Principle and Performance Analysis of Sequential TDOA Positioning in LEO mega-Constellations

Low Earth orbit (LEO) mega-constellations feature the advantages of a large number of sensors and wide coverage, enabling the acquisition of multi-dimensional time-difference information and presenting new opportunities for improving emitter localization accuracy. However, due to the typically narrow beams of emitters, multi-satellite simultaneous observation is challenging, which in turn limits further enhancement of localization performance. To address the limitations of existing approaches and the gaps in related research, this paper proposes a sequential TDOA based passive localization method. This approach fully leverages the constellation configuration advantages by jointly processing electromagnetic signals detected by different observation satellites at different positions and times for the same emitter, enabling high-precision localization. This paper presents a systematic quantitative analysis of the impacts of orbital configuration and beam parameters on localization accuracy, and conducts sequential TDOA localization performance simulations using a realistic constellation model established in STK. The results show that this method can improve localization accuracy to the order of hundreds of meters, which is significantly better than traditional methods, thus providing theoretical foundation for the optimization and design of reconnaissance systems supported by LEO mega-constellations.