University of Michigan
This project presents a new innovative radio frequency (RF) localization solution with unprecedented accuracy and power efficiency. The proposed RF localization solution for wireless tags is small, low energy, and rapidly deployable without heavy infrastructure investment. The proposed approach enables decimeter-level (tens of centimeter) accuracy in large indoor environments where non-line-of-sight (NLOS) scenarios dominate and GPS
does not reliably operate. To date, no existing solution addresses this set of challenging specifications which is critical to a collection of public safety applications including workforce, robots and equipment tracking for emergency search and rescue operations, and epidemiology analysis. -July 2019
Principle Investigator: Hun-Seok Kim
University of Michigan
The Decimeter Accurate, Long Range Non-Line-of-Sight RF Localization Solution for Public Safety Applications research project focuses on a new innovative radio frequency (RF) localization solution with unprecedented accuracy and power efficiency. This RF localization solution for energy-autonomous wireless tags is small (≤ 10cm3 including RF antennas), low cost (less than a dollar per tag), and rapidly deployable without heavy infrastructure investment. The approach targets a stringent average power budget of 10s of micro Watts (µW) while meeting decimeter (10cm) accuracy in large (>100m per dimension) indoor environments where non-line-of-sight (NLOS) scenarios dominate. To date, no existing solution addresses this set of challenging specifications which is critical to a collection of public safety applications including workforce, robots and equipment tracking for emergency search and rescue operations, and epidemiology analysis.
This solution introduces the use of a new ‘active RF reflector’ on the wireless tag which echoes back a frequency-shifted orthogonal frequency division multiplexing (OFDM) signal originally generated from an anchor. The proposed technique is based on time-of-flight (ToF) estimation in the frequency domain that effectively eliminates inter-carrier interference in multipath-rich indoor NLOS channels. The system uses a relatively narrow bandwidth of ≤ 80MHz which does not require an expensive very high sampling rate ADC. Unlike ultra-wideband (UWB) systems, the active reflection scheme is designed to operate at a relatively low carrier frequency (e.g., 2.4GHz) that can penetrate building walls and other blocking objects making it applicable to NLOS scenarios. Since the bandwidth at lower frequencies (2.4GHz) is severely limited, the team will lean on novel signal processing algorithms as well as machine learning techniques to significantly enhance the localization resolution given a constrained bandwidth of the proposed system.
By leveraging the PIs’ expertise in ultra-low power (ULP) IC and system design, new power-optimized ICs will be fabricated and integrated with an energy-autonomous wireless tag system within a centimeter-scale form-factor. In addition to the active reflector, the fabricated IC will feature a new ULP wakeup receiver and narrowband transmitter to significantly lower average power consumption and to enable long distance (up to 10km) beacon transmission in emergency.
This project addresses design cost as well as overall performance, leading to lower power and more affordable electronics in a wide array of applications. Adopting the proposed tag IC would incur minimal power and complexity overhead to conventional dual-band (5.8GHz and 2.4GHz) RF transceivers that are ubiquitously adopted in smartphones and wearable devices. This project also includes the development of the infrastructure anchor prototype using commercial off-the-shelf components and software-defined radio (SDR) platforms. The anchor prototype will be mobile and rapidly deployable (simply pulling them into outlets) since the proposed technique efficiently eliminates the need for accurate synchronization among infrastructure anchors and tags.
The final goal is a demonstration of the end-to-end localization system and field-trials using the tag ICs and anchors in realistic indoor / outdoor environments, including university campus buildings and hospital environments. The miniaturized tags will be disseminated to the broad sensor network community to dramatically accelerate the research on a wide array of localization-aware Internet-of-Things and public safety applications.