A Source Prediction System for Dynamic Networks Based on TDOA Measurements

A source prediction system for dynamic networks is proposed using the Time Difference Of Arrival (TDOA) measurements. In the proposed system, the starting position and velocity of mobile source are predicted by considering the velocity constant. Firstly, Unconstrained Linear Least Square (ULLS) algorithm is proposed to predict the position and velocity when assuming the signal Propagation Speed (PS) to be known. Then Rank-One Semi-Definite Programming (ROSDP) algorithm is also put forward to improve the performance. The ULLS and ROSDP algorithms are also extended to the situation of unknown PS. We also conduct the experiments in both a simulated environment and a real system to demonstrate the effectiveness of the proposed algorithms. The results show that the proposed ROSDP outperforms the ULLS in terms of the position and velocity prediction whether the PS is known or not.