Hauptseminar Digitale Kommunikationssysteme
|Stellung in Studienplänen||Siehe TUMonline|
- 18.10.2018 15:00-16:00 N2408, Seminarraum
25.10.2018 15:00-17:15 N2408, Seminarraum*
- 08.11.2018 15:00-17:15 N2408, Seminarraum
15.11.2018 15:00-17:15 N2408, Seminarraum* 22.11.2018 15:00-17:15 N2408, Seminarraum* 29.11.2018 15:00-17:15 N2408, Seminarraum* 06.12.2018 15:00-17:15 N2408, Seminarraum* 13.12.2018 15:00-17:15 N2408, Seminarraum*
- 20.12.2018 15:00-17:15 N2408, Seminarraum
- 10.01.2019 15:00-17:15 N2408, Seminarraum
- 17.01.2019 15:00-17:15 N2408, Seminarraum
24.01.2019 15:00-17:15 N2408, Seminarraum* 31.01.2019 15:00-17:15 N2408, Seminarraum* 07.02.2019 15:00-17:15 N2408, Seminarraum*
Teilnahmekriterien & Anmeldung
Lehr- und Lernmethoden
Construction and Decoding of Golay Codes
Supervisor: Tobias Prinz
The Golay code has been introduced by M. J. Golay in 1949 as the first perfect binary error correcing code . It was used by NASA during the Voyager 1 and 2 deep space missions to transmit color pictures of Jupiter and Saturn. Many construction methods  and decoding algorithms [3,4] have been proposed for the Golay code since 1949.
The student's task is to give an overview about construction and decoding methods for the Golay code, as well as a performance comparison of the decoding methods and some historical background.
-  M. J. Golay, “Notes on digital coding,” Proc. Inst. Radio Eng., vol. 37, no. 6, p. 657, 1949.
-  X.-H. Peng and P. G. Farrell, “On construction of the (24, 12, 8) Golay codes,” IEEE Trans. Inf. Theory, vol. 52, no. 8, pp. 3669–3675, Aug. 2006.
-  S. Sarangi and S. Banerjee, “Efficient hardware implementation of encoder and decoder for Golay code,” IEEE Trans. Very Large Scale Integr. (VLSI) Syst., vol. 23, no. 9, pp. 1965–1968, Sep. 2015.
- A. Vardy, “Even more efficient bounded-distance decoding of the hexacode, the Golay code, and the Leech lattice,” IEEE Trans. Inf. Theory, vol. 41, no. 5, pp. 1495–1499, Sep. 1995.
Vehicle-to-anything (V2X) communication
Supervisor: Markus Staudacher
Vehicle-to-anything (V2X) communications refer to information exchange between a vehicle and various elements of the intelligent transportation system (ITS), including other vehicles, pedestrians, Internet gateways, and transport infrastructure (such as traffic lights and signs). The technology has a great potential of enabling a variety of novel applications for road safety, passenger infotainment, car manufacturer services, and vehicle traffic optimization. Today, V2X communications is based on one of two main technologies: dedicated short-range communications (DSRC) and cellular networks. However, in the near future, it is not expected that a single technology can support such a variety of expected V2X applications for a large number of vehicles.
The student should provide an overview of the current V2X technology. Furthermore, the student should analyze how V2X could benefit from upcoming cellular communication standards (5G).
- K. Abboud, H. A. Omar and W. Zhuang, "Interworking of DSRC and Cellular Network Technologies for V2X Communications: A Survey," in IEEE Transactions on Vehicular Technology, vol. 65, no. 12, pp. 9457-9470, Dec. 2016.
Trellis Coded Modulation
Supervisor: Fabian Steiner
Trellis Coded Modulation was one of the first approaches to combine higher-order modulation with forward error correction codes and builds upon the principle of designing both jointly. That is, for a constellation with 2^m constellation points, a rate (m-1)/m convolutional code is employed. In this seminar, the student is expected to understand the scheme and summarize its underlying principles.
- G. Ungerboeck: Trellis-coded modulation with redundant signal sets Part I: Introduction
- G. Ungerboeck: Channel coding with multilevel/phase signals
Learning-Based Communication System
Supervisor: Peihong Yuan
The idea of (deep) learning-based communications systems goes back to the original definition of the communication problem and seeks to optimize transmitter and receiver jointly without any artificially introduced block structure.
The students task is to give a brief introduction to learning-based communications systems.
Optical Signal Processing by Four-Wave Mixing in Few-Mode Fibers
Supervisor: Tasnad Kernetzky
The nonlinear optical process Four-Wave Mixing (FWM) can be utilized for the signal processing tasks wavelength switching and optical phase conjugation. As FWM is highliy sensitive to phase-matching, the processe's efficiency can be improved if different fiber modes are taken into consideration.
Between two modes of an optical Few-Mode fiber, the two FWM processes Bragg Scattering and Phase Conjugation can be efficiently maintained and the diffetential mode delay can compensate for Chromatic Dispersion.
The Student's task is to understand the concept of fiber modes, FWM and phase-matching and perform a small Matlab Simulation similar to that in .
- Nonlinear Optics
- Fiber modes and therefore also Maxwell's equations to some extent
- : ieeexplore.ieee.org/document/8412265/
- : www.osapublishing.org/oe/abstract.cfm
- : ieeexplore.ieee.org/document/6648413/
Effect of TX and RX I/Q skew on High Baud rate optical transponders
Supervisor: Ginni Khanna
In this seminar, the student is required to program a Python script capable of assessing effects in an optical transponder. The tasks involves:
- Mainly assessing the skew limitations of such transponders namely, the skew at the TX and RX.
- Analyse the effectiveness of the scheme and compare it with a traditional brute force scheme
- Analyse the penalty on system performance for any residual skew in the TX And RX.
- Analyse the penalty on system performance with residual phase noise at the RX.
Note: Please make sure you have a slight knowledge of Python and a bit of programming while applying for this seminar
Simple Clustering Algorithms for Soliton Transmission with Detection in the Nonlinear Fourier Domain
Supervisor: Benedikt Leible
Achievable information rates (AIRs) of state of the art optical transmission systems are limited for high input powers, due to fiber nonlinearity. One can attempt to increase the AIRs of optical communication systems in the high input power region, by mapping data to transmission pulses in the nonlinear Fourier domain. This is in general done by using the nonlinear Fourier transform (NFT). In a simplified approach, the used nonlinear spectrum can be limited to its discrete part, resulting in solitonic transmission pulses. When the information is recovered from these pulses in the nonlinear Fourier domain, accumulated noise can no longer be considered to be Gaussian distributed. Thus hard decision detection for the discrete spectral points of these pulses yields non-optimal results. In an attempt to increase the AIR, clustering algorithms can be used to adapt detection to the unknown noise characteristics.
The students tasks include:
- Getting a coarse idea of the concept of the nonlinear Fourier transform and especially soliton transmission
- Understanding the concepts of two clustering algorithms (Expectation Maximization (EM) and K-means)
- Implementing the K-means algorithm in Matlab for the case of N-eigenvalue soliton transmission
- Comparing both algorithms (EM Matlab code provided) for the given transmission system in terms of complexity and mitigation capabilities
- Optical Communications Systems lecture (highly recommended)
- Solid Matlab programming capabilities
(Not all of this references have to be read completely. This list is just intended to give you an broad overview):
- Gerhard Kramer - Capacity Limits of Optical Fiber Networks
- Mansoor I Yousefi - Information Transmission Using the Nonlinear Fourier Transform, Part I: Mathematical Tools
- Darko Zibar - Nonlinear impairment compensation using expectation maximization for dispersion managed and unmanaged PDM 16-QAM transmission
- T. Kanungo - An Efficient K-Means Clustering Algorithm: Analysis and Implementation
Noisy processing in QKD
Supervisor: Roberto Ferrara
Quantum Key Distribution (QKD) exploits the transmission of quantum states to communicate a classical shared key that is provably secure from the any eavesdropper. Some aspects of QKD protocols can be understood as equivalently correcting errors in noisy shared entanglement to produce maximally entangled states that can then be use to extract the perfect key. However, there exist protocols distill key at larger rates than their the best protocols for entanglement distillation.
The goal is to understand the quantum error correcting codes in 1) and explain how they are used to improve a key rate in 2).
Security proofs of QKD protocols.
Supervisor: Roberto Ferrara
Quantum Key Distribution (QKD) exploits the transmission of quantum states to communicate a classical shared key that is provably secure from the any eavesdropper. While QKD does exploit quantum communication and measurements, these are normally used only to obtain a raw classical key and a bound on the information of the eavesdropper, then classical information theory is used to extract the perfect key where the security of the key can be proven thanks to the quantum assumptions. 1) and 2) provide security proofs for broad but specific classes of QKD protocols.
The goal is to understand the steps and structure of the QKD protocols and explain their security proofs.
Supervisor: Roberto Ferrara
Quantum information cannot be amplified, only error corrected. At the same time there exists a duality between quantum states and channels, and entangled states can be used to communicate quantum information using only local quantum operations and classical communication (with no quantum communication happening beyond the sharing of the entangled states). The most extreme example are maximally entangled states, which can be used to implement identity quantum channels using what is known as quantum teleportation.
Since amplification is impossible, quantum repeater stations are necessary to overcome noise in long distance communications. These repeater stations then help communicate quantum information by performing the error correction, or by distilling maximally entangled states used in teleportation.
The goal is to understand and report on a quantum repeater protocol. The student can choose between the original proposal 1), or one of the newest proposal 2).
Belief Propagation Decoding of Polar Codes
Supervisor: Thomas Wiegart
Polar codes  are capacity achieving codes with a successive cancellation decoder. Their finite-length performance is, however, relatively bad. Several approaches to improve the performance have been presented, such as list decoding and CRC-aided list decoding, which come with the cost of increased complexity. An other approach is to interpret the polar code as factor graph and perform belief propagation (BP) decoding . BP is a well established iterative decoding procedure (known e.g. from LDPC codes) which usually allows efficient and parallel decoding.
The student's task is to understand polar codes and the idea of belief propagation (BP) decoding for polar codes. Furthermore, the student should summarize recend advances in BP decoding (such as ,).