Research On Indoor WiFi Location Optimization Based On MUSIC Algorithm

In recent years,with the popularity of wireless networks,location-based services have occupied an important position in people's life.At present,the Internet of things technology is becoming increasingly mature.Under the Internet of things,the resource deployment optimization based on location information also brings a strong driving force for the research of indoor positioning technology.The scene of the Internet of everything is always supported by location information.In the scheme of indoor location information acquisition,the channel state information defined in the IEEE 802.11 WiFi protocol becomes a parameter that can better reflect the channel state than the received signal strength information,thus becoming the current hot research direction.However,the current positioning system based on channel state information is still unable to meet the precision requirements of the service scenarios of the Internet of things and smart home due to the accumulation of multi-party errors.In order to achieve high-precision indoor positioning system,there are still many challenges.For example,the classical MUSIC algorithm,as an important algorithm for positioning,has a large error in the estimation of the angle of arrival in the case of low signal-to-noise ratio and cannot be estimated due to the influence of coherent signals in the complex environment.In view of this,this article based on iot laboratory and a domestic company cooperation project "EMUI Router WiFi coverage optimization",based on the typical indoor complex scenes,for indoor location at different stages of the process,respectively,to study the error of measurement problem,arrival angle estimation problem under complex scene,and based on the angle of the location problem,an improved doas estimation algorithm is proposed and a set of positioning system error optimization method,specific as follows:First.Based on the classical MUSIC algorithm of arrival angle estimation,this paper proposes a Smooth ES-MUSIC algorithm,which effectively expands the virtual array of measurement under the communication architecture based on commercial network card.Different from the classical MUSIC algorithm,which only uses noise subspace,Smooth es-music algorithm makes full use of signal subspace information,which not only solves the problem of low signal-to-noise ratio(SNR)arrival estimation accuracy under non-visual distance or long distance,but also displays superior performance in coherent signal discrimination.The experiment shows that the error of the real scene is 30% lower than that of classical MUSIC.Second.A complete positioning system optimization scheme is proposed.In the data measurement stage of the positioning system,the minimum error multi-packet linear fitting method was used to eliminate the error,and DBSCAN clustering algorithm was used to eliminate the jitter error.In the calculation stage,Smooth ES-MUSIC algorithm was adopted to estimate the angle of arrival,and AP-logistic algorithm was proposed to optimize the principal path discrimination accuracy,and the signal attenuation model was combined with the position estimation model to determine the positioning target.The experiment shows that the accuracy of main path discrimination is 3.3% higher than the control system,the average positioning accuracy is 36% higher,and the robustness is better.