Research On Mobile Wireless Sensor Networks Localization Technology Based On Monte Carlo Method

Wireless sensor networks have powerful capability of information collecting, information processing, and information transmission. As a rapidly developing technology, WSN is getting more and more attention and application. After deploying a large quantity of micro sensor nodes in the surveillance area, the sensor nodes will form a network, thus WSN could monitor the whole area effectively. WSN could be applied in military use, environment monitoring, medical treatment, smart home, space exploration, husbandry, and other fields. The development prospect of WSN is vast.In practical applications, sensor nodes monitor the events’ happening and collect correlated information to send it to the observer. The information which records the event’s characteristics will become valueless if no position information is recorded. Since the sensor nodes are usually deployed in the surveillance area in abundance randomly, it would be unrealistic to make the sensor nodes equipped with GPS for the restriction of cost and volume. Thus localization for the sensor nodes is essential. The existing research on localization mainly focuses on the static WSN whose sensor nodes never move once deployed. The achievements on mobile WSN whose sensor nodes keep moving from one place to another are quite limited. Mobile WSN could make use of sensor nodes’ mobile feature and has intrinsic advantages in area surveillance. Nowadays mobile WSN has been applied more and more widely. So the localization research for mobile WSN has already become a hotspot issue.This paper reviews wireless sensor networks firstly, and describes the localization technology for mobile WSN in detail, including the localization principle of sensor nodes, related research of localization for mobile WSN, and performance indicators of localization algorithms. Then the paper explains the application principle of sequential Monte Carlo method in mobile WSN’s localization. On that basis, this paper proposes a new sequential Monte Carlo localization algorithm for mobile WSN, namely the TSBMCL algorithm. The algorithm of TSBMCL grants nodes better localization results by cutting more precise Monte Carlo Box and adding more particle filtering requirements. Large scale simulations show that this algorithm can lock nodes’ location area more precisely and improve the sampling efficiency. Compared with classic Monte Carlo localization schemes for mobile WSN, TSBMCL improves the localizing precision greatly.