Node placement optimization using extended virtual force and cuckoo search algorithm in wireless sensor network

• Main Author: Puteri Azwa, Ahmad
• Format: Thesis
• Language: English; English
• Published: 2014
• Subjects: T58.6-58.62 Management information systems; QA75 Electronic computers. Computer science
• Online Access: https://etd.uum.edu.my/4412/1/s810427.pdf; https://etd.uum.edu.my/4412/7/s810427_abstract.pdf; https://etd.uum.edu.my/4412/

Node placement is one of the fundamental issues that affects the performance of coverage and connectivity in Wireless Sensor Network (WSN). In a large scale WSN, sensor nodes are deployed randomly where they are scattered too close or far apart from each other. This random deployment causes issues such as coverage hole, overlapping and connectivity failure that contributes to the performance of coverage and connectivity of WSN. Therefore, node placement model is develop to find the optimal node placement in order to maintain the coverage and guaranteed the connectivity in random deployment. The performance of Extended Virtual Force-Based Algorithm (EVFA) and Cuckoo Search (CS) algorithm are evaluated and EVFA shows the improvement of coverage area and exhibits a guaranteed connectivity compared to CS algorithm. Both algorithms have their own strength in improving the coverage performance. The EVFA approach can relocate the sensor nodes using a repulsive and attractive force after initial deployment and CS algorithm is more efficient in exploring the search of maximum coverage area in random deployment. This study proposed Extended Virtual Force and Cuckoo Search (EVFCS) algorithm with a combination of EVFA and CS algorithm to find an optimal node placement. A series of experimental studies on evaluation of proposed algorithm were conducted within simulated environment. In EVFCS, the algorithm searches the best value of threshold distance and relocated the new position of sensor nodes. The result suggested 18.212m is the best threshold distance that maximizes the coverage area. It also minimizes the problems of coverage hole and overlapping while guaranteeing a reasonable connectivity quality. It proved that the proposed EVFCS outperforms the EVFA approach and achieved a significant improvement in coverage area and guaranteed connectivity. The implementation of the EVFCS improved the problems of initial random deployment.