Research On Indoor Localization Algorithm Based On Received Signal Strength

With the development of information technology,location information has become the focus of public attention.Sophisticated GPS technology is mainly used in outdoor environment,however,for complex indoor environments,GPS technology cannot work effectively.Therefore,indoor localization technology has received widespread attention.Because of the relatively low complexity,indoor localization algorithms based on Received Signal Strength(RSS)and Time of Arrival(TOA)algorithms based on Energy Detection(ED)has become a hot spot.Firstly,the development status of indoor localization technology is introduced in this dissertation.Several commonly used localization technologies and methods are analyzed.Then,several typical indoor localization algorithms based on RSS and TOA estimation algorithms based on ED are studied.Afterwards,an improved localization algorithm(MLA)which piecewise linearizes the exponential relationship curve between distance and RSS is proposed,and it is verified by simulation and experiment.Finally,an improved TOA estimation algorithm(MKS)based on ED is proposed,and various simulation analysis and discussion are made for MKS algorithm.In the simulation,the channel model CM3 and CM4 of IEEE 802.15.4a are applied.In MLA algorithm,the estimation process of all unknown parameters can be omitted by the linearization operation which piecewise linearizes the exponential relationship curve between distance and RSS with unequal intervals,so the offline training of the propagation model can be omitted in entire localization process.The lines after the linearization are very close to the original exponential curve.Therefore,real-time localization of the target node can be realized in MLA algorithm.To some extent,compared to the traditional algorithm based on exponential propagation model and the algorithm which linearizes the exponential curve using one line,the complexity of the positioning process and the localization error are reduced respectively.MKS algorithm mainly make the kurtosis analysis of energy blocks which are energy samples of the received ultra-wideband(UWB)signal and select the point with maximum kurtosis as the TOA estimation.Compared to other algorithms mentioned in this dissertation,the accuracy of MKS is improved.About the conclusion that the TOA corresponds to the maximum kurtosis,this dissertation gives specific theory analysis and formula derivation.The simulation and experiment results show that the performance of algorithms proposed in this dissertation are superior to those of other algorithms mentioned.