Research On Energy In Het Net-based Indoor Localization System Under The Group Mobility Mode

WLAN-based indoor localiztion system is a promising application system and draws a great of attention from academic and industrial fields,whose most outstanding shortcoming is high energy consumption.To overcome above problem,this paper proposes a novel localization frame based on heterogeneous network under the mode of goup mobility,which introduces a low-power WPAN in typical prue WLAN-based indoor localization system.The proposed indoor localization in this paper characterizes the group mobility by clustering and splits all portable smart devices into two kinds: cluster-header node and non-cluster-header node.The localization methods of CH node and non-CH node are relied on fingerprint matching algorithm and trilateration algotithm respectively.This paper studies,furtherly,the tradeoff scheme between energy and accuracy based on the proposed novel indoor localization system,which includes two aspects: 1)local dynamic re-clustering and multi-CH node dynamic switching scheme,2)optimal locating sampling scheme.The local dynamic re-clustering and multi-CH node dynamic switching scheme mainly studies how to carry out optimal re-clustering and multiple CH nodes dynamic switching during the indoor localizationprocess.Generally,those portable smart devices acting as CH node will consume more amount of energy than those acting as non-CH node.In order to maintain the fairness and proportionality of portable smart device,the multiple CH nodes should be changed rather than holding the line during localization.Therefore,this paper proposes the concept of incentive value based on remaining amount of energy and arrangements.Besides,the mobility of portable smart device results in the mismatching between initial clustering information and truly information as time goes on,thus re-clustering for some portable smart devices is necessary.The simulation results indicate that the proposed scheme can work well to keep the balance between energy and accuracy.The optimal locating sampling scheme mainly studies when non-CH nodes communicate with CH nodes.This paper determines the quantified expressions of energy consumption and location error based on the analysis and assumption.Then modeling this problem based on the constrained Markov Decision Process model and by solving this linear problem,we can get the optimal locating sample scheme.The simulation results indicate that the proposed scheme can realize the tradeoff between portable smart device energy consumption and location accuracy.