AG Kommunikationstheorie


Optimizing Power Allocation in Sensor Networks with Application in Target Classification


Recent technological developments in distributed sensor systems need intelligent solutions for a wide range of new applications. One of the most crucial requirements is low power consumption, which demands for energy-aware design and operation. In order to balance power consumption and overall system performance, precise mathematical modeling is of great importance. Finding optimal solutions becomes increasingly complex as the number of deployed sensor nodes becomes large. Present numerical methods are too slow to optimize the system performance for realistic network sizes. It is the aim of the present thesis to improve the performance of wireless sensor networks that are used for target detection and classification. New explicit analytical solutions and algorithms are developed. Major contributions concern optimal power allocation, target classification and the corresponding classification probability. After a short overview of the entire thesis, the average error probability (AEP) for randomly varying channels is highlighted. The explicit evaluation of the AEP leads to an extremely difficult integration problem. The general form of the corresponding integral is of high relevance for many problems in signal processing and communication theory, particularly for determining the average symbol error probability in Nakagami-distributed fading channels. Concise representations of the integral are developed and presented. Out of this, accurate bounds for the AEP and their relative errors are derived in closed form and graphically illustrated.

zurück zur Terminübersicht