Picture of Johannes Rosenberger

M.Sc. Johannes Rosenberger

Technical University of Munich

Chair of Communications Engineering (Prof. Kramer)

Postal address

Postal:
Theresienstr. 90
80333 München

Biography

I received the Bachelor's degree in 2019 and my Master's degree in 2021, both from Technical University of Munich (TUM). My focus in the Bachelor was mainly on High Frequency Engineering. Later, my interests shifted towards Information and Coding Theory. I wrote my Bachelor's thesis about Coding for DNA Storage at the Professorship for Coding and Cryptography (COD), and during the master I worked on Identification over Compound and Arbitrarily Varying Broadcast Channels at the Chair of Communications Engineering (LNT).

Since October 2021, I am a research assistant at the Chair of Communications Engineering. Currently, I am working on theory for communications systems beyond Shannon's approach in the project 6G-life by the German Federal Ministry of Education and Research.

Research Interests

  • semantic/goal-oriented communication theory
    • message identification over channels (Textbook, Christian's page)
    • computing functions over channels/function compression and connections to (K-)identification
  • multi-user information theory
  • interference, channel uncertainty and active jammers trying to inhibit communication
  • quantum information theory

Teaching

Theses

Available Theses

Theses in Progress

[identification] Typicality verifier in Deterministic Identification

Keywords:
identification, typical, sequences, information theory, encoder, decoder, verifier

Description

Identification is a communication scheme that allows rate advantages over transmission, with the tradeoff that identities cannot be decoded (as messages do) but can only be verified.
More precicely, while in transmission the receiver tries to determine the message m encoded by the sender, in identification the receiver is interested in his own specific identity i and want to verify whether the sender encoded i or not, without any attempt at figuring out which i' was exactly encoded.

The rates achievable by identification are remarkably dependent on whether stochastic or only deterministic encoders are allowed.
Here, we consider only deterministic identification.

The seminal work that started identificaion with deterministic encoders can be found at

while identification with stochastic encoders was initiated in

Identification come with the possibility of rending feasible the use of typicality.
In transmission, random codes and typicality decoders are impractical due to the exponential cost of defining and running such encoders and decoders, as every conditionally typical set of each possible codeword must be checked.
However, since in identification we only want to verify a specific message, only a single conditionally typical set needs to be verified, opening the possibility or rending typicality a valid and efficient solution for the receiver.

The goal of the project is to study the complexity and capacity of such a typical verifier both analytically and numerically, in the case of q-ary alphabets.

Related works on deterministic identification can be found at

Prerequisites

Knowledge of typical sequences and typical sets is recommended, but not required.

Supervisor:

Securing OFDM Against Jamming Attacks

Keywords:
wireless communication,jamming attacks,robust communication,common randomness

Description

In wireless systems, a jammer may send interfering signals to disrupt legitimate communication. Orthogonal frequency-domain multiplexing (OFDM) is especially vulnerable to jamming attacks [1]. Liang, Ren and Li [1] proposed a modified OFDM scheme which is secured against jamming attacks by introducing randomized phase shifts that are coordinated between sender and receiver. It is known in general [2] that randomized protocols can mitigate even adversarial jamming. True randomness is difficult to get, but a very small amount is sufficient to achieve full channel capacity [3].


The task of the student is to understand the vulnerability of OFDM and to
review the secured OFDM scheme by Liang, Ren and Li [1]. If possible, the student should consider whether the amount of randomness used there can be reduced.



[1] Y. Liang, J. Ren, and T. Li, “Secure OFDM System Design and Capacity
Analysis Under Disguised Jamming,” 2020. doi: 10.1109/TIFS.2019.2929449.
[2] D. Blackwell, L. Breiman, and A. J. Thomasian, “The Capacities of Certain
Channel Classes Under Random Coding,” 1960. doi: 10.1214/aoms/1177705783.
[3] R. Ahlswede, “Elimination of correlation in random codes for arbitrarily
varying channels,” 1978. doi: 10.1007/BF00533053.

Prerequisites

Information Theory

Supervisor:

Publikationen

2022

  • Pereg, Uzi; Rosenberger, Johannes; Deppe, Christian: Identification Over Quantum Broadcast Channels. 2022 IEEE International Symposium on Information Theory (ISIT), 2022 more… Full text ( DOI )
  • Rosenberger, J.: Deterministic Identification and Multi-User Models. Munich Doctoral Seminar on Communications, 2022 more…
  • Rosenberger, J.: Functional Compression with Side Information and Randomization. Munich Doctoral Seminar on Communications, 2022 more…
  • Rosenberger, J.: Deterministic Identification and Multi-User Models. 6G-life Project Meeting WP 1.3 (Post-Shannon), 2022 more…
  • Rosenberger, Johannes; Pereg, Uzi; Deppe, Christian: Identification over Compound MIMO Broadcast Channels. ICC 2022 - IEEE International Conference on Communications, 2022 more…

2021

  • Rosenberger, J.: Local Testability of Codes and Identification. Munich Doctoral Seminar on Communications, 2021 more…