Multi-AP-Aided Joint Localization Enabled by Near-Far-Field Codebook Design

By utilizing the cooperation among multiple access points (APs), 6G systems are capable of supporting both communication services and sensing tasks simultaneously. To avoid a high overhead of channel information acquisition due to the large-scale antennas at each AP, codebook design and beam training provide an efficient method for wireless communications by searching the optimal codeword in the codebook. However, a new challenge arises how to achieve accurate localization via joint beam training of multiple APs, especially when users may randomly locate in the near field and far field of each AP, rendering traditional far/near-field localization schemes not applicable enough. In this paper, we propose a joint multi-AP localization scheme based on fast beam training. We design a generalized hierarchical codebook for both near- and far-field users, where the coverage area of the codewords is progressively narrowed down across layers to acquire the user angle relative to each AP. Leveraging the angular information obtained from multiple APs, we develop a joint multi-AP localization algorithm aimed at minimizing the localization error. Simulation results reveal that the designed codebook can serve both near-field and far-field users. The proposed joint localization scheme achieve a higher localization accuracy while reducing the localization overhead compared to the traditional schemes.