The Design And Implementation Of WSN Routing Control Strategy Based On Genetic And Tabu Optimization

Wireless sensor nodes are randomly distributed and powered by a battery, the mostimportant features is that the energy is limited. Sending, receiving, and processing data arerequired energy consumption, but the maximum energy consumption come from transiting data.To save energy, a transmission path that consume little energy needs to be find, at the sametime,it can extend the network life cycle.Genetic algorithm is a global optimization algorithm which simulates the naturalevolutionary process, it chooses population using selection, crossover and mutation three geneticoperators and judges the merits of the individual through fitness function. The scope ofapplication is very broad, the algorithm can design different search programs for specificproblems. Many people also use genetic algorithms to find the data transmission path in orderto achieve energy-saving purpose. However, long-term observation of the experiment found thatthe genetic algorithm is easy precocious. It is because selection of the genetic algorithm makesindividual with greater fitness function value have greater chance to heredity to the nextgeneration, at the same time, makes individual with small fitness function value less likely toheredity to the next generation. In actual operation, the good individual quickly filled the entirepopulation which lead to the lack of diversity in the population, coupled with the fitness functionis set by human factors, the algorithm converges quickly, which lead to local minimum, and theresult is not the global optimal solution. While, the tabu search algorithm can solve theprematurity phenomenon. Th tabu search algorithm is heuristic algorithm solving optimizationproblems, it taboos over the optimal through the taboo table avoiding the solution which havebeen searched be repeatedly searched at the next step to open up a new search space which canexpand the search range and prevent repeatedly search around the local optima. Tabu searchalgorithm requires an initial solution, a good initial solution can speed up the search, and geneticalgorithms can provide this initial solution, there for, in this paper, the tabu search algorithm iscombined with the genetic algorithm, the solution found by genetic algorithm as the initialsolution in the tabu search algorithm. We propose making the template of the chromosome astaboo object, using taboo table taboo templates, searching the solution space outside the bestsolution and expanding the searching range in order to reach the global optimal solution.Simulation results show that this strategy can effectively solve the problem of prematureconvergence and balance the network energy.