Indoor Localization with Distributed Nodes Using Single-Frame CSI

Communication-enabled indoor localization is an essential aspect and a key enabling technology of integrated sensing and communication (ISAC). In future wireless networks such as 6G, the demand for higher localization accuracy and system efficiency under constrained hardware and signaling overhead becomes increasingly critical. To this end, this paper proposes a distributed node-based scheme, achieving accurate indoor localization using only single-frame uplink channel state information (CSI) measurements. The proposed scheme utilizes time-synchronized cooperative nodes to estimate the relative time of flight (rToF) from uplink CSI and performs hyperbolic localization for user devices. To ensure high localization accuracy, the scheme optimizes the placement of the distributed nodes by minimizing the Cramér-Rao bound. Additionally, an inverse discrete Fourier transform-based method incorporating a novel prior sidelobes suppression filter is proposed to enable accurate rToF estimation from a single-frame CSI measurement. The scheme also introduces the ranging signal-to-noise ratio metric to characterize the rToF estimate quality by applying a weighted iterative position estimator. Comprehensive experiments conducted in various environments show that the proposed scheme achieves a median positioning error of 0.72 m to 1.08 m with a single-frame CSI measurement, outperforming state-of-the-art systems that require multiple-frame CSI measurements.