/p&gt;</description></item><item><title>Distributed multisensor ISAC</title><link>https://sebastiansemper.de/pub/2026/thomae2026distributed_ms_isac/</link><pubDate>Thu, 01 Jan 1970 02:33:46 +0200</pubDate><guid>https://sebastiansemper.de/pub/2026/thomae2026distributed_ms_isac/</guid><description>&lt;p&gt;Integrated Sensing and Communications (ISAC) will become a service in future mobile communication networks. It enables the detection and recognition of passive objects and environments using radar-like sensing. The ultimate advantage is the reuse of the mobile network and radio access resources for scene illumination, sensing, data transportation, computation, and fusion. It enables building a distributed, ubiquitous sensing network that can be adapted for a variety of radio sensing tasks and services. In this article, we develop the principles of multi-sensor ISAC (MS-ISAC). MS-ISAC corresponds to multi-user MIMO communication, which in radar terminology is known as distributed MIMO radar. First, we develop basic architectural principles for MS-ISAC and link them to example use cases. We then propose a generic MS-ISAC architecture. After a brief reference to multipath propagation and multistatic target reflectivity issues, we outline multilink access, coordination, precoding and link adaptation schemes for MS-ISAC. Moreover, we review model-based estimation and tracking of delay/Doppler from sparse OFDMA/TDMA frames. We emphasize Cooperative Passive Coherent Location (CPCL) for bistatic correlation and synchronization. Finally, issues of multisensor node synchronization and distributed data fusion are addressed.&lt;/p&gt;</description></item><item><title>Integrated Radio Sensing Capabilities for 6G Networks: AI/ML Perspective</title><link>https://sebastiansemper.de/pub/2026/shatov2026/</link><pubDate>Thu, 01 Jan 1970 02:33:46 +0200</pubDate><guid>https://sebastiansemper.de/pub/2026/shatov2026/</guid><description>&lt;p&gt;The sixth-generation wireless communications (6G) is often labeled as &amp;ldquo;connected intelligence&amp;rdquo;. Radio sensing, aligned with machine learning (ML) and artificial intelligence (AI), promises, among other benefits, breakthroughs in the system&amp;rsquo;s ability to perceive the environment and effectively utilize this awareness. This article offers a panoramic view of radio sensing by unifying physical object sensing and spectrum sensing. To this end, while staying in the framework of integrated sensing and communication (ISAC), we expand the term &amp;ldquo;sensing&amp;rdquo; from radar, via spectrum sensing, to miscellaneous applications of radio sensing like non-cooperative transmitter localization. We formulate the problems, explain the state-of-the-art approaches, and detail AI-based techniques to tackle various objectives in the context of wireless sensing. Finally, we discuss the potential integration of various radio sensing functions into a common AI-enhanced framework, emphasizing the possible benefits and the challenges to overcome. In addition to the tutorial-style core of this work based on direct authors&amp;rsquo; involvement in 6G research problems, we review the related literature, and provide both a good start for those entering this field of research, and a topical overview for a general reader with a background in wireless communications.&lt;/p&gt;</description></item><item><title>Benchmarking CFAR and CNN-Based Peak Detection Algorithms in ISAC Under Hardware Impairments</title><link>https://sebastiansemper.de/pub/2025/tosi2025cfar_benchmark/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/tosi2025cfar_benchmark/</guid><description>&lt;p&gt;Peak detection is a fundamental task in radar and has therefore been studied extensively in radar literature. However, Integrated Sensing and Communication (ISAC) systems for sixth generation (6 G) cellular networks need to perform peak detection under hardware impairments and constraints imposed by the underlying system designed for communications. This paper presents a comparative study of classical Constant False Alarm Rate (CFAR)-based algorithms and a recently proposed Convolutional Neural Network (CNN)-based method for peak detection in ISAC radar images. To impose practical constraints of ISAC systems, we model the impact of hardware impairments, such as power amplifier nonlinearities and quantization noise. We perform extensive simulation campaigns focusing on multi-target detection under varying noise as well as on target separation in resolution-limited scenarios. The results show that CFAR detectors require approximate knowledge of the operating scenario and the use of window functions for reliable performance. The CNN , on the other hand, achieves high performance in all scenarios, but requires a preprocessing step for the input data.&lt;/p&gt;</description></item><item><title>Cubic B-Splines for Hardware-Accelerated Antenna Beampattern Interpolation</title><link>https://sebastiansemper.de/pub/2025/naviliat2025splines_beampattern/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/naviliat2025splines_beampattern/</guid><description>&lt;p&gt;For applications that require fast evaluations of an antenna&amp;rsquo;s beampattern, we propose a computationally efficient interpolation technique based on cubic B-Splines. We expand the pattern into a sum of finite-support polynomial functions that allow a fixed runtime evaluation irrespective of the antenna&amp;rsquo;s aperture size. Further, we show how B-Splines can be implemented efficiently on GPU hardware leveraging so-called texture memory. We outline how oversampling, i.e. decreasing the distance between adjacent B-Splines, can approximate ideal low-pass filtering. Comparison to the Effective Aperture Distribution on measured antenna calibration data shows a 10-fold speedup in runtime while attaining an accuracy of -100 dB. When used for direction estimation using antenna arrays, we show that the proposed B-Spline interpolation yields a statistically efficient estimator , as it attains the Cramer-Rao Lower Bound.&lt;/p&gt;</description></item><item><title>Dual-Band Sensing for Passive Target Surveillance in ISAC Systems</title><link>https://sebastiansemper.de/pub/2025/zhao2025dualband_sensing_isac/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/zhao2025dualband_sensing_isac/</guid><description>&lt;p&gt;A ubiquitous and high-performance sensing service is required in the next generation of wireless communication systems. The legacy Sub-6GHz bands can provide a larger sensing range, but the available bandwidth limits the resolution. In contrast, the millimeter-wave (mmWave) frequencies enhance the sensing performance with wider bandwidth. However, high-directional beamforming poses new challenges in passive target surveillance. The conventional beam sweeping in mm Wave is costly due to the significant overhead, which conflicts with the goal of communication on efficient data transmission. To reduce the overhead caused by target surveillance, we propose a dual-band sensing approach. By combining the advantages of both Sub-6GHz and mmWave frequencies, our method enables fine target detection in a wide area. Compared to beam sweeping, the pro-posed method significantly reduces the communication overhead and time consumption, while also lightening the computational load of radar signal processing.&lt;/p&gt;</description></item><item><title>Elowen: MIMO Channel Sounding and Characterization in an Urban Macro Cell</title><link>https://sebastiansemper.de/pub/2025/semper2025elowen/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/semper2025elowen/</guid><description>&lt;p&gt;The measurement campaign was performed in Cologne (Germany) using a RUSK MIMO channel sounder operating at 2.53 GHz. The fixed transmitter has been mounted on a roof top of an approx. 30m high building whereas the mobile receiver has been set up in a car. The application of high resolution parameter estimation algorithms requires adequate antenna arrays. Therefore the measurements have been done using a stacked uniform circular array at the base station side (Tx) as well as at the mobile station side (Rx). Both arrays are linear dual polarized. In order to obtain an antenna independent characterization of the radio channel the high resolution parameter estimation framework RIMAX was used. RIMAX models the channel as the superposition of a deterministic (specular) and a stochastic (dense multipath) part. The deterministic part resembles a multitude of plane waves that are resolvable by the estimator. Each plane wave is hereby described using its time-delay, angle-of-arrival (in azimuth and elevation), angle-of-departure (in azimuth and elevation) and the complex polarimetric path-weight. In order to render the data self contained, we have also added suitable calibration data of both antenna arrays. Only through the angle-, polarization- and port-dependent description of the antenna arrays one can achieve correct deembedding of the measurement equipment. In addition to the measurements and calibration data, we also supply four different sets of high resolution parameter estimates that are derived form the data based on RIMAX where different strategies for model-order selection and over-equalization were employed. Since these different settings impose different objective functions and parametric models onto the data, the resulting parameter estimations have signficantly different statistics, which sparks the need for further analysis that is facilitated by this dataset.&lt;/p&gt;</description></item><item><title>Estimating Angular Diffuse Components and Model Misspecification in MIMO Channel Sounding</title><link>https://sebastiansemper.de/pub/2025/semper2025angle_dmc/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/semper2025angle_dmc/</guid><description>&lt;p&gt;Parameter estimation for MIMO channel sounding data aims at accurately describing channel measurements with physically realistic and interpretable parameters. The performance of model-based approaches, e.g. maximum likelihood, is determined by the accuracy of the imposed signal model. For channel sounding data it has turned out to be beneficial to use two distinct concepts for the description of the propagation process. The specular components account for the dominant propagation paths of plane waves, whereas diffuse components model the weaker but more diverse propagation processes by means of a colored noise process. In order to improve the accuracy of the model for the diffuse components we propose a simple but still flexible parametric covariance model that allows to account for a smooth power angle profile that describes the correlation in the spatial domain. Moreover, the model for the deterministic part of the signal is usually contaminated by calibration errors, which in turn deteriorate the reliability of the specular path estimates. This is most prominently visible by the estimation of so-called ghost paths. To mitigate this we introduce a new model order selection scheme based on the so-called misspecified Cram \textasciiacute{}er Rao bound which accounts for the unavoidable modeling errors. Additionally, to avoid the fitting of ghost paths caused by the faulty modeling of strong specular components we locally decrease the estimated SNR in time domain around already estimated ones. Further, as these changes to the signal model require more computational resources compared to existing algorithms, we also showcase how necessary quantities like likelihoods, score functions and Fisher information matrices can still be computed efficiently. We implement our proposed extensions within the RIMAX framework. We also showcase that they improve the reliability of the produced estimates compared to plain vanilla RIMAX on real measurement data.&lt;/p&gt;</description></item><item><title>Experimental Performance Validation of Fisher Information-Optimized Multicarrier Waveforms for Sub-THz Channel Sounding</title><link>https://sebastiansemper.de/pub/2025/gedschold2025signal_design_thz_meas/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/gedschold2025signal_design_thz_meas/</guid><description>&lt;p&gt;This paper describes an experimental performance validation of Fisher information-based waveform designs for channel sounding and high-resolution propagation parameter estimation. These waveform designs require prior knowledge about the parameters which can be estimated from a preceding measurement. We describe a setup to implement this intermediate parameter estimation and waveform optimization step. A subsequent performance validation via the empirical MSE derived from measurement-based Monte Carlo trials is also presented. This paper builds upon a previous simulation-based study that found that the CRB for delay estimation improves using the optimized waveforms. This finding can be reproduced for the delay MSE with the presented experiments. An additional insight is that the accuracy of the intermediate estimates does not significantly impact the MSE, as long as the SNR is sufficient to detect the required paths reliably.&lt;/p&gt;</description></item><item><title>Joint Delay-Doppler Estimation Using OFDMA Payloads for Integrated Communications and Sensing</title><link>https://sebastiansemper.de/pub/2025/miranda2025joint_est_ofdma/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/miranda2025joint_est_ofdma/</guid><description>&lt;p&gt;The use of future communication systems for sensing offers the potential for a number of new applications. In this paper, we show that leveraging user data payloads in multinode Orthogonal Frequency Division Multiple Access (OFDMA) networks for estimating target delay and Doppler-shift parameters can yield a significant advantage in SNR and addressable bandwidth. However, gaps in the frequency-time resources, reference signal boosting and amplitude modulation schemes introduce challenges for estimation at the sensing receiver. In this work, we propose a joint delay and Doppler-shift model-based estimator designed to address these challenges. Furthermore, we demonstrate that incorporating knowledge of the device model into the estimation procedure helps mitigate the effects of the nonideal radar ambiguity function caused by amplitude-modulated user payloads and sparse reference signals. Simulation results demonstrate that the estimator achieves the theoretical lower bound on estimation variance.&lt;/p&gt;</description></item><item><title>Measurement-Based Evaluation of CNN-Based Detection and Estimation for ISAC Systems</title><link>https://sebastiansemper.de/pub/2025/schieler2025measurements_dnn/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/schieler2025measurements_dnn/</guid><description>&lt;p&gt;In wireless sensing applications, such as Integrated Sensing and Communication (ISAC), one of the first crucial signal processing steps is the detection and estimation targets from a channel estimate. Effective algorithms in this context must be robust across a broad Signal-to-Noise Ratio (SNR) range, capable of handling an unknown number of targets, and computationally efficient for real-time implementation. During the last decade, different Machine Learning methods have emerged as promising solutions, either as standalone models or as complementing existing techniques. However, since models are often trained and evaluated on synthetic data from existing models, applying them to measurement is challenging. All the while, training directly on measurement data is prohibitive in complex propagation scenarios as a groundtruth is not available. Therefore, in this paper, we train a Convolutional Neural Network (CNN) approach for target detection and estimation on synthetic data and evaluate it on measurement data from a suburban outdoor measurement. Using knowledge of the environment as well as available groundtruth positions, we study the detection probability and accuracy of our approach. The results demonstrate that our approach works on measurement data and is suitable for joint detection and estimation of sensing targets in ISAC systems.&lt;/p&gt;</description></item><item><title>Metrology of Multicarrier-Based Delay-Doppler Channel Sounding for sub-THz Frequencies</title><link>https://sebastiansemper.de/pub/2025/gedschold2025multicarrier_metrology/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/gedschold2025multicarrier_metrology/</guid><description>&lt;p&gt;Developing channel models typically requires aggregating channel measurements and the corresponding extracted propagation parameters from different research institutions to form a sufficiently large data basis. However, uncertainties arising from limitations of the sounding hardware and algorithms may greatly impact the comparability between sounding results. Especially, (sub-)THz channel sounders do not allow simultaneous spatially and timely resolved measurements as known from sub-6 GHz and mm-wave applications (right now), limiting the possibilities of a hardware-independent channel characterization. At the same time, a high Doppler bandwidth may occur due to the high carrier frequencies, limiting the time spans for coherent or incoherent data processing. Hence, assessing the sounder&amp;rsquo;s performance and limits is important before interpreting the measurement results. Evaluating the sounder performance requires a traceable reference allowing tracing back measurements (or estimated propagation parameters) to a physical ground truth. Therefore, we propose and discuss an over-the-air artifact allowing a joint verification of delay and Doppler parameters in a multipath scenario. The evaluations of exemplary sub-THz measurements with a multicarrier-based sounder highlight the strong interplay between sounder hardware and estimation algorithms, especially when coping with the mutual interference of parameters from multiple propagation paths. Hence, a metrological assessment always requires considering the full processing pipeline from the unprocessed measurements up to the extracted propagation parameters.&lt;/p&gt;</description></item><item><title>Model-Based Joint Delay-Doppler Estimation for ICAS with Sparse OFDM Resources</title><link>https://sebastiansemper.de/pub/2025/miranda2025mbpe_sparse_ofdma/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/miranda2025mbpe_sparse_ofdma/</guid><description>&lt;p&gt;Studies have shown that OFDM waveforms are promising candidates for dual functionality radar and communication networks. However, when the transmitted OFDM frame contains sparsely and arbitrarily allocated resources, periodogram based target estimation suffers from high side-lobes and poor multi-path resolution. We propose an iterative multi-path joint delay-Doppler estimator for OFDM waveforms with sparse and possibly unstructured Frequency-Time resources. We show through simulations that model-based methods achieve greater than Rayleigh resolution without the need for additional interpolation of the sampled channel transfer function. Furthermore, the estimation error as function of delay and Doppler separation is studied and shown to approach the CRB. To address issues of computational complexity, we simplify the exhaustive search step commonly found in Maximum-Likelihood based estimators and propose the use of Fischer-scored gradient optimization that requires only a small number of updates to enhance estimation accuracy.&lt;/p&gt;</description></item><item><title>Multi-Bistatic Coordinated Multipoint JCAS: Exploring the Synchronization Requirements</title><link>https://sebastiansemper.de/pub/2025/elgamal2025sync_requirements/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/elgamal2025sync_requirements/</guid><description>&lt;p&gt;This paper investigates synchronization accuracy and derived synchronization requirements for multi-bistatic Joint Communication and Sensing (JCAS) systems, with a focus on Coordinated MultiPoint (CoMP) architectures. JCAS technology, especially in Vehicle-to-everything (V2X) networks, can enhance road safety by integrating radar detection with communication systems. Multi-bistatic CoMP configurations minimize hardware complexity by distributing transmitters and receivers to remote radio units (RRUs). Using the cooperative passive coherent location (CPCL) technique, we evaluate key synchronization issues , including carrier frequency offset (CFO), timing offset (TO) and sampling frequency offset (SFO), as well as their impact on radar performance. Simulations show how these offsets affect range resolution, Doppler resolution and detection accuracy. We demonstrate that while CFO and TO lead to misestimation of target velocity and position, SFO additionally reduces resolution along both position axes and velocity. These insights provide system designers recommendations on how to adjust multi-bistatic CoMP JCAS performance.&lt;/p&gt;</description></item><item><title>Multidimensional Parameter Estimation for Channel Sounding</title><link>https://sebastiansemper.de/pub/2025/gedschold2025sounding_chapter/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/gedschold2025sounding_chapter/</guid><description>&lt;p&gt;This chapter deals with parametric multidimensional channel models, with the estimation of the corresponding parameters (such as delays, path weights and angles) from channel sounding data and with the metrological aspects associated to this processing chain. Deterministic and stochastic components of the channel model are discussed, focusing on decomposing the channel into a superposition of discrete multipath components and non-resolvable dense multipath components (DMC). Terahertz frequency bands readily enable the deployment of wide bandwidths; hence, models designed for narrow bandwidths need to be revisited and adapted. From a metrological point of view, the limits of parameter estimation need to be considered. It is shown how these limits relate to the choice of the sounding waveform and how they can aid in determining a reasonable number of propagation paths (model order selection).&lt;/p&gt;</description></item><item><title>Newtonized Orthogonal Matching Pursuit for High-Resolution Target Detection in Sparse OFDM ISAC Systems</title><link>https://sebastiansemper.de/pub/2025/shah2025nomp_ofdm/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/shah2025nomp_ofdm/</guid><description>&lt;p&gt;Integrated Sensing and Communication (ISAC) is a technology paradigm that combines sensing capabilities with communication functionalities in a single device or system. In vehicle-to-everything (V2X) sidelink, ISAC can provide enhanced safety by allowing vehicles to not only communicate with one another but also sense the surrounding environment by using sidelink signals. In ISAC-capable V2X sidelink, the random resource allocation results in an unstructured and sparse distribution of time and frequency resources in the received orthogonal frequency division multiplexing (OFDM) grid, leading to degraded radar detection performance when processed using the conventional 2D-FFT method. To address this challenge, this paper proposes a high-resolution off-grid radar target detection algorithm irrespective of the OFDM grid structure. The proposed method utilizes the Newtonized orthogonal matching pursuit (NOMP) algorithm to effectively detect weak targets masked by the sidelobes of stronger ones and accurately estimates off-grid range and velocity parameters with minimal time and frequency resources through Newton refinements. Simulation results demonstrate the superior performance of the proposed NOMP-based target detection algorithm compared to existing compressed sensing (CS) methods in terms of detection probability, resolution, and accuracy. Additionally, experimental validation is performed using a bistatic radar setup in a semi-anechoic chamber. The measurement results validate the simulation findings , showing that the proposed algorithm significantly enhances target detection and parameter estimation accuracy in realistic scenarios.&lt;/p&gt;</description></item><item><title>Real-Time 141 GHz JCAS Channel Sounder: Near-Field Switched Beamforming, Carrier Multiplexing, and Context Awareness</title><link>https://sebastiansemper.de/pub/2025/bang2025sounder141ghz/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/bang2025sounder141ghz/</guid><description>&lt;p&gt;We have developed a real-time 141 GHz joint communications and sensing (JCAS) channel sounder, with focus on this article on 3-D near-field imaging (less than 50 cm range) for applications such as finger recognition. To achieve this, we integrated phased-array antennas into the channel sounder&amp;ndash;to the best of our knowledge, we are the first to do so in the D-band&amp;ndash;and adapted the switched beamforming technique we developed previously, to realize real-time near-field beamforming. We also propose another novel technique, coined carrier multiplexing, to extend the instantaneous bandwidth of the phased arrays to their full radio frequency (RF) passband&amp;ndash;by a factor of two for our current arrays and by a factor of ten for our next edition. Finally, we integrated camera and Lidar systems that are temporally and spatially synchronized with the channel sounder to generate real-time digital twins of the target while sounding, enabling machine learning on multimodal data. We showcase system performance with controlled experiments on metal cylinders, demonstrating imaging resolution on the order of a centimeter.&lt;/p&gt;</description></item><item><title>Sensing-Aided Beamforming: The Impact of Distributed Sensing Network Geometry</title><link>https://sebastiansemper.de/pub/2025/zhao2025sensing_aided_bf/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/zhao2025sensing_aided_bf/</guid><description>&lt;p&gt;As modern wireless systems increasingly focus on high-frequency bands, accurate beamforming has become an essential component for ensuring high-rate data transmission and robust connections. To reduce the overhead and latency caused by beam training, radar sensing can be used to support beamforming by narrowing the search space. Following the concept of Integrated Sensing and Communication (ISAC), we employ a distributed sensing network with multiple bistatic links to localize the target. The estimated target position is then used to assist beamforming. In this paper, we specifically investigates the underlying sensing performance for the given network geometry. Our results show that the spatially inhomogeneous distribution of the Cram'er-Rao Lower Bound (CRLB) can influence the Angle of Departure (AoD) estimation of beamforming, which highlights the importance of beamformer placement in the sensing network. Furthermore, we present a beamforming accuracy map, which can serve as a reference for optimizing placement strategies.&lt;/p&gt;</description></item><item><title>Wireless Propagation Parameter Estimation with Convolutional Neural Networks</title><link>https://sebastiansemper.de/pub/2025/schieler2025prop_est_cnn/</link><pubDate>Thu, 01 Jan 1970 02:33:45 +0200</pubDate><guid>https://sebastiansemper.de/pub/2025/schieler2025prop_est_cnn/</guid><description>&lt;p&gt;Wireless channel propagation parameter estimation forms the foundation of channel sounding, estimation, modeling, and sensing. This paper introduces a deep learning approach for joint delay and Doppler estimation from frequency and time samples of a radio channel transfer function. Our work estimates the 2D path parameters from a channel impulse response containing an unknown number of paths. Compared to existing deep learning-based methods , the parameters are not estimated via classification but in a quasi-grid-free manner. We employ a deterministic preprocessing scheme that incorporates a multichannel windowing to increase the estimator&amp;rsquo;s robustness and enables the use of a convolutional neural network (CNN) architecture. The proposed architecture then jointly estimates the number of paths along with the respective delay and Doppler shift parameters of the paths. Hence, it jointly solves the model order selection and parameter estimation task. We also integrate the CNN into an existing maximum-likelihood estimator framework for efficient initialization of a gradient-based iteration, to provide more accurate estimates. In the analysis, we compare our approach to other methods in terms of estimate accuracy and model order error on synthetic data. Finally, we demonstrate its applicability to real-world measurement data from a anechoic bistatic RADAR emulation measurement.&lt;/p&gt;</description></item><item><title>Broadband DRA with Uniform Angular Dependent Delay for Indoor Localization</title><link>https://sebastiansemper.de/pub/2024/rashidifar2024wideband_dra/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/rashidifar2024wideband_dra/</guid><description>&lt;p&gt;Estimating the Time Difference of Arrival (TDoA), is a simple yet reliable technique to accurately perform an indoor monostatic localization. To implement TDoA estimation, one approach is to utilize a broadband radar system equipped with multiple receiving antenna elements. To obtain the Time of Arrival (ToA) at each antenna element, the round-trip time is required. However, the round-trip time does not only consist of the propagation delay in free space but the propagation delay within the antenna as well. To perform the localization precisely, it is desired that an antenna element introduces a uniform delay in all directions. To this end, a compact rectangular dielectric resonator antenna is designed for the operating frequency of 6.5 GHz with a fractional bandwidth of 20%. Al2O3 with a dielectric constant of 9.8 is used for the substrate as well as the dielectric resonator. The antenna is designed to provide a high correlation between the input and the output pulses. To investigate the correlation, the antenna is excited with a modulated Gaussian pulse and the radiated pulses are studied. The antenna possesses an excellent behavior in terms of pulse preservation for the upper hemisphere. Therefore, when incoming pulses from the same distance but different directions impinge on the antenna, they reach the port of the antenna at a similar time. It is shown that this feature of the proposed antenna allows the utilization of TDoA estimation without the need for a calibration step. The characteristics of the antenna are verified by simulation and measurement.&lt;/p&gt;</description></item><item><title>Distributed Computing and Model-Based Estimation for Integrated Communications and Sensing: A Roadmap</title><link>https://sebastiansemper.de/pub/2024/semper2024medicas/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/semper2024medicas/</guid><description>&lt;p&gt;The recent advances in Integrated Sensing and Communications (ICAS) essentially transform mobile radio networks into a diverse , dynamic and heterogeneous sensing network. For the application of localization, the acquired sensing data needs to be processed to estimates of the state vectors of the targets in a timely manner. This paper aims at providing a roadmap for the development of a suitable computing system for ICAS. We propose to embed the signal processing into the concept of edge computing. It provides the necessary theoretical computing framework, since it alleviates the need for communication with a remote cloud. To obtain localization information in such a distributed, asynchronous and heterogeneous scenario, we study how existing maximum likelihood estimation techniques can be transformed into algorithms that can be orchestrated close to the edge. The advantage of these approaches is that they have well studied statistical properties and efficient algorithmic implementations exist. We propose to study how to derive a graph that encodes these algorithms&amp;rsquo; processing by relating individual and isolated computations in terms of the input/output-behavior of so-called compute nodes. This compute graph structure can then be flexibly distributed across multiple devices and even whole processing/sensing units. Moreover, modern computing architectures leverage such graph structures to optimize the efficient use of computing hardware. Additionally, once this graph is constructed we have laid the groundwork for the possibility to exchange certain compute steps by deep learning architectures. For instance, this allows to sidestep some costly iterative part of traditional maximum likelihood estimators, which further contributes to the low-latency of the localization task. Moreover, deep learning methods bear the promise of being more robust to model mismatches in contrast to the conventional model based approaches. As a consequence, we can then study the relation between those classical methods and the new deep learning based methods and analyze the achievable performance. One key final result will be a deeper understanding of how well the maximum likelihood approach can be applied to ICAS and how much it profits from the combination with modern deep learning techniques.&lt;/p&gt;</description></item><item><title>Excitation Signal Design for THz Channel Sounding and Propagation Parameter Estimation</title><link>https://sebastiansemper.de/pub/2024/gedschold2024signal_design_thz/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/gedschold2024signal_design_thz/</guid><description>&lt;p&gt;In this publication, we analyze how the performance of propagation parameter estimation for THz channel sounding can be improved by the power spectrum design of a multicarrier waveform. To this end, we discuss the Fisher information of the propagation parameters and the corresponding deterministic Cram`er-Rao lower bound (CRB) as well as their relation to the carrier powers of the excitation signal. We use these quantities to design waveforms that improve range estimation. In practice, evaluating the Fisher information requires prior knowledge of the propagation scenario which is usually unavailable. Hence, we propose two solutions which we compare numerically to the classical approach of equal power distribution. The evaluation shows that an optimized power distribution can improve the CRB comparable up to a 4 dB gain in signal-to-noise ratio (SNR) if knowledge about the propagation scenario is available and slightly less if the scenario is unknown.&lt;/p&gt;</description></item><item><title>Grid-Free Harmonic Retrieval and Model Order Selection Using Convolutional Neural Networks</title><link>https://sebastiansemper.de/pub/2024/schieler2024grid_free_ht_conv/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/schieler2024grid_free_ht_conv/</guid><description>&lt;p&gt;Harmonic retrieval techniques are the foundation of radio channel sounding, estimation and modeling. This paper introduces a Deep Learning approach for joint delay- and Doppler estimation from frequency and time samples of a radio channel transfer function. Our work estimates the two-dimensional parameters from a signal containing an unknown number of paths. Compared to existing deep learning-based methods, the signal parameters are not estimated via classification but in a quasi-grid-free manner. This alleviates the bias, spectral leakage, and ghost targets that grid-based approaches produce. The proposed architecture also reliably estimates the number of paths in the measurement. Hence, it jointly solves the model order selection and parameter estimation task. Additionally, we propose a multi-channel windowing of the data to increase the estimator&amp;rsquo;s robustness. We also compare the performance to other harmonic retrieval methods and integrate it into an existing maximum likelihood estimator for efficient initialization of a gradient-based iteration.&lt;/p&gt;</description></item><item><title>Improving the Spatial Correlation Characteristics of Antenna Arrays using Linear Operators and Wide-band Modelling</title><link>https://sebastiansemper.de/pub/2024/miranda2024wideband_correlation/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/miranda2024wideband_correlation/</guid><description>&lt;p&gt;The analysis of wireless communication channels at the mmWave, sub-THz and THz bands gives rise to difficulties in the construction of antenna arrays due to the small maximum inter-element spacing constraints at these frequencies. Arrays with uniform spacing greater than half the wavelength for a certain carrier frequency exhibit aliasing side-lobes in the angular domain, prohibiting non-ambiguous estimates of a propagating wave-front&amp;rsquo;s angle of arrival.In this paper, we present how wide-band modelling of the array response is useful in mitigating this spatial aliasing effect. This approach aims to reduce the grating lobes by exploiting the angle- and frequency-dependent phase-shifts observed in the response of the array to a planar wave-front travelling across it. Furthermore, we propose a method by which the spatial correlation characteristics of an array operating at 33 GHz carrier frequency with an instantaneous bandwidth of 1 GHz can be improved such that the angular-domain side-lobes are reduced by 5-10 dB. This method, applicable to arbitrary antenna array manifolds, makes use of a linear operator that is applied to the base-band samples of the channel transfer function measured in space and frequency domains. By means of synthetically simulated arrays, we show that when operating with a bandwidth of 1 GHz, the use of a derived linear operator applied to the array output results in the spatial correlation characteristics approaching those of the array operating at a bandwidth of 12 GHz. Hence, non-ambiguous angle estimates can be obtained in the field without the use of expensive high-bandwidth RF front-end components.&lt;/p&gt;</description></item><item><title>Misspecification of Multiple Scattering in Scalar Wave Fields and its Impact in Ultrasound Tomography</title><link>https://sebastiansemper.de/pub/2024/perez2024multiple_scattering/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/perez2024multiple_scattering/</guid><description>&lt;p&gt;In this work, we investigate the localization of point-like targets in the presence of multiple scattering. We focus on the often omitted scenario in which measurement data is affected by multiple scattering, and a simpler model is employed in the estimation. We study the impact of such model mismatch by means of the Misspecified Cramer-Rao Bound (MCRB). In numerical simulations inspired by tomographic inspection in Ultrasound Nondestructive Testing (UNDT), the MCRB is shown to correctly describe the estimation variance of localization parameters under misspecification of the wave propagation model. We provide extensive discussion on the utility of the MCRB in the practical task of verifying whether a chosen misspecified model is suitable for localization based on the properties of the maximum likelihood estimator and the nuanced distinction between bias and parameter space differences. Finally, we highlight that careful interpretation is needed whenever employing the classical CRB in the presence of mismatch through numerical examples based on the Born approximation and other simplified propagation models stemming from it. Although motivated by UNDT, the analysis and discussion throughout this work generally apply to localization tasks based on scalar wave field measurements.&lt;/p&gt;</description></item><item><title>Misspecification of Multiple Scattering in Scalar Wave Fields and its Impact in Ultrasound Tomography</title><link>https://sebastiansemper.de/pub/2024/perez2024pre_misspecification/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/perez2024pre_misspecification/</guid><description>&lt;p&gt;In this work, we investigate the localization of targets in the presence of multiple scattering. We focus on the often omitted scenario in which measurement data is affected by multiple scattering, and a simpler model is employed in the estimation. We study the impact of such model mismatch by means of the Misspecified Cram'er-Rao Bound (MCRB). In numerical simulations inspired by tomographic inspection in ultrasound nondestructive testing, the MCRB is shown to correctly describe the estimation variance of localization parameters under misspecification of the wave propagation model. We provide extensive discussion on the utility of the MCRB in the practical task of verifying whether a chosen misspecified model is suitable for localization based on the properties of the maximum likelihood estimator and the nuanced distinction between bias and parameter space differences. Finally, we highlight that careful interpretation is needed whenever employing the classical CRB in the presence of mismatch through numerical examples based on the Born approximation and other simplified propagation models stemming from it.&lt;/p&gt;</description></item><item><title>Super-resolution Localization and Tracking in WiFi Sensing</title><link>https://sebastiansemper.de/pub/2024/wang2024superres_sensing/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/wang2024superres_sensing/</guid><description>&lt;p&gt;Integrated sensing and communication (ISAC) systems have been investigated by the research and standardization communities in the recent past. Accurately localizing the target and tracking the target&amp;rsquo;s movement are critical for numerous smart Internet of Things (IoT) systems (smart manufacturing, smart transportation, etc.). This paper aims to realize super-resolution localization and tracking in WiFi sensing by leveraging the IEEE 802.11ad beamforming training procedure. We leverage the CLEAN-Space-Alternating Generalized Expectation-maximization (CLEAN-SAGE) algorithm on a single beam sweeping cycle for target localization and investigate the targets&amp;rsquo; delays and angle estimation. For tracking moving targets, we design mechanisms to estimate the target&amp;rsquo;s motion, including the target&amp;rsquo;s velocity and motion pattern, such as estimating the target&amp;rsquo;s spatial positions over time to obtain the Doppler shift or tracking its trajectory using a Kalman filter. In order to prove that our approach works effectively, we conduct a thorough performance evaluation study. Our evaluation results confirm that the CLEAN-SAGE algorithm can achieve estimation performance beyond the ISAC system&amp;rsquo;s inherent bandwidth and beamwidth constraints. Furthermore, we provide insights into how system configurations, including antenna size, beam overlap, and the number of iterations in the SAGE algorithm, influence its performance.&lt;/p&gt;</description></item><item><title>Using Temporal Consistency for Compressed Sensing in High-Resolution mmWave Sounding</title><link>https://sebastiansemper.de/pub/2024/semper2024temporal_cs/</link><pubDate>Thu, 01 Jan 1970 02:33:44 +0200</pubDate><guid>https://sebastiansemper.de/pub/2024/semper2024temporal_cs/</guid><description>&lt;p&gt;Switched phased array systems operating at high sample rates generate large amounts of data during measurements of radio channels, but many scenarios contain only few multipath components. Compressed Sensing suggests in these cases Nyquist-rate samples are wasteful in terms of data size. It has been observed that structural parameters, i.e., time of flight and angle of arrival, of the propagation paths are temporally consistent, as they vary slowly in time. Hence, we propose a local temporally consistent signal model that includes delay and angle of arrival, and also their time-derivatives, coherently connecting multiple radio channel snapshots. This allows to use a cyclic compression scheme consisting of a few compression matrices that extract mutually incoherent information from adjacent snapshots. Last, we present an algorithm to extract specular multipath components from these compressive measurements.We verify our findings on simulated data and real measurements. On simulated data we observe a good agreement of the estimates with the available ground-truth and show that the proposed cyclic compression scheme improves estimation accuracy. On measured data, we compare the estimates from data obtained at Nyquist rate to data compressed to 17% of the original size and find good agreement as well.&lt;/p&gt;</description></item><item><title>4D Joint Harmonic Retrieval and Model Order Estimation with Convolutional Neural Networks</title><link>https://sebastiansemper.de/pub/2023/schieler2023joint_harmonic_4d/</link><pubDate>Thu, 01 Jan 1970 02:33:43 +0200</pubDate><guid>https://sebastiansemper.de/pub/2023/schieler2023joint_harmonic_4d/</guid><description>&lt;p&gt;Harmonic retrieval is essential in radio channel sounding, estimation, and modeling. In our previous work, we proposed a CNN-based approach combined with additional steps on the likelihood function. This paper extends the approach to perform joint 4D harmonic retrieval by utilizing the samples from a multi-antenna receiver in frequency, time, and the spatial domains of a radio channel transfer function. The proposed architecture also reliably estimates the number of spectral components on measurements. Hence, our approach can estimate four-dimensional parameters from a signal without prior knowledge of the unknown number of paths. Therefore, the architecture jointly solves the model order selection problem and the parameter estimation task in 4D&lt;/p&gt;</description></item><item><title>A Framework for Developing and Evaluating Algorithms for Estimating Multipath Propagation Parameters from Channel Sounder Measurements</title><link>https://sebastiansemper.de/pub/2023/sayeed2023framework/</link><pubDate>Thu, 01 Jan 1970 02:33:43 +0200</pubDate><guid>https://sebastiansemper.de/pub/2023/sayeed2023framework/</guid><description>&lt;p&gt;A framework is proposed for developing and evaluating algorithms for extracting multipath propagation components (MPCs) from measurements collected by channel sounders at millimeter-wave frequencies. Sounders equipped with an omni-directional transmitter and a receiver with a uniform planar array (UPA) are considered. An accurate mathematical model is developed for the spatial frequency response of the sounder that incorporates the non-ideal cross-polar beampatterns for the UPA elements. Due to the limited Field-of-View (FoV) of each element, the model is extended to accommodate multi-FoV measurements in distinct azimuth directions. A beamspace representation of the spatial frequency response is leveraged to develop three progressively complex algorithms aimed at solving the single-snapshot maximum likelihood estimation problem: greedy matching pursuit (CLEAN), space-alternative generalized expectation-maximization (SAGE), and RiMAX. The first two are based on purely specular MPCs whereas RiMAX also accommodates diffuse MPCs. Two approaches for performance evaluation are proposed, one with knowledge of ground truth parameters, and one based on reconstruction mean-squared error. The three algorithms are compared through a demanding channel model with hundreds of MPCs and through real measurements. The results demonstrate that CLEAN gives quite reasonable estimates which are improved by SAGE and RiMAX. Lessons learned and directions for future research are discussed.&lt;/p&gt;</description></item><item><title>Dynamic delay-dispersive UWB-Radar Targets: Modeling and Estimation</title><link>https://sebastiansemper.de/pub/2023/gedschold2023widemax/</link><pubDate>Thu, 01 Jan 1970 02:33:43 +0200</pubDate><guid>https://sebastiansemper.de/pub/2023/gedschold2023widemax/</guid><description>&lt;p&gt;This publication proposes a parametric data model and a gradient-based maximum likelihood estimator suitable for the description of delay-dispersive responses of multiple dynamic ultrawideband (UWB)-radar targets. The target responses are estimated jointly with the global target parameters range and velocity. The large relative bandwidth of UWB has consequences for model-based parameter estimation. On the one hand, the Doppler effect leads to a dispersive response in the Doppler spectrum and to a coupling of the target parameters that both need to be considered during modeling and estimation. On the other hand, the shape of an extended target results in a dispersive response in range, which can be resolved by the radar resolution. We consider this extended response as a parameter of interest, e.g., for the purpose of target recognition. Hence, we propose an efficient description and estimation of it by a finite impulse response (FIR) structure only imposing a restriction on the target&amp;rsquo;s dispersiveness in range. We evaluate the approach on simulations, compare it to state-of-the-art solutions, and provide a validation of the FIR model on measurements of a static scenario.&lt;/p&gt;</description></item><item><title>High-Resolution Parameter Estimation for Wideband Radio Channel Sounding</title><link>https://sebastiansemper.de/pub/2023/semper2023wideband_channel_sounding/</link><pubDate>Thu, 01 Jan 1970 02:33:43 +0200</pubDate><guid>https://sebastiansemper.de/pub/2023/semper2023wideband_channel_sounding/</guid><description>&lt;p&gt;Multidimensional channel sounding measures the geometrical structure of mobile radio propagation. The parameters of a multipath data model in terms of directions, time-of-flight, and Doppler shift are estimated from observations in frequency, time, and space. A maximum likelihood estimation framework allows joint high resolution in all dimensions. The prerequisite for this is an appropriate parametric data model that represents the multipath propagation correctly. At the same time, a device data model is necessary that typically results from calibration measurements. The used model should be as simple as possible, since its structure has a considerable effect on the estimation effort. For instance, the inherent effort in parameter search is reduced if the influence of the parameters is kept independent. Therefore, the data model is characterized by several approximations. The most important is the ``narrowband assumption , &amp;rsquo;&amp;rsquo; which assumes a low relative bandwidth and also avoids considering any frequency response in magnitude and phase. We extend the well-known multidimensional Richter maximization approach (RIMAX) parameter estimation framework by including proper frequency responses. The advantage reveals itself with high bandwidth in the mmWave and sub-THz range. It allows for a more realistic modeling of antenna arrays, and it breaks with the usual narrowband model and allows a better modeling of mutual coupling and time delay effects. If the interacting object extends over several delay bins (hence, an extended target in radar terminology), we propose a model that assigns a short delay spread and a frequency response to the propagation path that associates it with the respective object. We verify the validity of the device model by numerical experiments on simulated and measured antenna data and compare it with RIMAX. In addition, we use synthetic data based on ray-tracing results and measurements both ranging from 27.0 to 33GHz with known ground-truth information and show that the proposed estimator delivers better performance for higher relative bandwidths than the conventional RIMAX implementation.&lt;/p&gt;</description></item><item><title>Misspecification under the Narrowband Assumption: A Cramér--Rao Bound Perspective</title><link>https://sebastiansemper.de/pub/2023/semper2023mcrb/</link><pubDate>Thu, 01 Jan 1970 02:33:43 +0200</pubDate><guid>https://sebastiansemper.de/pub/2023/semper2023mcrb/</guid><description>&lt;p&gt;To efficiently extract estimates about the propagation behavior of electromagnetic waves in a radio environment it is common to invoke the narrowband-assumption. It essentially states that the relative bandwidth of the measurement system is so low that the frequency response of a single propagation path only depends on it Time-of-Flight and the response of the measurement device can be calibrated independently of the measured channel. Recent advances into higher relative bandwidths and antenna arrays with larger spatial aperture render this assumption less likely to be satisfied, which leads to a model mismatch during estimation. In this case estimates are inherently biased and have a special statistical behavior. This behavior can be captured by the so-called Misspecified Cram'er-Rao Bound, which formulates a lower bound for the variance of estimates that are biased due to model mismatch. We analyze this bound in contrast to the traditional Cram'er-Rao Bound and show the shortcomings in the setting of joint ToF-DoA estimation in the mmWave spectrum. The conducted numerical studies also show that planar array geometries inherently suffer from violation of the narrowband assumption irrespective of the individual elements&amp;rsquo; frequency response, whereas circular structures show it to a lesser degree.&lt;/p&gt;</description></item><item><title>Time Variant Directional Multi-Link Channel Sounding and Estimation for V2X</title><link>https://sebastiansemper.de/pub/2023/stanko2023dynkavi_measurements/</link><pubDate>Thu, 01 Jan 1970 02:33:43 +0200</pubDate><guid>https://sebastiansemper.de/pub/2023/stanko2023dynkavi_measurements/</guid><description>&lt;p&gt;We present our new scalable multi-channel and multi-node sounder , the ILMSound G3. It is configurable in terms of the number of switched Tx and parallel Rx nodes. The basic structures of the Tx and Rx nodes are given with consideration of the measurement system requirements. The ILMSound G3 is validated via a proof of concept measurement at 2.53 GHz in an urban environment. The system was configured as a sounder with two Tx nodes and one Rx node, which records two MIMO links simultaneously. Each node uses an antenna array. The transmitters were moved by cars whereas the receiver was elevated 20 m above the ground by a lifting platform acting as a base station. The used system configuration results in a snapshot rate of approx. 0.5 kHz, which covers the expected time variance of the chosen environment.The measurement results were analyzed using HRPE, providing a parametric description of the specular propagation paths of the radio channel per measurement link. Each estimated path is described by its directions of departure and arrival, delay, Doppler shift, and fully polarimetric complex path weights.&lt;/p&gt;</description></item><item><title>A Frequency Domain Approach for Estimating the Angular Dependent Delay of an UWB Antenna</title><link>https://sebastiansemper.de/pub/2022/rashidifar2022freq_dep_ant/</link><pubDate>Thu, 01 Jan 1970 02:33:42 +0200</pubDate><guid>https://sebastiansemper.de/pub/2022/rashidifar2022freq_dep_ant/</guid><description>&lt;p&gt;The procedure of estimating the angular dependent delay of an Ultra Wideband (UWB) antenna, directly in frequency domain, is presented. It is shown that in this approach any occurrences that can lead to inaccurate results, such as the windowing effect, can be entirely avoided. It alleviates the need to design a specific pulse, since the transfer function can be obtained by a frequency sweep. Moreover, polarization mismatch is accounted in the analysis. To this end, an UWB monopole antenna is designed and it is simulated with time and frequency domain solvers. The simulation results demonstrate that only frequency domain analysis is sufficient for the delay estimation. The outcome of this work provides a clear procedure for obtaining the angular dependent delay of an antenna in an accurate way.&lt;/p&gt;</description></item><item><title>Efficient algorithms and data structures for compressive sensing</title><link>https://sebastiansemper.de/pub/2022/semper2022dissertation/</link><pubDate>Thu, 01 Jan 1970 02:33:42 +0200</pubDate><guid>https://sebastiansemper.de/pub/2022/semper2022dissertation/</guid><description>&lt;p&gt;Along with the ever increasing number of sensors, which are also generating rapidly growing amounts of data, the traditional paradigm of sampling adhering the Nyquist criterion is facing an equally increasing number of obstacles. The rather recent theory of Compressive Sensing (CS) promises to alleviate some of these drawbacks by proposing to generalize the sampling and reconstruction schemes such that the acquired samples can contain more complex information about the signal than Nyquist samples. The proposed measurement process is more complex and the reconstruction algorithms necessarily need to be nonlinear. Additionally, the hardware design process needs to be revisited as well in order to account for this new acquisition scheme. Hence, one can identify a trade-off between information that is contained in individual samples of a signal and effort during development and operation of the sensing system. This thesis addresses the necessary steps to shift the mentioned trade-off more to the favor of CS. We do so by providing new results that make CS easier to deploy in practice while also maintaining the performance indicated by theoretical results. The sparsity order of a signal plays a central role in any CS system. Hence, we present a method to estimate this crucial quantity prior to recovery from a single snapshot. As we show, this proposed Sparsity Order Estimation method allows to improve the reconstruction error compared to an unguided reconstruction. During the development of the theory we notice that the matrix-free view on the involved linear mappings offers a lot of possibilities to render the reconstruction and modeling stage much more efficient. Hence, we present an open source software architecture to construct these matrix-free representations and showcase its ease of use and performance when used for sparse recovery to detect defects from ultrasound data as well as estimating scatterers in a radio channel using ultra-wideband impulse responses. For the former of these two applications, we present a complete reconstruction pipeline when the ultrasound data is compressed by means of sub-sampling in the frequency domain. Here, we present the algorithms for the forward model, the reconstruction stage and we give asymptotic bounds for the number of measurements and the expected reconstruction error. We show that our proposed system allows significant compression levels without substantially deteriorating the imaging quality. For the second application, we develop a sampling scheme to acquire the channel Impulse Response (IR) based on a Random Demodulator that allows to capture enough information in the recorded samples to reliably estimate the IR when exploiting sparsity. Compared to the state of the art, this in turn allows to improve the robustness to the effects of time-variant radar channels while also outperforming state of the art methods based on Nyquist sampling in terms of reconstruction error. In order to circumvent the inherent model mismatch of early grid-based compressive sensing theory, we make use of the Atomic Norm Minimization framework and show how it can be used for the estimation of the signal covariance with R-dimensional parameters from multiple compressive snapshots. To this end, we derive a variant of the ADMM that can estimate this covariance in a very general setting and we show how to use this for direction finding with realistic antenna geometries. In this context we also present a method based on a Stochastic gradient descent iteration scheme to find compression schemes that are well suited for parameter estimation, since the resulting sub-sampling has a uniform effect on the whole parameter space. Finally, we show numerically that the combination of these two approaches yields a well performing grid-free CS pipeline.&lt;/p&gt;</description></item><item><title>Estimating Multi-Modal Dense Multipath Components using Auto-Encoders</title><link>https://sebastiansemper.de/pub/2022/schieler2022mcdeep/</link><pubDate>Thu, 01 Jan 1970 02:33:42 +0200</pubDate><guid>https://sebastiansemper.de/pub/2022/schieler2022mcdeep/</guid><description>&lt;p&gt;We present a maximum-likelihood estimation algorithm for radio channel measurements exhibiting a mixture of independent Dense Multipath Components. The novelty of our approach is in the algorithms initialization using a deep learning architecture. Currently, available approaches can only deal with scenarios where a single mode is present. However, in measurements, two or more modes are often observed. This much more challenging multi-modal setting bears two important questions: How many modes are there, and how can we estimate those? To this end, we propose a Neural Net-architecture that can reliably estimate the number of modes present in the data and also provide an initial assessment of their shape. These predictions are used to initialize for gradient- and model-based optimization algorithm to further refine the estimates. We demonstrate numerically how the presented architecture performs on measurement data and analytically study its influence on the estimation of specular paths in a setting where the single-modal approach fails.&lt;/p&gt;</description></item><item><title>fastmat: Efficient linear transforms in Python</title><link>https://sebastiansemper.de/pub/2022/wagner2021fastmat/</link><pubDate>Thu, 01 Jan 1970 02:33:42 +0200</pubDate><guid>https://sebastiansemper.de/pub/2022/wagner2021fastmat/</guid><description>&lt;p&gt;Scientific computing requires handling large linear models, which are often composed of structured matrices. With increasing model size, dense representations quickly become infeasible to compute or store. Matrix-free implementations are suited to mitigate this problem at the expense of additional implementation overhead, which complicates research and development effort by months, when applied to practical research problems. Fastmat is a framework for handling large structured matrices by offering an easy-to-use abstraction model. It allows for the expression of matrix-free linear operators in a mathematically intuitive way, while retaining their benefits in computation performance and memory efficiency. A built-in hierarchical unit-test system boosts debugging productivity and run-time execution path optimization improves the performance of highly-structured operators. The architecture is completed with an interface for abstractly describing algorithms that apply such matrix-free linear operators, while maintaining clear separation of their respective implementation levels. Fastmat achieves establishing a close relationship between implementation code and the actual mathematical notation of a given problem, promoting readable, portable and re-usable scientific code.&lt;/p&gt;</description></item><item><title>Verification of THz Channel Sounder and Delay Estimation with Over-The-Air Multipath Artifact</title><link>https://sebastiansemper.de/pub/2022/dupleich2022thz_artifact/</link><pubDate>Thu, 01 Jan 1970 02:33:42 +0200</pubDate><guid>https://sebastiansemper.de/pub/2022/dupleich2022thz_artifact/</guid><description>&lt;p&gt;In the present paper we introduce the empirical results of measurements with an over-the-air based propagation artifact for verification and validation of sub-THz and THz channel sounders and parameter estimation algorithms. This experiment produces a fixed number of multipath components with traceable propagation properties in the different domains that can be used to test resolution and performance. Because of the inherent characteristics of the measurement hardware, we have introduced an adaptation on a parametric high resolution estimation algorithm to account the imperfections of the channel sounder. The results have shown to account for a relative good performance of the sounder and the tested parametric and non-parametric estimation algorithms.&lt;/p&gt;</description></item><item><title>Compressed Ultrasound Computed Tomography in NDT</title><link>https://sebastiansemper.de/pub/2021/perez2021tomography/</link><pubDate>Thu, 01 Jan 1970 02:33:41 +0200</pubDate><guid>https://sebastiansemper.de/pub/2021/perez2021tomography/</guid><description>&lt;p&gt;Ultrasound Computed Tomography (UCT) is challenging due to phenomena such as strong refraction, multiple scattering, and mode conversion. In NDT, large speed of sound contrasts lead to strong artifacts if such phenomena are not modeled correctly; however, enhanced models are computationally expensive. In this work, a two-step framework for Compressed UCT based on the integral approach to the solution of the Helmholtz equation is presented. It comprises a physically motivated forward step and an imaging step that solves a suitable inverse problem. Multiple scattering is accounted for through the use of Neumann series. Convergence problems of Neumann series in high contrast settings are addressed via Pad'e approximants. Compressed sensing is employed to reduce the computational complexity of the reconstruction procedure by reducing data volumes directly at the measurement step, avoiding redundancy in the data and allowing the ability to steer the admissible computational effort at the expense of reconstruction quality. The proposed method is shown to yield high quality reconstructions under heavy subsampling in the frequency and spatial domains.&lt;/p&gt;</description></item><item><title>Frequency Subsampling of Ultrasound Nondestructive Measurements: Acquisition, Reconstruction, and Performance</title><link>https://sebastiansemper.de/pub/2021/kirchhof2020fouriersubsampling/</link><pubDate>Thu, 01 Jan 1970 02:33:41 +0200</pubDate><guid>https://sebastiansemper.de/pub/2021/kirchhof2020fouriersubsampling/</guid><description>&lt;p&gt;In ultrasound nondestructive testing (NDT), a widespread approach is to take synthetic aperture measurements from the surface of a specimen to detect and locate defects within it. Based on these measurements, imaging is usually performed using the synthetic aperture focusing technique (SAFT). However, SAFT is suboptimal in terms of resolution and requires oversampling in the time domain to obtain a fine grid for the delay-and-sum (DAS). On the other hand, parametric reconstruction algorithms give better resolution, but their usage for imaging becomes computationally expensive due to the size of the parameter space and a large amount of measurement data in realistic 3-D scenarios when using oversampling. In the literature, the remedies to this are twofold. First, the amount of measurement data can be reduced using state-of-the-art sub-Nyquist sampling approaches to measure Fourier coefficients instead of time-domain samples. Second, parametric reconstruction algorithms mostly rely on matrix–vector operations that can be implemented efficiently by exploiting the underlying structure of the model. In this article, we propose and compare different strategies to choose the Fourier coefficients to be measured. Their asymptotic performance is compared by numerically evaluating the Cram'er–Rao bound (CRB) for the localizability of the defect coordinates. These subsampling strategies are then combined with an l1 -minimization scheme to compute 3-D reconstructions from the low-rate measurements. Compared to conventional DAS, this allows us to formulate a fully physically motivated forward model matrix. To enable this, the projection operations of the forward model matrix are implemented matrix-free by exploiting the underlying two-level Toeplitz structure. Finally, we show that high-resolution reconstructions from as low as a single Fourier coefficient per A-scan are possible based on simulated data and measurements from a steel specimen.&lt;/p&gt;</description></item><item><title>Hardware Architecture for Ultra-Wideband Channel Impulse Response Measurements Using Compressed Sensing</title><link>https://sebastiansemper.de/pub/2021/wagner2020csuwb/</link><pubDate>Thu, 01 Jan 1970 02:33:41 +0200</pubDate><guid>https://sebastiansemper.de/pub/2021/wagner2020csuwb/</guid><description>&lt;p&gt;We propose a compact hardware architecture for measuring sparse channel impulse responses (IR) by extending the M-Sequence ultra-wideband (UWB) measurement principle with the concept of compressed sensing. A channel is excited with a periodic M-sequence and its response signal is observed using a Random Demodulator (RD), which observes pseudo-random linear combinations of the response signal at a rate significantly lower than the measurement bandwidth. The excitation signal and the RD mixing signal are generated from compactly implementable Linear Feedback Shift registers (LFSR) and operated from a common clock. A linear model is derived that allows retrieving an IR from a set of observations using Sparse-Signal-Recovery (SSR). A Matrix-free model implementation is possible due to the choice of synchronous LFSRs as signal generators, resulting in low computational complexity. For validation, real measurement data of a time-variant channel containing multipath components is processed by simulation models of our proposed architecture and the classic M-Sequence method. We show successful IR recovery using our architecture and SSR, outperforming the classic method significantly in terms of IR measurement rate. Compared to the classic method, the proposed architecture allows faster measurements of sparse time-varying channels, resulting in higher Doppler tolerance without increasing hardware or data stream complexity.&lt;/p&gt;</description></item><item><title>Cramér-Rao Bounds for Flaw Localization in Subsampled Multistatic Multichannel Ultrasound Ndt Data</title><link>https://sebastiansemper.de/pub/2020/perez2020crbsubsampling/</link><pubDate>Thu, 01 Jan 1970 02:33:40 +0200</pubDate><guid>https://sebastiansemper.de/pub/2020/perez2020crbsubsampling/</guid><description>&lt;p&gt;The localization of defects is a prevalent task in ultrasound nondestructive testing. Multi-channel techniques like Full Matrix Capture (FMC) measurements are employed in this regard for their better spatial accuracy compared to single-channel synthetic aperture measurements at the expense of larger data volumes and increased measurement time. In this paper, we analyze a compressed sensing scenario in which location parameters of point-like scatterers are estimated from subsampled FMC data. Particularly, the impact of the specific choice of Tx and Rx elements is studied by means of the Cram'er-Rao Bound (CRB). We derive the CRB of lateral and vertical position of the scatterers estimated from FMC data, as well as expressions for the CRB that arise in the far-field scenario. These expressions are useful for two reasons. First, they provide insights about the impact of number and location of channels on the localization performance. Second, we can use them to optimize the sensor positions in the subsampled array, which we demonstrate by introducing a CRB-based array design technique. The far-field expressions reveal that only two channels are required for the CRB of the lateral case to become finite, and also indicate a far-field gain when using a larger subsampled array.&lt;/p&gt;</description></item><item><title>eadf: Representation of far-field antenna responses in Python</title><link>https://sebastiansemper.de/pub/2020/semper2020eadf/</link><pubDate>Thu, 01 Jan 1970 02:33:40 +0200</pubDate><guid>https://sebastiansemper.de/pub/2020/semper2020eadf/</guid><description>&lt;p&gt;Understanding and analysis of antennas is crucial for the development of modern telecommunication technologies, like 5G or massive MIMO. Research in this area has the need for software to measure, analyze and test antenna systems. For instance, one needs a convenient way to transform large amounts of antenna measurement data into a concise representation. The herein described eadf package aims at providing this functionality. On top of that, it offers routines to interpolate the beam-pattern of an antenna as well as compressing it in spatial frequency domain. As such, it can be used for the development of high resolution parameter estimation algorithms and analysis of antenna beam-patterns. In summary, the package provides a flexible and extensible framework for dealing with antennas.&lt;/p&gt;</description></item><item><title>A comparative investigation on the use of compressive sensing methods in computational ghost imaging</title><link>https://sebastiansemper.de/pub/2019/zhang2019ghostimaging/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/zhang2019ghostimaging/</guid><description>&lt;p&gt;Usually, a large number of patterns are needed in the computational ghost imaging (CGI). In this work, the possibilities to reduce the pattern number by integrating compressive sensing (CS) algorithms into the CGI process are systematically investigated. Based on the different combinations of sampling patterns and image priors for the L1-norm regularization, different CS-based CGI approaches are proposed and implemented with the iterative shrinkage thresholding algorithm. These CS-CGI approaches are evaluated with various test scenes. According to the quality of the reconstructed images and the robustness to measurement noise, a comparison between these approaches is drawn for different sampling ratios, noise levels, and image sizes.&lt;/p&gt;</description></item><item><title>ADMM for ND Line Spectral Estimation Using Grid-free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation</title><link>https://sebastiansemper.de/pub/2019/semper2018admm_lse/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/semper2018admm_lse/</guid><description>&lt;p&gt;This paper is concerned with estimating unknown multidimensional frequencies from linear compressive measurements. This is accomplished by employing the recently proposed atomic norm minimization framework to recover these frequencies under a sparsity prior without imposing any grid restriction on these frequencies. To this end, we give a rigorous derivation of an iterative scheme called alternating direction of multipliers method, which is able to incorporate multiple compressive snapshots from a multi-dimensional superposition of complex harmonics. The key result here is how to formulate the objective function minimized by this scheme and its partial derivatives, which become hard to manage if the dimensionality of the frequencies is larger than 1. Moreover we demonstrate the performance of this approach in case of 3D line spectral estimation and 2D DOA estimation with a synthetic antenna array.&lt;/p&gt;</description></item><item><title>Combining Matrix Design for 2D DoA Estimation with Compressive Antenna Arrays Using Stochastic Gradient Descent</title><link>https://sebastiansemper.de/pub/2019/pawar2018sgddoaopt/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/pawar2018sgddoaopt/</guid><description>&lt;p&gt;Recently, compressive antenna arrays have been considered for direction of arrival (DoA) estimation with reduced hardware complexity. By utilizing compressive sensing, such arrays employ a linear combining network to combine signals from a larger set of antenna elements in the analog RF domain. In this paper, we develop a design approach based on the minimization of error between spatial correlation function (SCF) of the compressive and the uncompressed array resulting in the estimation performance of the two arrays to be as close as possible. The proposed design is based on grid-free stochastic gradient descent (SGD) optimization. In addition to a low computational cost for the proposed method, we show numerically that the resulting combining matrices perform better than the ones generated by a previous approach and combining matrices generated from a Gaussian ensemble.&lt;/p&gt;</description></item><item><title>Defect Detection From Compressed 3-D Ultrasonic Frequency Measurements</title><link>https://sebastiansemper.de/pub/2019/semper20193dfista/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/semper20193dfista/</guid><description>&lt;p&gt;In this paper, we propose a compressed sensing scheme for volumetric synthetic aperture measurements in ultrasonic nondestructive testing. The compression is achieved by limiting the measurement to a subset of the Fourier coefficients of the full measurement data, where we also address the issue of a suitable hardware architecture for the task. We present a theoretic analysis for one of the proposed schemes in terms of the Restricted Isometry Property and derive a scaling law for the lower bound of the number of necessary measurements for stable and efficient recovery. We verify our approach with reconstructions from measurement data of a steel specimen that was compressed synthetically in software. As a side result, our approach yields a variant of the 3-D Synthetic Aperture Focusing Technique which can deal with compressed data.&lt;/p&gt;</description></item><item><title>Method for the acquisition of impulse responses, e.g. for ultra-wideband systems (pending)</title><link>https://sebastiansemper.de/pub/2019/wagner2019csuwbpatent/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/wagner2019csuwbpatent/</guid><description>&lt;p&gt;There are disclosed techniques (e.g., apparatus, methods) for estimating an Impulse response (30) of a linear System (4). An apparatus (100) is configured to generate (7) a transmit Signal (23) on the basis of a first sequence. The apparatus (100) is configured to obtain a receive Signal (24) and to multiply (9) the receive Signal (24) with a second sequence (25), to obtain a modified receive Signal (26), wherein the second sequence (25) is different from the first sequence. The apparatus (100) is configured to analog-to-digital, ADC, convert (11) an Integration result (27) in order to obtain a sample value (28), the Integration result (27) being based on an Integration (10) of the modified receive Signal (26) over a period (Ts) of time. The apparatus (100) is configured to obtain an estimate (30) of the Impulse response on the basis of a plurality (29) of sample values (28).&lt;/p&gt;</description></item><item><title>Packing Bounds for Outer Products with Applications to Compressive Sensing</title><link>https://sebastiansemper.de/pub/2019/semper2019khatriraopack/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/semper2019khatriraopack/</guid><description>&lt;p&gt;In order to obtain good reconstruction guarantees for typical compressive sensing scenarios, we translate the search for good compression matrices into a ball packing problem in a suitable projective space. We then derive such reconstruction guarantees for two relevant scenarios, one where the matrices are unstructured and one where they have to be Khatri-Rao products. Finally, we demonstrate how the proposed method can be implemented with a physically motivated numerical optimization scheme, and how it compares to a conventional scheme of random compression matrices.&lt;/p&gt;</description></item><item><title>Total Focusing Method with Subsampling in Space and Frequency Domain for Ultrasound NDT</title><link>https://sebastiansemper.de/pub/2019/perez2019tfmsubsampling/</link><pubDate>Thu, 01 Jan 1970 02:33:39 +0200</pubDate><guid>https://sebastiansemper.de/pub/2019/perez2019tfmsubsampling/</guid><description>&lt;p&gt;In this paper, we present a compressed sensing model for 2D Full Matrix Capture data from a uniform linear array. Data is reconstructed via a matrix-free implementation of the Total Focusing Method (TFM) combined with the Fast Iterative Shrinkage/ Thresholding Algorithm. This results in reduced measurement times and data volumes with-out sacrificing image quality. Our approach is com-pared to standard TFM by applying the techniques on real measurement data, both synthetically compressed and complete.&lt;/p&gt;</description></item><item><title>Compressive spatial channel sounding</title><link>https://sebastiansemper.de/pub/2018/ibrahim2018compressivechannel/</link><pubDate>Thu, 01 Jan 1970 02:33:38 +0200</pubDate><guid>https://sebastiansemper.de/pub/2018/ibrahim2018compressivechannel/</guid><description>&lt;p&gt;In this paper we investigate the application of Compressed Sensing (CS) to MIMO channel sounding in the spatial domain. A compressive spatial channel sounder is proposed and evaluated based on real scenarios showing advantages in terms of time, hardware complexity and resolution. In particular, in the case where we use time division duplex for measuring the MIMO channel (in the form of antenna switching at the transmitter and/or the receiver), the proposed approach reduces the total number of switching periods, which implies a reduced channel acquisition time and thus an improved Doppler bandwidth. Alternatively, if we use multiple receive RF chains for the measurement, the compression allows to reduce the number of RF chains, which is a relevant advantage in terms of the overall receiver complexity, the amount of data to be processed in the digital domain (e.g., FPGA), power consumption, as well as RF hardware calibration. On the other hand, for the same measurement time and/or hardware complexity, one can increase the number of array elements to cover a larger aperture and so achieving better performance in terms of resolution.&lt;/p&gt;</description></item><item><title>Defect Detection from 3D Ultrasonic Measurements Using Matrix-free Sparse Recovery Algorithms</title><link>https://sebastiansemper.de/pub/2018/semper2018defectdetection/</link><pubDate>Thu, 01 Jan 1970 02:33:38 +0200</pubDate><guid>https://sebastiansemper.de/pub/2018/semper2018defectdetection/</guid><description>&lt;p&gt;In this paper, we propose an efficient matrix-free algorithm to reconstruct locations and size of flaws in a specimen from volumetric ultrasound data by means of a native 3D Sparse Signal Recovery scheme using Orthogonal Matching Pursuit (OMP). The efficiency of the proposed approach is achieved in two ways. First, we formulate the dictionary matrix as a block multilevel Toeplitz matrix to minimize redundancy and thus memory consumption. Second, we exploit this specific structure in the dictionary to speed up the correlation step in OMP, which is implemented matrix-free. We compare our method to state-of-the-art, namely 3D Synthetic Aperture Focusing Technique , and show that it delivers a visually comparable performance, while it gains the additional freedom to use further methods such as Compressed Sensing.&lt;/p&gt;</description></item><item><title>GPU-Accelerated Matrix-Free 3D Ultrasound Reconstruction for Nondestructive Testing</title><link>https://sebastiansemper.de/pub/2018/kirchhof2018gpuus/</link><pubDate>Thu, 01 Jan 1970 02:33:38 +0200</pubDate><guid>https://sebastiansemper.de/pub/2018/kirchhof2018gpuus/</guid><description>&lt;p&gt;In this paper, we propose a matrix-free 3D ultrasonic reconstruction scheme based on the Fast Iterative Shrinkage-Thresholding algorithm (FISTA) implemented on a GPU. The matrix-free implementation allows to reconstruct images even for problem sizes that would be intractable when explicitly calculating the matrix. However, due to the matrix-free implementation, additional steps are necessary to estimate the stepsize parameter required by FISTA, since the optimal stepsize depends on the largest singular value of the operator matrix, which in the matrix-free version is unavailable and cannot be built due to its size. The estimation is performed based on a priori knowledge of the model. We compare our method to 3D SAFT and OMP images of volumetric ultrasound measurement data of a steel specimen to show how FISTA leads to sharper images facilitating sizing and locating of defects within a specimen.&lt;/p&gt;</description></item><item><title>Grid-Free Direction-of-Arrival Estimation with Compressed Sensing and Arbitrary Antenna Arrays</title><link>https://sebastiansemper.de/pub/2018/semper2018doaanm/</link><pubDate>Thu, 01 Jan 1970 02:33:38 +0200</pubDate><guid>https://sebastiansemper.de/pub/2018/semper2018doaanm/</guid><description>&lt;p&gt;We study the problem of direction of arrival estimation for arbitrary antenna arrays. We formulate it as a continuous line spectral estimation problem and solve it under a sparsity prior without any gridding assumptions. Moreover, we incorporate the array&amp;rsquo;s beampattern in form of the Effective Aperture Distribution Function (EADF), which allows to use arbitrary (synthetic as well as measured) antenna arrays. This generalizes known atomic norm based grid-free DOA estimation methods (that have so far been limited to uniformly spaced arrays) to arbitrary antenna arrays. In addition, our formulation allows to incorporate compressed sensing in form of special linear combinations of the antennas&amp;rsquo; output ports. We provide conditions for the successful reconstruction of a certain number of targets depending on the amount of compression and the EADF of the antenna array. Our results are applicable to measurement matrices from any sub-Gaussian distribution.&lt;/p&gt;</description></item><item><title>Sparsity Order Estimation From a Single Compressed Observation Vector</title><link>https://sebastiansemper.de/pub/2018/semper2018soe/</link><pubDate>Thu, 01 Jan 1970 02:33:38 +0200</pubDate><guid>https://sebastiansemper.de/pub/2018/semper2018soe/</guid><description>&lt;p&gt;In this paper, the problem of estimating the unknown degree of sparsity from compressive measurements without the need to carry out a sparse recovery step is investigated. While the sparsity order can be directly inferred from the effective rank of the observation matrix in the multiple snapshot case, this appears to be impossible in the more challenging single snapshot case. It is shown that specially designed measurement matrices allow to rearrange the measurement vector into a matrix such that its effective rank coincides with the effective sparsity order. In fact, it is proven that matrices that are composed of a Khatri-Rao product of smaller matrices generate measurements that allow to infer the sparsity order. Moreover, if some samples are used more than once, one of the matrices needs to be Vandermonde. These structural constraints reduce the degrees of freedom in choosing the measurement matrix, which may incur in a degradation in the achievable coherence. Thus, this paper also addresses suitable choices of the measurement matrices. In particular, Khatri-Rao and Vandermonde matrices are analyzed in terms of their coherence and a new design for Vandermonde matrices that achieves a low coherence is proposed.&lt;/p&gt;</description></item><item><title>Vita</title><link>https://sebastiansemper.de/vita/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://sebastiansemper.de/vita/</guid><description>&lt;figure class="floatImage"&gt;&lt;img src="https://sebastiansemper.de/me.webp"&gt;
&lt;/figure&gt;

&lt;p&gt;I studied mathematics at Technische Universität Ilmenau, (TU Ilmenau), Ilmenau, Germany and received the Bachelor and Master of Science degrees in 2013 and 2015, respectively.
Since 2015, I have been a Research Assistant with the Electronic Measurements and Signal Processing Group, which is a joint research activity between the Fraunhofer Institute for Integrated Circuits IIS, Ilmenau, and TU Ilmenau, Ilmenau.
In 2022 I finished my doctoral studies and received the doctoral degree (Dr.-Ing.) with honors in electrical engineering.
Since then, I have been a post doctoral student in the Electronic Measurements and Signal Processing Group.
In 2023 I spent six months with the Wireless Division at the National Institute of Standards and Technology (NIST) in Gaithersburg MD, where I worked on various channel sounding systems and related algorithms.
Since 2024 I have been leading the &amp;ldquo;Signal-Processing with Efficient Algorithms and Learning&amp;rdquo; (SEAL) group at Fraunhofer IIS, Ilmenau, Germany.
My research interests consist of signal processing for radio channels, parameter estimation, integrated communications and sensing, optimization, numerical methods algorithm design, artifical intelligence and compressive sensing.&lt;/p&gt;</description></item></channel></rss>