Data Analysis And Assessment System Of Multi-Radar

With the number of aims rapidly rise nowadays,the information that single-sensor can provided today could hardly meet the demands so that the multi-sensor has already replaced the position of it.Furthermore,data fusion technology has become a popular topic,which is also the main point in this paper.In the field of radar,analyzing the information of multi-sensor which including detecting,bond,relevance,estimation and combination to achieve estimation of acute condition and identity,is called data fusion.There are some main problems that we faced nowadays:hardly tract exactly because of the complicated environments and dismiss the module of evaluation to the performance of system.In order to solve the problems,we continue design and implement a system invalidate and improve fusion algorithm fusion system,designing and implementing a multi-sensor data fusion system which is based on a non-cooperative target and within a evaluation module.In this system,it includes correlation/filtering,state/identity estimation(location,attribute)and system evaluation.In order to improve the accuracy of the target state in noisy environment,the study compared the filtering effect in different environment.Secondly,in order to improve the correlation accuracy,the study compared the correlation effect in different situations.The combination of the two increases the accuracy of the target tracking(status/identity)in a complex environment.In addition,the evaluation module of evaluation index selection,and design the evaluation index of evaluation algorithm,through the data sources of different evaluation indexes classified into two categories,radar data and data fusion,and completed a review of two classes of data.In this paper,there is an new fusion algorithm of data fusion for improving tracking accuracy,increase the system evaluation module to evaluating the performance of the current system,which could also works as an interface on machine learning algorithm which could further reduce errors(this method is based on the cooperative target).