Edmund Coersmeier

Azimuth based localization for mobile phones

Based on measured acceleration and compass data of smartphones a concept for GPS-free localization and traveling route estimation is presented. The built-in compass is used to measure the relative direction changes of the traveling user. The resulting azimuth trajectories are then successively matched to a vectorized street map. In order to relate the measured trajectory to the street map data, it has to be scaled correctly by the current user velocity. To estimate the current velocity, pattern matching is applied to the acceleration sensor data to detect the current means of transportation (MOT). By associating typical velocities with each MOT, a set of possible scaled azimuth trajectories can be found.

Comparison between different adaptive pre-equalization approaches for wireless LAN

The IEEE 802.11a wireless LAN standard defines a high system performance and therefore requires a certain signal accuracy for the OFDM transmitter output. Taking the potential analog low pass, medium IF and RF filter imperfections into account, it is necessary to pre-equalize the signal stream digitally before transmitting. The paper concentrates on the description and comparison between different mathematical approaches for pre-equalization in an OFDM transmitter front-end architecture. For the favorite algorithm, the corresponding simulation results are presented.

Adaptive pre-equalization in analog heterodyne architectures for wireless LAN

The IEEE802.11a wireless LAN standard defines a high system performance and therefore requires a certain signal accuracy for the OFDM transmitter output. Taking the potential analog low, medium IF and RF filter imperfections into account, it is necessary to pre-equalize the signal stream digitally before transmitting. The paper concentrates on the description of the OFDM transmitter front-end architecture and the adaptive decision-aided pre-equalization in the digital domain. The mathematics of the pre-equalization process are derived from the well known post-equalization respective channel-equalization case. Finally simulation results are presented.

Improving the Performance of a Recurrent Neural Network Convolutional Decoder

The decoding of convolutional error correction codes can be described as combinatorial optimization problem. Normally the decoding process is realized using the Viterbi Decoder, but also artificial neural networks can be used. In this paper optimizations for an existing decoding method based on an unsupervised recurrent neural network (RNN) are investigated. The optimization criteria are given by the decoding performance in terms of bit error rate (BER) and the computational decoding complexity in terms of required iterations of the optimization network. To reduce the number of iterations and to improve the decoding performance, several design parameters, like shape of the activation function and level of self-feedback of the neurons are investigated. Furthermore the initialization of the network, the use of parallel decoders and different simulated annealing techniques are discussed.

A matrix-vector based approach to FFT implementations

Today discrete Fourier transforms (DFTs) are applied in various radio standards based on OFDM (orthogonal frequency division multiplex). To achieve a high computational speed with low power consumption, specialized fast Fourier transform (FFT) engines are used in mobile devices. However, in face of the software defined radio (SDR) development, more general (parallel) processor architectures are often desirable, which are not necessarily tailored to FFT computations. Therefore, alternative approaches are required to reduce the complexity of the DFT. Starting from a matrix-vector based description of the FFT idea, we will present different factorizations of the DFT matrix, which allow a reduction of the complexity. The resulting complexity lies between the original DFT and the minimum FFT complexity. The computational complexities of these factorizations and their suitability for implementation on different processor architectures are investigated.

Scalable and efficient car communication topology

Car communication acts as a base for advanced traffic and car services. There is a strong potential to utilize existing hardware to introduce first car communication applications and services. Potentially first car communication technologies should be seen as initial base to learn about car communication while not all expected functions can be realized with this setup. The more time critical the required communication for a specific car communication application is, the more optimized communication technologies are needed. Considering only already existing communication technologies like UMTS and WLAN, there is a significant potential to implement first car communication solutions in near future. Thus this paper investigates which network topologies provide enough scalability and efficiency to realize first car communication functions based on available hardware.