Research On Node Self-localization Algorithms For Wireless Sensor Networks

Wireless Sensor Network(WSN),an ad-hoc network,which consists of thousands of cheap and low-cost sensors,has widen our ways of data acquisition and connects the physical world to the logical information world.It has bright future in fields of the military,precision agriculture,Smart Home,medical care et al.Self-localization is a fundamental issue of many wireless sensor network applications,and that makes the research on self-localization quite significant.In WSN, anchor-based self-localization algorithms have higher accuracy than anchor-free ones usually. But in some harsh environments or in large scaled network, anchor nodes cannot be placed manually, and in some other situations, the cost of installing GPS on sensor nodes is too high. In these cases, the research on anchor-free self-localization is that important. It can be convenient to exploit event sources measuring the time of flight of the signals acquired by each sensor node if the distance measurements cannot be gathered easily.The above method always needs to solve a nonlinear problem to achieve the sensor nodes locations. Given linear methods have lower computation complexity, this paper proposed an enhanced bilinear self-localization algorithm based on the bilinear method originally proposed by Crocco et al.The algorithm explores the noisy time of flight(TOF) measurements that quantify the distances between sensor nodes to be localized and sources also at unknown positions.The newly proposed technique first obtains rough estimates of the sensor node and source positions,and then it refines the estimates via a linear least squares(LLS) estimation.The LLS estimator takes the geometrical constraints introduced by the desired global coordinate system into account to improve performance.Simulations show that the new technique offers superior localization accuracy over the original Crocco’s algorithm under small measurement noise condition.Seeing from the simulation results from the third chapter,the enhanced bilinear method is not suitable for high noise condition.That means the enhanced bilinear method is susceptible to noise.Therefore,a new algorithm is proposed based on the enhanced bilinear method. The rough estimates of sensor and source locations are found using singular decomposition(SVD) and the weighted linear least squares(WLLS) estimation is applied to refine the estimates.The formulation of the WLLS estimator takes the effects of TOF noises on the SVD results of the Crocco’s method into account.Simulations show that the new algorithm has improved localization accuracy over the original Crocco’s method.