Preliminary Localization Results With An RFID Based Indoor Guiding System

Abstract

This paper reports preliminary work with an RFID with a growing importance, Radio Frequency (RF) signals [1], based local positioning system (LPS) designed for location and [2]. Belonging to this last category, many different possibilities guidance of people and autonomous vehicles in indoor environ-have been explored: WiFi, Bluetooth, Ultra-WideBand (UWB), ments. The system consists of an RF reader carried by the Radio Frequency Identification (RFID), etc. It should be noted mobile user, and a number of active RFID tags, disseminated at known positions in the displacement region, which regularly that all these technologies (with the obvious exception of GPS emit RF signals with an identification code. Upon reception and UWB, which permit direct measurement of the signal of a signal, the range of the user to the corresponding tag is propagation times) were designed for digital communication estimated indirectly from the received signal strength (RSSI), using rather than for localization purposes. a previously obtained statistical model. A computationally efficient RFID systems, which provide easy multiple reading of the Bayesian localization method (particle filter) is used to process the RFID systion codes and an indirec eatin of the measurements and produce an estimation of the user's position. tag's identification codes and an indirect estimation of theirThe RFID-LPS is tested empirically in a displacement region range to the receiver through the perceived signal strength comprised of three adjacent rooms, with a total area of 250 m2, (RSSI), are becoming very popular for many applications. Its in which there are placed 21 tags. Our first results show a typical advantages include not requiring wiring or line-of-sight to the mean positioning error of 3.25 m, which compares favorably with reader, a relatively long read range (with active tags), an easy other systems reported in the literature. deployment of tags, with minimum infrastructure modification, KeVwords -Local positioning systems, RFID localization, particle a capability to remain operative in harsh environments, and an filters overall low cost. Besides, RFID has quickly become a mature technology, with a high degree of penetration in the market.

Accurate Pedestrian Indoor Navigation by Tightly Coupling Foot-Mounted IMU and RFID Measurements

“…This methodology should limit the total error growth and keep it bounded by a value that depends on the maximum accuracy obtainable by a typical RFID-LPS system (about 2 m according to most papers in literature [17], [33]). The same trajectories as in Fig.…”

“…Unfortunately, MEMS acelerometers and gyroscopes are subject to significant random noise and bias [16] that have to be estimated on-line in to partially attenuate their drift effects. A more effective way to eliminate the accumulated error, is to fuse PDR with some indoor absolute positioning references, such as a received signal strength (RSS)-based local positioning system or RSS-LPS [17] or time-of-arrival (TOA) positioning systems (TOA-LPS) [18]. This PDR+LPS integrated concept, as already implemented in many outdoor applications fusing INS and GPS [19], has the potential to provide an accurate driftfree positioning solution.…”

“…Two main approaches to integrate RSS measurements with Inertial data can be found in the literature: 1) RSS-centric; and 2) Inertial-centric. The RSS-centric approach is focused in position estimation from the RSS data (by fingerprinting, Bayesian estimation or fitting data to a path-loss model); some corrections from the inertial sensor (in the form of incremental stride distances traveled) are fed into the estimation Alter to improve the movement model and to smooth the estimated trajectory (otherwise quite noisy paths are normally obtained using only RSS values [17], [27], [28]). For example, in [21] a WiFi RSS-based fingerprinting method is presented to estimate the position using a particle Alter implementation; the stride length (SL) estimation is fed (only distance, no orientation) in the particle Alter as a movement model to sample the new particles positions in a circumference having a normally distributed radius with mean equal to the measured SL.…”

Pedestrian indoor navigation by aiding a foot-mounted IMU with RFID Signal Strength measurements

“…In this section we present the results of the full integrated processing method as presented in section IV. This methodology should limit the total error growth and keep it constant to a value that depends on the maximum accuracy obtainable by an RFID-LPS system (about 2 meters according to most papers in literature [12], [11]), and consequently, the error should not grow with the path length or number of iterations. The same trajectories than in Fig.…”

A computerized system to determine the provenance of finds in archaeological sites using acoustic signals

“…UWB, Pseudolites) (Fontana and Gunderson, 2002;Werb and Lanzl, 1998) or by using the Received Signal Strength (RSSI) (e.g. WiFi, Bluetooth, RFID) (Ni et al, 2004;Bahl and Padmanabhan, 2000;Koutsou et al, 2007). These proposed LPS solutions are mainly devoted to applications such as robotic guidance and person position monitoring.…”