Research On Localization Technology In Underwater Acoustic Sensor Networks

Underwater acoustic sensor network is a emerging application of wireless sensornetwork, as the marine environment ideal media, underwater acoustic sensor network has awide development prospects, such as: monitoring a wide range of target waters in real time,exploring oceanic resources, monitoring underwater environmental and assisted navigation,etc.The information collected by nodes in WSN (wireless sensor network) will make sensein case the locations of corresponding nodes can be obtained. Hence node localization isviewed as a classic question that numerous researchers have investigated for many years.Underwater acoustic channel has many characteristics, for example, serious path loss,multipath propagation, Doppler shift, big propagation delay and so on. As a result of the driftof ocean currents, underwater acoustic sensor network has lots of characteristics differentfrom WSN, for instance, high mobility, low reliability, low position accuracy, three dimensionspace, etc. The characteristics of underwater acoustic sensor network causes that locationtechnology in WSN can not be directly applied to underwater acoustic sensor networks.This paper researches the localization technology based on the mobile nodes’range-based. Our research mainly contains two parts. Firstly, this paper proposes alocalization algorithm based on particle filter in underwater acoustic sensor network (referredto as PF_UWSN algorithm). PF_UWSN localization algorithm is proposed on the basis of theDNR algorithm using the ideal node mobility model. PF_UWSN localization algorithm mapsout the forecast area according to the bounding box method, blends the particle filteralgorithm into the forecast period and the filter stage, and launches the expectation of lowerlimit of its estimation error by theoretical analysis. The experiments show that PF_UWSNlocalization algorithm overmatches the DNR localization algorithm on the positioningaccuracy. Secondly, on the basis of the effect large-scale underwater acoustic sensor networklocation algorithm (referred to as ELSL algorithm), this paper proposes a weightedlocalization technology in underwater acoustic sensor network (referred to as WUWSNalgorithm) with high weight thinking. Compared with ELSL algorithm, WUWSN algorithmis to first select the potential positions of all possible nodes, then weights all potentialpositions and last estimates position information not only by solving a unique solution,because the role played by the potential positions of high weight is larger than other potentialpositions. Simulation experiments set node movement model based on LaGrange oceancurrents kinematic model. Experimental results show that the WUWSN algorithm has excellent performance in terms of positioning coverage, the average location error andaverage communication overhead.