636 research outputs found
A cyclic projection algorithm via duality
No full textA cyclic projection algorithm via duality / Norbert Gaffke ; Rudolf Mathar. - In: Metrika. 36. 1989. S. 29-5
A cyclic projection algorithm via duality
No full textA cyclic projection algorithm via duality / Norbert Gaffke ; Rudolf Mathar. - In: Metrika. 36. 1989. S. 29-5
Linear minimax estimation and related Bayes-L-optimal design
No full textLinear minimax estimation and related Bayes-L-optimal design / N. Gaffke ; R. Mathar. - In: Symposium on Operations Research : Proceedings / ed. by: Benno Fuchssteiner ... - Frankfurt am Main : Hain, 1990. - S. 617-628. - (Methods of operations research ; 60
Optimizing power allocation in sensor networks with application in target classification
No full textIn the present work, a system of distributed wireless sensor networks with applications in target detection and classification is investigated. In particular, aspects of mathematical modeling, methodologies for resource management, and performance analyses are studied. The ultimate goal of present investigations is to find analytical solutions in closed-form for optimizing the power allocation and hence minimizing the energy consumption of the entire network while keeping the classification performance constant. Since closed-form solutions are very difficult to obtain and only achievable in particular cases, this work presents first novel results for dealing with power allocation in distributed radar systems. Passive and active radar systems are investigated under sum-power and individual power constraints per sensor node. In addition, the corresponding classification performance, exemplarily for the active radar system, is determined. This leads to the investigation of a certain class of integrals concerning the evaluation of the average classification probability for communication over fading channels. Since the evaluation of average classification probability and average symbol-error probability for communication over fading channels are closely related, the corresponding mathematical integrals are thoroughly scrutinized. Finally, new equivalent equations and novel mathematical expressions as well as accurate analytical bounds in closed-form are presented
Design and implementation of adaptive OFDM in a software defined radio framework
No full textCurrent broadband wireless standards are based on Orthogonal frequency division multiplexing (OFDM), a multicarrier modulation scheme that provides strong robustness against multipath fading. However, due to the increasing demand for new high data rate services, an important task in the design of future OFDM systems is to exploit frequency diversity offered by a broadband channel. The efficient way to preserve power and bandwidth efficiency is to perform an efficient resource allocation by adapting transmission parameters (bandwidth, coding/data rate, power) while satisfying certain system requirements. However, the investigation and assessment of information theoretic concepts in a radio frequency (RF) environment is limited in the literature. This is mostly due to the lack of commercial hardware that can support adaptable transmission parameters. Currently, these functionalities are only offered by Software Defined Radio (SDR) technology supported by general purpose hardware. This work aims at bringing information theoretic concepts for wireless resource management closer to the practical application by identifying and addressing the important challenges that arise from the implementation of an adaptive OFDM communication system. This comprises the presence of a real radio frequency (RF) channel, as well as the critical influence of the SNR estimation and hardware imperfections on the system performance. The contributions in this theses are of theoretical and practical nature. Because the signal-to-noise ratio (SNR) provides the measure of signal quality often used as an input parameter for many resource allocation algorithms, the theoretical contribution of this work focuses largely on proposing an efficient algorithm for the SNR estimation in wireless OFDM systems. Furthermore, this work contains the study of the rate adaptive resource allocation algorithms in a synchronization mismatch scenario with the proposed SNR estimator. This discussion provides information about performance degradation of a real implementable transceiver with respect to an ideal transmission system with perfectly estimated parameters. The main practical contribution of this work is the design and implementation of an SDR-based reconfigurable framework for testing the capacity-achieving adaptive OFDM transmission in a wireless RF environment. The implementation comprises a complete OFDM physical layer transceiver with a large set of reconfigurable parameters, which are normally static in real systems. This framework enables adaptivity of individual subcarriers such that practical evaluation of resource allocation algorithms can be performed. The transceiver performance is characterized and analyzed through the concept of SNR loss and corresponding rate-power function. The performance evaluation of the resource allocation algorithms in real RF conditions using the TIGR framework is presented. The results are compared with the simulation results showing the applicability of hardware imperfections models and robustness of the proposed SNR estimator to the design of efficient resource allocation algorithms
Performance analysis and control of wireless communication networks with multi-user receivers
No full text
- …
