Genetic algorithms applied in routing protocols for wireless sensor networks

Wireless sensors have the capacity to gather information like temperature, sound, motion or pollutants, being distribute in an environment that has to be monitored and controlled for many domains like industry, healthcare or military applications. Grouped in networks with one Base Station gathering the useful data, the sensors are characterized by limited battery life and low processing power, which results in a limited network lifetime. We propose a spanning tree topology for the WSN that dynamically changes according to the nodes' residual energy, in order to maximize the usage of the network. The main processing is done using genetic algorithms, which are most appropriate solution for optimization problems. Our contribution related to the protocol design consists in the way the spanning tree is built, based on the randomly selected edges, the fitness function is used in genetic algorithm and the recalculation of the topology according to the energy levels.