5G Massive MIMO for Localization and Tracking

Abstract

• Multipath distortion has been a ill-posted problem for many years.
• Resolution/compacity ability which means how many clients you can serve simultaneously.

• Different from most of the localization methods which need to distinguish LoS and NLoS signal components, we treat LoS and NLoS signal as a whole and explore the statistical property of the received signal. We prove that the strength of the autocorrelation function (ACFS) of the received signal in massive MIMO systems exhibits like a stable Sinclike focusing beam around the receiver in spatial domain. By using ACFS, we bypass the problem that the LOS and NLOS signals are hard to be distinguished in severe NLOS environments.

• Variation of the imping angle/delay of each specific signal component (regardless of LOS and NLOS) caused by the channel changes (e.g., target movement may cause the changes of wireless channel between the BSs and target) does not impact much about our algorithm since the statistical property of the ACFS is stable.
• The system leverages the existing massive antennas on the BSs and only 1 antenna is needed on the receiver side, which makes it a low-cost design.
• The system can support massive concurrent targets/clients without affecting the channel capacity and preserve privacy since a target/client only listen to the BSs without transmitting any signal.
• Owing to the benifit of using ACFS, the system robustly achives < 0.2m localization error regardless of the environments and thus making the it scalable to various of applications such as navigations.
  

For more details. please refer to our paper (PDF）

Key words: Massive MIMO, Time Reversal, Multipath, Localization and Tracking.