Wireless Sensor Network Localization Algorithm Based On Hybrid Ranging

For many applications and network services of wireless sensor networks, combining the perception of information and its location is crucial. For example, applications of fire rescue and warning, mechanical failure, battlefield surveillance, etc. Do not know the location of perceiving data sensor nodes is of no significance. Ranging accuracy is the important factor of affecting the accuracy and reliability of range-based wireless sensor network localization algorithm. This paper will focus on innovative work based on hybrid ranging localization algorithm of RSSI and TOF.Firstly, according to the latest research works, the paper summarizes the status of wireless sensor networks based on range based localization algorithm. In addition, introduce and analyze the structure and characteristics of wireless sensor networks as well as practical applications problems of wireless sensor network localization, Raises the positioning problem of wireless sensor network. Focus on discussing the key technologies of range-based localization algorithm.Secondly, aiming at the problems that received signal strength localization technology is vulnerable to environmental factors interference, low positioning accuracy, a new hybrid ranging method is proposed. Respectively by means of piecewise combination and weighted fusion the novel TOF and RSSI, improve the range and location accuracy. Through theoretical derivation of Cramer-Rao lower bound of hybrid ranging algorithm, its theoretical optimal value provide theoretical basis for the analysis and design localization algorithm.Finally, put forward the multidimensional scaling localization algorithm based on hybrid ranging. By structuring dissimilarity matrix of distance, double centralization and singular value decomposition, compute the coordinates of unknown nodes. Then comparing the performances of hybrid and single range localization algorithm. Simulation results show that:the use of hybrid positioning algorithms ranging positioning error is significantly less than under the same kinds of conditions ranging single location algorithm positioning errors while mixing distance location algorithm has a more stable position error convergence.