CSI-Based Multi-Antenna and Multi-Point Indoor Positioning Using Probability Fusion

Channel state information (CSI)-based fingerprinting via neural networks (NNs) is a promising approach to enable accurate indoor and outdoor positioning of user equipment (UE), even under challenging propagation conditions. In this paper, we propose pipeline for wireless LAN MIMO-OFDM systems which uses uplink CSI measurements obtained from one or more unsynchronized access points (APs). For each AP receiver, novel features are first extracted the that robust system impairments arising in real-world transceivers. These inputs NN extracts probability map indicating likelihood UE being at given grid point. The output then fused across multiple APs provide final position estimate. We experimental results with line-of-sight (LoS) non-LoS conditions an 80MHz bandwidth IEEE 802.11ac using two-antenna transmit two receivers four antennas. Our shown achieve centimeter-level median distance error, order magnitude improvement over conventional baseline.