Performance analysis of node positioning strategies in cognitive wireless sensor networks based on IEEE 802.22 protocol

This thesis investigates the effects of node positioning strategies on network performance within the framework of a Cognitive Wireless Sensor Network (CWSN) based on the IEEE 802.22 protocol. Within the scope of the study, an architecture running on TV White Spaces (TVWS) and using Wireless Regional Area Network (WRAN) infrastructure has been configured. Within this architecture, the communication between the Customer Premises Equipment (CPE) and the Base Station (BS) is realized using Time Division Multiple Access (TDMA). Each CPE node continuously measures key network performance metrics such as delay, jitter, and throughput, and transmits this data to a centralized base station. In this context, a fixed location-based deployment strategy has been developed, aiming to provide maximum coverage with a minimum number of nodes, thus reducing the total installation and operating costs of the system and increasing its operational efficiency. The designed architecture and the developed algorithms were tested in the OPNET (now Riverbed Modeler) simulation environment, and the performance values for each CPE were analyzed in the form of time series according to delay, jitter, and throughput parameters. These data were evaluated in a fuzzy logic-based decision support system to ensure the sustainability of the network under varying conditions. In this system, decision processes were carried out for the selection of the relay node, which routes the incoming data to the next destination, usually performing a routing or coverage extension function. The performance measurements obtained from each CPE were processed according to fuzzy logic rules to determine the optimal relay node. Simulation results show that the proposed node placement method and the fuzzy logic-based relay node selection algorithm provide significant improvements to network performance. Achieving a wider coverage area with fewer nodes reduces the installation and operation costs of the system, which directly contributes to strengthening economic sustainability. The findings reveal that intelligent decision support systems offer powerful and feasible solutions in cognitive network architectures not only in terms of high efficiency but also cost-effectiveness.