Constructing Angular Power Maps in Massive MIMO Networks Using Measurements Without Location Labels

Online Channel state information (CSI) acquisition is a challenging problem in massive multiple-input multipleoutput (MIMO) networks. Although the current 5G massive networks collect a large amount of CSI data for radio resource management, such data is discarded immediately after use, because the collected CSI data is not associated with location labels. Radio maps provide a promising solution for radio resource management by reducing online CSI acquisition. However, conventional approaches for radio map construction require location-labeled CSI data, which is challenging in practice. This paper investigates unsupervised angular power map construction based on large-scale CSI data collected in a massive MIMO network without location labels. A hidden Markov model (HMM) model is built to connect the hidden trajectory of a mobile with the CSI evolution of a massive MIMO channel. As a result, the mobile location can be estimated, enabling the construction of an angular power map. Based on reference signal received power (RSRP) data collected in a real 5G network, an average localization error of 17.9 meters can be achieved for location labeling of CSI. By using the angular power map constructed from the estimated locations, the proposed method achieves promising CSI prediction performance, demonstrating at least a 20% improvement compared to the traditional approach.