Energy efficient camera node activation control in multi-tier wireless visual sensor networks

Multi-tier topology model is an appropriate choice to accomplish the application-specific tasks in many wireless visual sensor networks. In a general multi-tier architecture, the first tier is composed of scalar nodes. This tier activates the camera nodes of the second tier within its communication range only when it is necessary. However, in many cases the activated camera nodes do not detect any events because of the directionality of their sensing range. In this paper, we present field of view clustering as a new clustering method to reduce the unneeded visual activations of camera nodes and overcome the waste of energy. We also improve this clustering method by presenting a solution named Random Neglect. As the simulation results indicate, our proposed solutions are successful in saving the energy of camera nodes and prolonging network lifetime while providing high detection performance.