This study proposes a novel hybrid localization method that combines the time difference of arrival (TDOA) and angle of arrival (AOA) measurements to achieve improved accuracy with two base stations. Unlike other subspace-based methods that require calculating the eigenbasis, our proposed method constructs a Krylov subspace basis using the residual vectors of the Conjugate Gradient (CG) algorithm, which reduces computational complexity. In our approach, the measurements obtained from AOA estimation serve as initial estimates in our Adaptive Conjugate Gradient Algorithm (ACGA). ACGA incorporates an innovative adaptation mechanism that utilizes the residual error for updating and refining the estimated target position. This adaptation strategy allows our algorithm to iteratively improve localization accuracy, especially in the presence of NLOS errors. To assess the performance of our method, we conduct exhaustive simulations, considering various NLOS error scenarios and comparing them against conventional methods and extensively developed algorithms. The results demonstrate that our proposed method surpasses existing techniques, achieving a more densely spaced estimated position of the target in noisy environments with only two base stations. Furthermore, our method approaches the Cramer Rao Lower Bound (CRLB), indicating its high accuracy and efficiency in estimating the target position in typical NLOS environments. The effectiveness and potential of the proposed hybrid TDOA-AOA localization method are evident from the superior performance demonstrated in our simulations. These findings highlight the practical significance of our approach for real-world applications where accurate target localization is crucial, such as wireless communication, radar systems, and autonomous navigation.
Read full abstract