Realization of optimization of wireless sensor networks by electromagnetism-like algorithm

Wireless Sensor Networks (WSNs) are used to observe physical or environmental conditions such as heat, moisture, sound, object movements in different conditions by using sensors which behave independently from each other in coverage area. The optimization of dynamic deployments after the random deployment of sensors in WSN increases the coverage rate of the related network and as a result develops directly proportional to the performance of the network. Furthermore, the realization of efficient dynamic deployment which determines the positions in the related area of mobile sensors that are in WSN enables the sensors to converge faster and as a result locate in the optimal position in a short period of time. In this thesis, our purpose is to achieve dynamic deployment of the sensors in the area in an optimal way so as to optimize the dynamic deployment of sensors in WSN with Electromagnetism-like (EM) algoritm which is a population based meta-heuristic algoritm and works according to loaded particle theorem in electromagnetic area. Optimal Sensor Detection Algorithm (OSDA-EM), which considers the proposed EM algorithm as a basis, is improved and by applying it to deployment models, and optimal coverage area in WSN is obtained. Firstly, OSDA-EM, which has been improved, has been applied to binary detection model using only mobile sensors in the related area and then the change of the coverage rate has been searched by applying it to probabilistic detection model both with mobile and stationary sensors. The application of the improved OSDA-EM in MATLAB has been performed and the results, which have been obtained, have been compared with Particle Swarm Optimization (PSO) and Artificial Bee Colony (ABC) based algortims in literature, which is proposed in the dynamic deployment of WSN. Finally, the efficiency of OSDA-EM has been measured by performing applications in an authentic environment. Consequently, it has been determined that the improved OSDA-EM has provided optimal results in dynamic deployment of the sensor nodes and it has been discussed that it can be preferred in the optimization of the distributed WSN.