Trakpoint Solutions, Inc.

Jon Siann

Christopher Guy Williams

Aspects of the present invention provide systems and methods for distributed signal processing of indoor localization signals wherein statistical algorithms and machine learning are used in place of a fingerprint map. The disclosure relates to calculation of angle and distance based on measurements of an indoor localization signal, followed by energy-efficient distribution of signal processing. Local signal processing is performed using any of multiple eigen structure algorithms or a linear probabilistic inference, before cloud-based signal processing is performed using a nonlinear probabilistic inference and machine learning that's been trained with historical data transmitted by the base stations and time-of-day location patterns. Without having to generate and constantly update an energy-exorbitant fingerprint map, the disclosed system reduces localization error to merely 50 cm with 95% probability without compromising energy-efficiency to rival the accuracy of indoor localization systems that utilize fingerprinting.

Energy-efficient localization of wireless devices in contained environments

The present invention is directed to optimization and failure detection of a wireless base station network. Based on RF link attenuation measurement, e.g., a Received Signal Strength Indication (RSSI) measurement, a server determines an optimal transmission sequence. Each base station of the optimal transmission sequence, a predecessor and a successor, are designated. Each base station of the sequence generates a packet. Most distant base station (relative to the server) transmits its packet to its successor. Each base station of the sequence (in turn) receives the packet from its predecessor, combines the received packet with its own generated packet, transmits the combined packet to its successor, and so on until the combined packet is relayed to a super base station at the end of the sequence. The super base station transmits the packet to the server. Based on the packet size, server can ascertain which base station (if any) failed.

Optimization and failure detection of a wireless base station network

The present invention is directed to energy-efficient hibernation in indoor wireless localization systems. A tag passively associates with a detection point (DP) and establishes a reveille time. The tag will awaken at the reveille time and send or receive a beacon to or from its associated DP. If the tag is receiving a beacon, it will awaken, receive, phase-lock its clock based on when the beacon was expected and when it was actually received, and return to hibernation. The DP transmits a scattershot of beacons, one for every tag in the system. If the tag is sending a beacon, it will awaken, send its beacon, and return to hibernation. The DP will receive the beacon and adjust its own clock based on the delay between when the beacon was expected and when it was actually received. The tag will broadcast its location to the DP on a set interval.

Method for waking from energy-efficient hibernation

The present invention is directed to optimization and failure detection of a wireless base station network. Based on an RF link attenuation measurement, e.g., a Received Signal Strength Indication (RSSI) measurement, a server determines an optimal transmission sequence. For each base station of the optimal transmission sequence, a predecessor and a successor are designated. Each base station of the sequence generates a packet. The most distant base station (relative to the server) transmits its packet to its successor. Each base station of the sequence (in turn) receives the packet from its predecessor, combines the received packet with its own generated packet, transmits the combined packet to its successor, and so on until the combined packet is relayed to a super base station at the end of the sequence. The super base station transmits the packet to the server. Based on the packet size, the server can ascertain which base station (if any) failed.

The present invention is directed to optimization and failure detection of a wireless base station network. Based on Received Signal Strength Indication (RSSI) measurements, a cloud server determines an optimal transmission sequence. For each base station of the optimal transmission sequence, a predecessor and a successor are designated. Each base station of the sequence generates a packet. The most distant base station (relative to the cloud server) transmits its packet to its successor. Each base station of the sequence (in turn) receives the packet from its predecessor, combines the received packet with its own generated packet, transmits the combined packet to its successor, and so on until the combined packet is relayed to a super base station at the end of the sequence. The super base station transmits the packet to the cloud server. Based on the packet size, the cloud server can ascertain which base station (if any) failed.

Growing community of inventors

San Diego, CA, United States of America

Jon Siann