Research On Wireless Sensor Networks Locating Algorithm Based On RSSI And Split-step Particle Swarm Optimization

In recent years, wireless sensor network technology has been developed rapidly as the core supporting technology of the Internet of things, especially positioning technology. In the process of sensor network monitoring, the location information of sensor node plays a vital role in positioning. There also has been extensive application prospect for the positioning technology in military, medical, environmental monitoring fields. It becomes the key and hot point that how to apply these massive, randomly arranged, miniature, cheap sensor nodes to scheme out a localization algorithm or system which is highly precise, low power consumption, low cost with a long lifecycle, and that how to apply them to real life to improve people’s living standard and make it more humanized and intelligent.Based on massive reading and research on domestic and foreign references, this thesis is mainly focused on the research about the localization algorithm of sensor nodes:Firstly the background of positioning technology—wireless sensor network is summarized briefly, the relevant theoretical knowledge and the research significance of positioning technology are presented in detail, several classic range-based algorithm and range-free algorithm are compared and analyzed intensively. Based on this, taking into account the demand of practical application, the ranging method based on RSSI is chosen for locating. In the process of ranging, propagation loss model corrected is chosen to avoid ranging error brought by model as far as possible, at the same time, considering the uncertain error caused by randomly deployed nodes, anchor node elimination mechanism is proposed. Keep anchor nodes in line with the triangle inequality, exclude those does not comply with the conditions, so as to improve positioning accuracy. Based on this, an improved split-step particle algorithm is proposed and used to optimize the initial positioning for further. In the process of updating algorithm, considering the weights with dependence on environment and the anchor node performance, adaptive weight approach is adopted. Through a large number of experiments, the most appropriate weight is selected and the idea of survival of the fittest is adopted, the anchor nodes are binary phased out, so the anchor nodes with excellent performance are always maintained in locating, the locating precision is effectively improved.Based on RSSI locating algorithm and combined with the use of particle swarm algorithm, the initial ranging result is optimized deeply and a good locating effect is obtained. Through detailed and rational analysis of the experiment data, the algorithm effectiveness is testified.