Carrier Phase Tracking Architecture for Positioning in LTE Networks Under Channel Fading Conditions

Author: Dr. Pai Wang Postdoctoral Researcher for the SeNSe Lab at University of Colorado Boulder
Presented at The Institute of Navigation's GNSS+ 2020 Virtual Meeting
Full Paper Available Here: https://www.ion.org/gnss/virtual-abst...

Satellite Navigation and Sensing Lab website: http://ccar.colorado.edu/sense
CCAR Research Center website: https://www.colorado.edu/ccar/
Aerospace Engineering Sciences Department website: https://www.colorado.edu/aerospace/
University of Colorado Boulder: https://www.colorado.edu/

Meeting webportal: https://www.ion.org/gnss/index.html
Institute of Navigation website: https://www.ion.org/

Abstract:
In urban canyons or indoors, the global navigation satellite system (GNSS) signals may be blocked and experience multipath effects, resulting in degraded navigation solutions. Significant efforts have been devoted to developing complementary positioning solutions in urban and indoor areas. Cellular signals, such as the current long-term evolution (LTE) signals, have emerged as promising signals-of-opportunity for positioning and navigation. The LTE network-based position methods include assisted-GNSS, enhanced cell ID, positioning reference signal-based observed time-difference-of-arrival (TDOA), and sounding reference signal-based uplink TDOA. Measuring the time-of-arrivals (TOAs) of the first arriving path using various LTE downlink synchronization and reference signals from multiple eNodeBs enables user equipment (UE)-based localization. Fingerprinting-based localization technique using readily available received signal strengths has also been studied in LTE networks. Moreover, the ranging measurements obtained from LTE systems can be integrated with other navigation systems to form hybrid positioning schemes. Recently, LTE carrier phase measurement has attracted attention owing to its potential higher ranging accuracy compared to the TOA measurement. Both open-loop (OL) and closed-loop (CL) architectures have been investigated to address the carrier phase estimation of the LTE or the generic orthogonal frequency division multiplexing (OFDM) signals. The OL estimator suffers from a high computational burden and a large estimation noise. The CL approach with a loop filter can enhance the estimation accuracy, but is vulnerable to high signal dynamics such as the channel fading in multipath environments. For the common channel models mentioned in the LTE technical specification, each propagation path can be modeled as a fading process which is characterized by frequent deep amplitude fading and fast phase fluctuation. Therefore, there is an urgent need to design and analyze carrier phase tracking loops for LTE signals under channel fading conditions. In this contribution, we present a CL architecture for tracking LTE carrier phase in fading channels using the cell-specific reference signal (CRS). This tracking loop is based on a generalized state space carrier tracking framework with the proportional integral filter (PIF) design in GNSS receivers. For the OFDM-modulated LTE downlink physical layer, we first derive the received signal model in both time and frequency domains under non-ideal synchronization and channel fading conditions. The state and measurement models for the LTE signal carrier tracking are then constructed by taking into consideration the specific subcarrier allocation pattern for the CRS transmission. Since the CRS is only transmitted in one or two OFDM symbols within each LTE slot, the time period used for generating the average carrier phase error measurement differs from the state update period. This results in a modified state estimation gain matrix in comparison with that for GNSS signals. In addition, the initialization procedures tailored to the Doppler frequency and initial carrier phase estimations of LTE signals are provided. Simulations have been conducted to validate the carrier phase tracking capability of the proposed architecture for LTE signals in fading channels. We assume that the UE moves towards the eNodeB along the line-of-sight direction with a constant speed and the LTE signal is propagated through a frequency flat Rician fading channel. The TOA of the received signal is tracked by using a delay lock loop. The carrier tracking performance is evaluated under various operating conditions, such as the Rician fading factor, signal-to-noise power ratio, and loop parameters. The proposed method is shown to be robust against channel fading effects and provides more accurate carrier phase measurements compared to the OL estimator. The carrier phase tracking algorithm developed in this contribution serves as a foundation for developing advanced signal processing algorithms for LTE positioning receivers in multipath fading channels. Adaptively adjusting the loop parameters has the potential to further increase the ability of the proposed architecture in handling harsh signal conditions.