An Intelligent System Based on CCM, SA and FDTD for Sensor Node Placement Optimization in Wireless Visual Sensor Networks

A Wireless Visual Sensor Network (WVSN) has a mesh mechanism and performs multi-hop communication between multiple sensor nodes equipped with visual sensors. They can be used to monitor infrastructure facilities and rivers. However, the placement of sensor nodes has a significant impact on wireless communication connectivity and transmission loss. In addition, the imaging range of visual sensors is limited, which limits the area that can be imaged by the WVSN. Therefore, to cover all events with the imageable range of the WVSN, it is necessary to decide the optimal placement of sensor nodes in the WVSN and the imaging direction of the visual sensors. Since sensor nodes may be densely placed, it is needed to extend the communication range of sensor nodes and the range over which visual sensors are able to image the entire target area. In this paper, we propose an intelligent system based on CCM, SA and Finite Difference Time Domain Method (FDTD) for sensor node placement optimization in WVSN. We evaluate the proposed system by computer simulations. The simulation results show that the SGC is always maximized and all sensor nodes are connected for all methods. The NCE for CCM at the end of iterations covers 37 [unit] events, while for CCM-based SA and AEFSEM all 48 [unit] events are covered. For AEFSEM the sensor nodes are spread over a wider area and all events are covered in the imaging range of visual sensors. The ARSE for CCM-based SA at the end of iterations is 57 [ \(\%\) ], while for AEFSEM is 72 [ \(\%\) ], which shows an improvement of about 15 [ \(\%\) ] and the stable communication area is increased.