Font Size: a A A

Research On Indoor Location System Based On ZigBee Node Network

Posted on:2017-05-04Degree:MasterType:Thesis
Country:ChinaCandidate:B QiFull Text:PDF
GTID:2308330485966767Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the rapid growth of data services and multimedia services, people increase the demand for positioning and navigation, especially in complex indoor environment,such as airport lobby, exhibition hall, warehouse, supermarket, library, underground parking, etc. The position of one indoor mobile terminal or its holder is often necessary to determine. However, compared with outdoor positioning, indoor positioning is limited by time, accuracy and complex environment, and such that outdoor positioning methods cannot be effectively applied to indoor environment.Therefore, to develop indoor positioning system with high accuracy, good stability,simple realization and low cost using wireless sensor networks has become a research hotspot.This paper presents firstly with an overview of the research status of indoors positioning technology, and then compares common indoor wireless positioning technology. Afterwards, it focuses on ZigBee wireless location systems. On this basis,through the comparison of the present several location algorithms, this paper selects the received signal strength indication(RSSI) convenient localization algorithm.Because of the interference of environmental noise in the process of RSSI ranging, the positioning error is very large beyond a certain distance, so this paper presents a region partition and centroid weighted algorithm, which divides the positioning region into several subregions to realize close positioning. This algorithm also takes the sum of reciprocal of distance between nodes instead of the reciprocal of sum of distance as the weighted value, and has been corrected according to the weighted factors in order to make full use of the node information. The simulation results show that this algorithm can provide an additional relative precision gain of 25% than the original weighted centroid algorithm, and the robustness of the improved algorithm is also better.Finally, because the error caused by indoor complex environment is inevitable, theperformance of the traditional algorithms will be greatly affected, while the BP artificial intelligence network has good fault tolerance ability and nonlinear mapping ability, so it can achieve good positioning performance. The simulation results show that the average localization error of the network in the area of 6m×3m is 0.21 m. In addition, aiming at the problem that the BP neural network is easy to fall into local minimum, this paper uses particle swarm optimization(PSO) to optimize the initial weights and thresholds of BP neural network. The experiments show that the improved algorithm accuracy can reach 0.16 m, which is 23.8% higher than that of the ordinary BP neural network.
Keywords/Search Tags:indoor location, received signal strength indication, correcting weight, BP artificial intelligence network, PSO algorithm
PDF Full Text Request
Related items