Chair of Communications

Theses

Offered Theses

Additional topics for Master Theses are available. Please contact Prof. Dr.-Ing. Stephan Pachnicke or the assistants directly.

  1. Stokes Vector Receiver
    Motivation: While coherent receivers are capable of detecting the whole optical field, classical direct detection (DD) systems can only determine the intensity of the incoming signal. This advantage of coherent detection comes at the expense of a more complex hardware. The Stokes vector receiver is a DD scheme that can track polarization changes and decode information in multiple dimensions. This yields the advantage, that the data rate can be increased compared to DD systems, which use only a single photodiode, without increasing the symbol rate. The Stokes receiver needs additional hardware compared to classical DD systems and can therefore be considered as an intermediate step between these and coherent systems.
    In the scope of a master thesis the Stokes vector receiver shall be investigated in simulations.
     Keywords: Stokes Vector Receiver, Direct Detection, Digital Signal Processing, Optical Communications
     Supervisor: Tom Wettlin, Room C-015, Phone: 880-6312,
    tomw@tf.uni-kiel.de

     

  2. Investigation of transmitter side imperfections on NFT-based soliton transmission
    Motivation: Solitons are pulses that propagate linearly through the optical fiber. Due to their special shape, non-linear Kerr effects (self-phase modulation) and the dispersion balance each other out, which is why the fiber can optimally be regarded as an AWGN channel. With the help of the nonlinear Fourier transform (NFT) a mathematical tool has been developed in recent years which makes it possible to use high-level modulation formats for soliton communication. These developments have aroused a new interest in solitons. A disadvantage of the NFT, however, is that it only describes the optical channel and does not react to other imperfections. On the transmission side in particular, it is important to generate solitons as accurately as possible in order to enable linear development in the channel.
    The aim of this work is to investigate transmit-side disturbances such as modulator nonlinearities and imbalances, low-pass behaviour or quantization. Possible concepts for the equalization of these disturbances shell be developed and tested.
     Keywords: Nonlinear Fourier Transform (NFT), Numerical Calculations, Modeling, complex modulation format
     Supervisor: Jonas Koch, Room C-031, Phone: 880-6305,
    jonas.koch@tf.uni-kiel.de

     

  3. Nonlinear Fourier Transform
    Motivation: The Nonlinear Fourier Transform (NFT) has the potential to remove the nonlinear limitations for higher capacity with increasing signal to noise ratio (SNR) in the optical fiber transmission system. The time signal can be split into two different nonlinear spectra similar to the Fourier spectrum. The signal can be composed of harmonic waves (continuous spectrum) and nonlinear waves (discrete spectrum), which are orthogonal to each other. These spectra can be modulated separately. The propagation of those spectra along the optical fiber can be characterized through a linear phase shift. It is therefore a linearization process of the nonlinear evolution inside an optical fiber. This makes the demodulation relatively easy.
    However, the numerical calculations for generating (inverse NFT) and demodulating (NFT) of the signals are still computationally complex. Furthermore, modulation formats have to be adapted to the nonlinear spectrum in order to compete with state-of-the-art commercially deployed techniques.
    A part of this research area shall be covered in the course of a master’s thesis. This can be the investigation of different applications as well as the development or evaluation of different modulation formats.
     Keywords: Nonlinear Fourier Transform (NFT), Numerical Calculations, Modeling, complex modulation format
     Supervisor: Jonas Koch, Room C-031, Phone: 880-6305,
    jonas.koch@tf.uni-kiel.de

     

  4. Machine Learning in Optical Communication
    Motivation: Next Generation Digital Signal Processing - The Kerr-Effect leads to nonlinear impairments inside the optical fiber, which impose a limit for the achievable transmission rate with increasing signal-to-noise-ratio (SNR). Nonlinear compensators using state-of-the-art digital signal processing (DSP) often struggle to improve the signal quality. Additionally, the high complexity makes it impractical for real-time use. Moreover, only deterministic nonlinearities are being compensated. There are, however, other nonlinear impairments in a core network, which are the result of the interactions between random noise (e.g. concatenated Erbium-doped amplifiers, EDFA) and the Kerr-Effect.
    Especially stochastic nonlinearities can be identified, characterized and equalized through machine learning algorithms. In addition, we can build a probabilistic model to describe those impairments, which we can use to optimize the demodulation algorithms. A part of this research area shall be covered in the course of a master’s thesis. This can be the investigation of different applications as well as the development or evaluation of different modulation formats.
     Keywords: Machine Learning, Digital Signal Processing, Nonlinear Compensation, Characterization of nonlinear statistical impairments
     Supervisor: Rebekka Weixer, Room C-015, Phone: 880-6312,
    rebekka.weixer@tf.uni-kiel.de

     

  5. Analysis of the Energy Efficiency of Hybrid Optical-Electrical Data Center Connects
     Motivation: In next generation data centers thousands of servers need to be interconnected with very high bit rates. Current architectures mainly rely on so called “fat-tree” approaches using a multi-layer electrical switching matrix. For further increasing data rates such architectures become more and more inefficient and consume a tremendous amount of electrical energy. A potential solution is to use additional optical connects between blades or racks purely in the optical layer without any electrical components in between. The optical switching of data connects will allow a significant reduction of the energy consumption because no optical-electrical conversion is required along the path. As an example an electrical 10 Gb/s Ethernet connection requires approx. 10 W of power whereas a pure photonic solution can be realized with only a fraction of the energy consumption.
     Keywords: Modelling, Optimization, Data Center
     Supervisor: Mihail Balanici, Raum C-014, Telefon: 880-6311,
    mba@tf.uni-kiel.de