Analysis Of Dilution Of Precision In Wireless Sensor Network Location

Owing to the low-cost,large-scale,and self-organizing nature,wireless sensor networks play a crucial role in various fields,including both military and civilian applications.Positioning technology serves as the core and foundation for wireless sensor networks applications,as monitoring information without accurate location data is essentially meaningless.Consequently,the pursuit of improved positioning accuracy with lower computational complexity has become a prominent research topic.In this thesis,the positioning problem in wireless sensor networks based on ranging is investigated,with a primary focus on collaborative dilution of precision,network dilution of precision,and asynchronous dilution of precision in three systems:collaborative localization,network localization,and asynchronous localization for dynamic targets.These three factors decouple certain characteristics of localization,enabling an analysis of how to improve localization accuracy from specific perspectives.For the aforementioned systems,corresponding closed-form localization algorithms are proposed.The main content of this article includes:1.By calculating the CRLB(Cramer-Rao Lower Bound)closed-form solution for the ratio of cooperative positioning to non-cooperative positioning,collaborative dilution of precision is proposed.The value of this factor is solely related to the additional distance measurement amount and its distribution.Simulations verify that performance improvement in collaborative positioning compensates for positioning estimation errors by incorporating more measurements.Additionally,a subspace-based multi-dimensional scaling collaborative positioning method is studied,with simulation results indicating higher accuracy than non-cooperative positioning algorithms.2.For the network distance loss caused by network topology and communication constraints,network dilution of precision is proposed.The value of this factor is exclusively related to the missing distance measurement amount and its distribution.Simulation results confirm that positioning accuracy decreases due to the absence of partial elements.Subsequently,a network collaborative positioning algorithm based on low-rank matrix recovery is proposed.This algorithm performs collaborative localization through convex optimization of the low-rank property of the kernel norm constrained ranging matrix,reconstructing missing ranging information.Simulation results demonstrate that the algorithm achieves high accuracy in distance reconstruction and positioning.3.In situations where a large number of missing measurements between nodes in wireless sensor networks lead to synchronous positioning failures under extreme conditions,asynchronous dilution of precision is proposed.The value of this factor is independent of measurement errors,being only related to factors such as target speed and the accumulation of asynchronous positioning.Simulations verify the relationship between this value and target speed and accumulation time.Subsequently,a multi-station limited measurement asynchronous positioning algorithm based on least squares is proposed.Simulation results show that the algorithm can accurately estimate the speed and initial position of dynamic targets with sufficient accumulation time,but computational complexity is high.Therefore,an improved asynchronous positioning method based on recursive least squares is proposed.Simulation results indicate that this method achieves similar positioning accuracy with lower complexity compared to the asynchronous positioning method based on least squares.