Research on BLE AOA estimation method based on FOCUSS improved algorithms

Bluetooth digital keys, as a hot research direction in current automotive intelligence, are difficult to meet the positioning requirements solely relying on traditional RSSI technology. Although the Bluetooth 5.1 specification introduces direction finding functionality, in practical engineering applications, traditional AOA estimation algorithms require collecting a large amount of independent snapshot data. Due to the limited specifications of Bluetooth receiving antennas in narrow vehicle spaces, they cannot support the data requirements of traditional algorithms. When observation data and dictionary matrices contain noise, it further affects the algorithm's estimation accuracy and computational complexity. In response to this situation, this paper proposes an arrival angle estimation algorithm that divides the FOSUSS algorithm into multiple observation matrices using Bluetooth RSSI information based on sparse representation theory. Simulation results and actual measurements demonstrate the effectiveness of the proposed algorithm. Compared to other methods, this algorithm achieves higher AOA estimation accuracy in vehicular environments and holds significant reference value for direction estimation in other application areas.