Virtual Force Disposal Algorithm Of Visual Wireless Sensor Network

The network with the sensor as the core node has been greatlyvalued in China. With the development of the Internet of Things, itis important to study the deployment algorithm for wireless sensornetworks. Wireless sensor network deployment problem can beformulated as an optimization problem. Such optimization problemscontaining more objective function, and its objective function islikely to nonsmooth and discontinuous. Therefore, the use oftraditional optimization method to deal with the optimizationproblem makes the efficiency and the results cannot be guaranteed.Especially when the sensor model is directed perception model,such as video sensor model, variable increased, the problem willbecome more complex. Optimizing the deployment of Internet ofThings with virtual force method, can avoid solving the optimizationequation, according to the location of the sensor nodes, calculatethe distance and direction of the sensor needs to be moved. Virtualforce algorithm optimizes the network deployment by the number ofsuch mobile.This paper improved virtual force algorithm, and applied it invideo sensor network optimization. In order to further improve theoptimization efficiency, this article adopts the method of changingthe force point location to reduce the effect of the direction of videosensor on optimization results. The main work of this paperincludes:1. View to the complexity of the model, this paper adopts themethod of changing the location of force point. This method couldreduce the effect of the direction of video sensor to a lower level inthe optimization of the sensor network. This article also putsforward some new methods for calculating the location of stresspoint, including the simulation of the force point, the inscribed circlecenter, and it is applied to optimize the sensor network deploymentvideo.2. To analog the uncertainties of video sensor in practical, thispaper presents a fuzzy video sensor model. The model is differentfrom the traditional0,1model. In the boundary region of the sensingrange, the sensor coverage intensity gradually weakened. In thispaper, using the integral method to calculate the weighted sensor area, and the location of the force point is calculated by the method,so as to optimize the sensor network deployment composed of suchsensors.3. According to the proposed method, this paper conducted anumber of experiments, and statistically analyzed the results. Thispaper optimized sensor networks containing different number ofsensor nodes by using the designed virtual force algorithm, andmake interpolation analysis for the results, including cubic HERMITEinterpolation, nearest point interpolation, etc. Finally, the paperpredict sensor networks optimization results by polynomial fitting ofthe experimental data.