Research On Technology Of Indoor Parking Lot Location Based On WiFi

In recent years,people are increasingly demanding fast and accurate location information for themselves and their vehicles in the large indoor parking lot environment.The global positioning system is difficult to achieve higher location accuracy in the complex indoor parking lot environment,and with the popularity of wireless local area networks(WLAN),users can use various mobile devices to connect wireless networks at any time.Therefore,the positioning technology of indoor parking lot has become a hot spot of research.In this paper,an improved fingerprint location algorithm based on Wireless-Fidelity(Wi Fi)is proposed and applied to indoor parking lot location.At present,most of the researches on the fingerprint location algorithm focus on the positioning accuracy,but ignore the time spent and the labor costs in the offline phase.In order to deploy fewer reference points(RPs)to reduce location costs,a partition-fitting fingerprint algorithm(P-FP)is proposed to reduce the location cost in the offline phase,and on this basis,put forward a fingerprint location algorithm based on sequential importance resampling(SIR)particle filter(PS-FP),which realizes real-time location and tracking of mobile targets.To reduce the workload of offline phase sampling,the P-FP algorithm firstly performs data filtering on the received signal strength indication(RSSI)in the offline phase,and eliminates the sample data which is quite different from the actual RSSI.Secondly,the whole target environment is partitioned and the virtual RP(VRP)is redeployed in each partition,and the environment coefficient of each partition is obtained by the method of partition fitting.Then,the error characteristic vector of each partition is obtained by using the polynomial logarithmic distance path loss model and the environmental coefficient,and the vector is taken as the noise of the location partition.Finally,according to the indoor propagation model,the RSSI of all VRPs is calculated to build offline fingerprint database,and the K-means clustering algorithm is applied to cluster the fingerprint database to further reduce the workload of online phase.To further improve the location accuracy of P-FP algorithm,and realize the tracking of the target,based on the P-FP algorithm,this paper proposes the PS-FP algorithm.Firstly,the location coordinates are obtained by P-FP algorithm,and the location information of the target at the next time is predicted by SIR particle filter.Secondly,the relative distance between the predicted coordinates of the SIR particle filter and the coordinates obtained by P-FP algorithm at the next time is calculated.Finally,a reasonable threshold is set.If the relative distance is greater than the threshold,the coordinates predicted by SIR particle filter will be taken as the final location coordinate.Otherwise,the coordinates obtained by P-FP algorithm will be the final location coordinates.To better analyze and verify the effectiveness and practicability of the proposed algorithm,this paper systematically experimented with two algorithms.Experimental results show that compared with the traditional WKNN algorithm,fewer RPs are deployed in the indoor parking lot,the proposed P-FP and PS-FP algorithms can effectively improve the location accuracy when deploying fewer RPs in indoor parking lot,and have a good application prospect.