Multi-objective Optimization Of The Coverage Control Algorithm For Heterogeneous Wireless Sensor Networks

Wireless sensor network (WSN) is a information acquisition and processingtechnology, and is a hot research topic in the field of international information.Compared with the traditional technologies, WSN can be used in military,environmental science, the internet of things and so on. How to choose the coveragescheme for different applications is the primary issue in WSN. However, sensor nodeshave limited power due to their small size. So, it is important for WSN care about theproblem of energy-efficiency. The coverage control schemes not only direct reflect thequality of service of apperception, which provided by the WSN, but also can reduce thenetwork cost and energy consumption to prolong the life of net, according to design thereasonable coverage control strategy to optimize the resources of the network.Most of existing coverage control schemes for WSN, which only consider the sametype of nodes, cannot be well suited for the heterogeneous WSN since its nodes’heterogeneity is general. And the Boolean perceptual model used by most algorithms,which does not take into account the uncertainties perceived by the test objectives. Inaddition, the heterogeneous WSN coverage control algorithms don’t consider thedifferent monitoring areas have the different requirements of coverage. In other words,the monitoring areas have "hot zone", which some locals have frequent data collectiontasks.A multi-objective optimization of heterogeneous wireless sensor network nodedeployment strategy, using the probability perception model, is proposed based on thebinary particle swarm optimization algorithm, to the objectives of guaranteeingcoverage of sensors, satisfaction of detection thresholds, and least energy consumption,The algorithm is based on Pareto dominance relationship to construct non-inferiorsolution set. The selected non-inferior solution set is stored in the external storage space.The algorithm adapt dynamic ε-dominated to update the external collection, making thealgorithm has a good solution of the distribution. And it uses disposable relationship byadding random weight allocation of the appropriate value to select individual extreme,and selects the global extreme based on crowding distance, for increasing the diversityof solutions. The algorithm doesn’t need a priori knowledge about the objective function,but has an extensive search in the target space. The simulation results show that thealgorithm can carry out extensive search in the target space and get the effective heterogeneous wireless sensor network node deployment scheme. Compared with themulti-objective optimization algorithm based on NSGA-II, the proposed method hasgood convergence, which can effectively improve network coverage and reducenetwork energy consumption.