Multi-Sensor TDOA Associating Method Based on Location Approximation Filtering

The existing TDOA passive location technique typically uses the reference-sensor pulse correlation method for TDOA association after estimation. However, this method fails to effectively correlate TDOA measurements from all observation sensors under incomplete common-view conditions. To solve this problem, we propose a novel multi-sensor TDOA pairing method using location approximation filtering, which enables TDOA pairing for five or more sensors even with poor co-visibility. This method is a three-step process. Firstly, a three-sensor TDOA unassociated localization strategy is implemented to generate initial position estimates via the Chan method. Secondly, a linearized distance difference function is driven through ENU coordinate transformation and TDOA gradient analysis, and a target TDOA window is constructed via the Pauta criterion. Finally, false location points are filtered and valid TDOA pairs are identified by combining TDOA window constraints with multi-sensor measurements. Simulation results demonstrate that, compared to existing methods, the proposed approach achieves higher TDOA associating success rates under the incomplete common-view state, and the real-time performance of the algorithm is acceptable.