Software Defined Radio (SDR)

The number of standards used in wireless transfer systems is increasing (e.g. GSM, UMTS, RFID, WLAN, Bluetooth, WiMAX). A support of all standards by dedicated hardware in mobile terminals will be always unacceptable. To increase the Terminal's velocity, the base-band signal processing will be done in software, which allows a fast reconfiguration of the Terminal according to the required standards.

As part of the project, a transciever based on a DSP (Digital Signal Processor) for the DRM (Digital Radio Mondiale) broadcasting standard will be developed, which can be easily modified because of the modular structure, for example, to verify new algorithms. In parallel, appropriate algorithms that distinguish for their adaptive properties in terms of mathematical complexity and performance will be developed. This will allow that the complexity of signal processing algorithms (for example, the channel decoding) adapts to the transmission conditions. Consequently, the energy efficiency of mobile terminals can be increased, leading to smaller designs and respectively longer operation times.

As part of another project, an RFID-system should be realized in SDR. The communication with commercial labels is done with a USRP (Universal Software Radio Peripheral). This hardware-device provides the connection of the RX/TX-Frontend and a usual host-PC, which is calculating the actual signal processing. This flexible tool can be used for further research.

Signal Processing for Energy Transmission Systems

In modern energy transmission systems, monitoring has become a central task. Nowadays, this monitoring has to be much more accurate and complex. In addition, it has to be done in real-time. This is important for realizing automatic system reconfiguration to enable the system to adapt to a changing environment. This environment is highly influenced by the weather (when thinking of regenerative energy generation), the electricity stock exchange, and the system operation near the upper limit.

The monitoring is done by observing and tracking of specific system parameters as the fundamental frequency, the phase offset, and the spectrum at distinct key locations. It is followed by the interpretation of these parameters. Monitoring shall not only capture the primary system features like fundamental frequency and amplitude, but also recognize, classify and localize disturbances. Crucial algorithms for this are parameter estimation (e.g. signal subspace estimation), parameter tracking (e.g. database), and classification methods (e.g. control charts), as well as their interaction. In addition, solutions shall be cheap and easy to integrate into the existing environment.

Localization through Radio Frequency Identification (RFID)

Radio Frequency Identification (RFID) is the automatic identification of objects via radio waves. For this a so-called Reader emits the radio waves which are received by a so-called Label or Tag. The Tag then backscatters the waves while changing its properties in a way that the Reader can read out a unique identification when receiving the waves.

With the help of this mechanism localization is possible next to the identification of the object. For this procedure different information need to be extracted and combined adequately. One of these parameters is the Received Signal Strength Indicator (RSSI) which specifies the intensity of the received wave at the Reader. Through this and other parameters it is possible to distinguish the distance between Reader and Tag and to track the Tag's path.

Algorithm for parallel Baseband Signalprocessing

The rising demand for data transmission rate in mobile communications systems makes high demands on the computing power of mobile terminals. To achieve the necessary data throughput, it is aimed to implement the base band singal processing in parallel architectures. These include both traditional hardware architectures (ASIC, FPGA) and multi-processor systems. In addition, the parallel signal processing allows energy-efficient implementations, as high speeds can be avoided.

To be able to use the aforementioned advantage, a description of signal processing tasks through algorithms with parallel structures is necessary. As part of this and previous projects will be therefore developed parallel descriptions for common baseband operations in OFDM and CDMA systems (such as detection, channel decoding, FFT). Especially, the possibility of overlapping block signal processing for various applications will be investigated.

Sparse Matrix Problems for Network-on-Chip Architectures

Sparse Matrix-Vector-Multiplication (SMVM) appears in many scientific and engineering applications. Iterative methods (Jacobi method, conjugate gradient method, Lanczos method) are used to solve the underlying numerical problems, as e.g. solving sparse systems of linear equations or computing the eigenvalue decomposition of sparse matrices.

Since all these methods are based on SMVMs, various ways to speed up the SMVM on general purpose processors as well as parallel hardware structures were presented. However, all these approaches are strongly depending on a specific sparsity structure of the given matrix. Here, by using a Network-on-Chip (NoC) an approach for dealing with arbitrary sparsity structures is taken.

NoC architecture is the proposed concept for replacing the traditional bus-based on-chip interconnections by packet-based switch network architecture. Therefore, the packets (vector elements, matrix elements) can be freely distributed over the parallel hardware structure. Furthermore, there is a wide range of topologies of the NoC architectures and the used routing schemes. In this project the NoC architecture is used to deal with the highly irregular communication structure of parallel SMVM operations. The proposed SMVM-NoC realizes a chip–internal packet–based switch network as the main transmission network for the data transfers required for the SMVM computations. Meanwhile, the concept will also be realized in FPGA prototypes as well as an ASIC implementation using TSMC 45nm technology library.

Convex Optimization in Communication Problems

In recent years, convex optimization is widely used as a powerful optimization tool to solve many engineering problems. Convex optimization is the minimization of a convex objective (cost) function subject to convex constraints. The importance of convex techniques is that: any local optimum is a global optimum therefore, the convex optimization techniques are a good choice to solve such engineering problems. Many wireless communication problems are convex problems or they can be converted into convex problems using some relaxation techniques.

In our lab, the idea of using the structure of the wireless communication system matrix (Toeplitz, block Toeplitz, circular,...) is combined with a convex relaxation of the detection problem to reduce the computational complexity of convex optimization problems. For example, the generalized minimum mean squared error (GMMSE) detector has a performance almost the same as minimum mean squared error (MMSE) detector but it does not require the knowledge of noise power.

Our research aim is to generate a detector which has a performance better than MMSE with less computational complexity. We hope that convex optimization techniques will achieve that.