Memetic algorithm applied to topology control optimization of a wireless sensor network

In most applications of wireless sensor networks the specification of the corresponding topology can be useful for the optimization of some important features, such as: node energy consumption, connectivity and coverage area. This is known as the Sensor Allocation Problem (SAP). Our work proposes an approach based on memetic algorithm concepts to find high-quality solutions. In our approach, each node can be associated with one of four operation modes (classified according to its maximum range). The algorithm optimizes the position of each node and produces solution clusters. In order to evaluate the efficiency of the method, we analyze case studies with different coverage areas that are then compared against results previously found in the literature. Our experiments show that in order to achieve a smaller energy consumption and an increase in network coverage area, one needs to operate with a sizeable number of sensors, but with few nodes operating in larger transmission power modes (which require an increased energy expenditure).