TY - JOUR
T1 - Tightly Coupled Positioning and Multipath Radio Channel Tracking
AU - Mannesson, Anders
AU - Yaqoob, Muhammad Atif
AU - Bernhardsson, Bo
AU - Tufvesson, Fredrik
PY - 2016
Y1 - 2016
N2 - Radio based localization is an active research topic with a wide range of applications. In this paper, we focus on localization of a radio receiver equipped with an inertial measurement unit. The localization is performed while simultaneously constructing a map of the small scale fading pattern in the local radio environment. The map in our case is a ray-trace-based multipath channel model. This solution is enabled by sensor fusion of information from the channel estimation data and the inertial sensors, and it does not assume any knowledge of, e.g., transmitter locations. The sensor data is fused in a recursive state space model that combines the kinematic motion model with the ray-based radio channel model, and the state vector is estimated using a particle filter. The choice of the particle filter is justified by the multimodal characteristics of the posterior likelihood distributions that follows from the nonlinearities of the problem. The work is assuming a single receiver antenna but the approach can also be transferred to multiple antenna systems. We study the performance of the approach under realistic assumptions, based on the performance of today’s low-cost inertial sensors and radio systems, including accelerometer and gyroscope noise, and also radio receiver frequency error and noise. Simulations show a significant improvement in long-term positioning performance, evaluated against dead reckoning. The work is concluded with experiments which serve as a proof of concept for the proposed technique, using no extra equipment compared to what can be found in a modern cellular phone.
AB - Radio based localization is an active research topic with a wide range of applications. In this paper, we focus on localization of a radio receiver equipped with an inertial measurement unit. The localization is performed while simultaneously constructing a map of the small scale fading pattern in the local radio environment. The map in our case is a ray-trace-based multipath channel model. This solution is enabled by sensor fusion of information from the channel estimation data and the inertial sensors, and it does not assume any knowledge of, e.g., transmitter locations. The sensor data is fused in a recursive state space model that combines the kinematic motion model with the ray-based radio channel model, and the state vector is estimated using a particle filter. The choice of the particle filter is justified by the multimodal characteristics of the posterior likelihood distributions that follows from the nonlinearities of the problem. The work is assuming a single receiver antenna but the approach can also be transferred to multiple antenna systems. We study the performance of the approach under realistic assumptions, based on the performance of today’s low-cost inertial sensors and radio systems, including accelerometer and gyroscope noise, and also radio receiver frequency error and noise. Simulations show a significant improvement in long-term positioning performance, evaluated against dead reckoning. The work is concluded with experiments which serve as a proof of concept for the proposed technique, using no extra equipment compared to what can be found in a modern cellular phone.
UR - https://www.scopus.com/pages/publications/84997503388
U2 - 10.1109/TAES.2016.140653
DO - 10.1109/TAES.2016.140653
M3 - Article
SN - 0018-9251
SP - 1522
EP - 1535
JO - IEEE Transactions on Aerospace and Electronic Systems
JF - IEEE Transactions on Aerospace and Electronic Systems
ER -