Research On The Optimization Of Clustering Algorithm Based On Support Vector Machine

As the rapid development of several modern technologies, such as Web Communication Technology and Micro Electronic Technology, people raised their requirements on various aspects of operating system, so as their standard of information evaluation within big area.Consequently, more researchers start to address their research on the Wireless Sensor Network.However, these sensor nods are usually placed at locations without surveillance carried with limited battery power. Once the carried power been used out, it’s going to be really difficult to refill these sensor nodes. Therefore, how to extend the life lengths of wireless sensor network sensor nodes by using the carried battery power of these sensors more effectively and efficiently has become one of the most popular and important topic in the research area of Wireless Sensor Network. In this article, a new approach which uses the SVM technique to improve the LEACH algorithm and GAF algorithm is proposed. A novel optimized cluster-based routing LEACH arithmetic based on SVM is given in this paper targeting at the deficiencies of LEACH algorithm including the uneven distribution of cluster-heads and the uncertainty of the number of cluster-heads in every loop. Moreover, an improved GAF algorithm based on SVM is also put forward in this paper aiming at driving the virtual cells problems.Specifically, the main process of described in this papers are given below:(1)An optimized cluster-based routing LEACH algorithm combined with SVM is proposed and been performed in the optimization of node deployment strategy for wireless sensor network. Two methods are put forward to solve the uneven distribution of cluster-heads and the uncertainty of the number of cluster-heads in every loop problems of LEACH algorithm.One method is trying to release the optimization based on the SVM techniques. In this method,the geographical locations of the wireless network sensor nodes are taken into consideration.The corresponding supported vectors are searched and used to realize dividing the network space which minimize the uneven distribution of cluster heads. Another methods aiming at improving the cluster heads selection scheme of LEACH algorithm through defining the cluster heads number in every loop, which can solve the uncertainty of the number of cluster-heads problem. Simulation results show that, the optimized algorithm extends the network lifetime and reduces the energy consumption.(2)An improved GAF algorithms based on the SVM techniques is discussed in this paper. Based on the traditional GAF algorithm, this paper suggested to further use the support vector regression to optimize the virtual cell division and tested the optimized algorithms with simulation experiments. The experimental results indicate that the optimized GAF algorithm based on SVM realized the optimization of node deployment strategy.