Unlocking the Beamforming Potential of LoRa for Long-range Multi-target Respiration Sensing

Despite extensive research effort in contact-free sensing using RF signals the last few years, there still exist significant barriers preventing their wide adoptions. One key issue is inability to sense multiple targets due intrinsic nature of relying on reflection for sensing: reflections from get mixed at receiver and it extremely difficult separate these each individual. This problem becomes even more severe long-range LoRa because range much larger compared WiFi acoustic based sensing. In this work, we address challenging multi-target issue, moving one big step towards practical adoption. The idea effectively utilize antennas gateway enable spatial beamforming support While traditional methods adopted Radar systems rely accurate channel information or transmitter-receiver synchronization, requirements can not be satisfied systems: transmitter are synchronized no state obtained cheap nodes. Another interesting observation that while helps increase signal strength, phase/amplitude which critical corrupted during process, eventually compromising capability. paper, propose novel processing issues above reparation with LoRa. Extensive experiments show our system monitor respiration rates five human simultaneously an average accuracy 98.1%.