An Improved TDOA Algorithm Based On CS-BPNN In UWB NLOS Environment

Nowadays,wireless positioning technology is more and more important to our daily life,and people's demand for its positioning accuracy is also higher and higher.Wireless location refers to the location technology that uses the combination of wireless communication and location algorithm to estimate the location of the label to be located relative to the base station coordinate.With the proposal of industry 4.0 standard,the continuous upgrading and development of Internet of things technology and the continuous integration of physical information system(CPS),People have higher and higher precision requirements for positioning system,such as item tracking,search and rescue,guidance and navigation,intelligent application and service,security monitoring and so on.Nowadays,wireless positioning technology has been widely used in military,commercial and public service fields.Due to the advantages of ultra wide band(UWB)positioning technology,such as fast transmission rate,high multi-path resolution,strong signal anti-interference ability and so on,it is rising in many indoor positioning technologies and widely used in the market.Among many location algorithms,time difference of arrival(TDOA)is widely used because of its low clock synchronization requirements,simple equipment and high location accuracy.In this paper,the basic concept of UWB positioning technology,the research status at home and abroad,and the composition of its positioning system are studied in depth.The related positioning algorithm is introduced and a positioning algorithm named TDOA in the positioning system is studied emphatically.In view of the low accuracy of traditional TDOA estimation algorithm and the low positioning accuracy caused by non line of sight(NLOS)error,this paper proposes an improved TDOA positioning algorithm based on cuckoo search(CS)and BP(back propagation)neural network.In the process of using the cuckoo search algorithm,we found that in the late iteration of the algorithm,the diversity of the population is poor and it is easy to fall into the local optimum.To solve this problem,we introduced the niche "punishment" mechanism to solve this problem.The algorithm uses the niche improved cuckoo search algorithm to estimate the non-linear equations of TDOA,and uses the estimated results as the input of the trained BP neural network to modify,and the output of BP neural network is the final location result.