Research On Novel Methods For Time-based Source Localization Using Moving Observers

Obtain the un-cooperative emitter location information via moving observers,is very significant wherever in the military or civilian field.Modern passive localization system develops towards the direction of high accuracy,using networks whose single node is simplest.Among these types of measurement used to passive localization,time-based measurements including time of arrival(TOA)or time difference of arrival(TDOA),theoretically,just a single channel is required to measure the TOA of the signal and for every node and make the load of each node simplest,moreover,TOA measurement has no relation with the modulation of signal,is the best choice of the network-based passive localization system.Aiming how to use time domain observations and moving observers to locate the emitter,this paper studies the basic principle,localization methods and localization performance analysis using single or multiple observers to locate the un-cooperative emitter with time-based observations.The main contents include:The second chapter studies how to establish the TOA localization modal based on a single or multiple observers.First,we analysis the periodic features of the emitter signal is required on that condtion that using a single observer and when the multiple observers can not detect the emitter's signal at th same time.We establish the localization modal using the periodic features,and obtain the observability conditions and the analyize that necessary localization condition.Then the periodic features of the radar and communication emitters are also analyzed.Aiming the problem that using the multiple moving receivers and these receivers detect the emitter's signal asynchronously.We establish two kinds of asynchronous TOA localization modals.One is based on multiple time references localization modal and the centralized time reference localization modal and analyzes the observabilities of the two modals.And at last,we analyze the ambiguous problem of parameters in the the centralized time reference localization modal when the distances among multiple observers are large.The third chapter studies the problem that TOA-based periodic source localization with a single moving observer.First,we derive the conditional cramer-rao lower bound(CRLB)of the problem and we proposed batched method to estimate the location and period of the emitter simultaneously based on Gassian newton iteration.Second,we establish the dynamic filter model of estimating emitter location and the period,we derived the recursive conditional CRLB and proposed the filter approach to estimating emitter location and the period simultaneously.At last,simulation studies are performed to verify the effectiveness of the batched and filter appoarches.The fourth chapter studies the problem that using the multiple moving receivers include: TOA localization approach based on multiple time references localization modal we derived the C_CRLB and proposed the method based on Leverberg-Marquardt iteration approach to estimate the location of the target and the period jointly.We study another asynchronous TOA localization method based on the centralized time reference localization modal;we derived the C_CRLB and based on the ambiguity analysis,we proposed three the estimating the emitter location and the period simultaneously include: the method based on time delay compensated order(TDCO),the method based on grid-based multiple intialization LM iteration(GB-LM)and the method based on(Particle Swarm Optimization-based LM iteration(PSO-LM).At last,simulation studies are performed to verify the two appoarches.The fourth chapter studies the problem TDOA-based emitter localization,using multiple moving receivers under the condition that the emitter signal can be dectected by all moving receivers at the same time.Aiming at the high computational complexity of the traditional iteration methods and the high noise threshold of the traditional algebraic solutions,Based on analysis of the poor performance of the two stage weighted least squares(TSWLS),we proposed a novel algebraic and close-form localization frame with the localization error refinement for TDOA localization,the theoretical effectiveness of the new localization frame are analyzed and proved.We also derived theoretical estimation bias of the proposed approach.And extend this new localization framework to TDOA-based source localization with sensor location errors and TDOA/TDOA-rate based localization problems.The effectiveness is also proved under the two localization scenarios.All indicates that the proposed approach has excellent extended feature.