Low Power Tracking of GNSS Signal by IMU Aided Intermittent Tracking Algorithm

L. Xu, M. Wang, R. Ying, P. Liu, W. Yu

Abstract:	Nowadays, most Global Navigation Satellites Systems (GNSS) receivers are based on ASIC and SoC, especially for hand-hold applications. On one hand, they adopt different power modes to reduce power consumption while maintaining a certain level of performance. One common power reducing technique is intermittent tracking, which means receiver or tracking unit operates for a certain period and shuts down for another period. The duty cycle is programmed by the user or adaptively changes according to the quality of received GNSS signal. Intermittent tracking reduces power consumption by powering down tracking unit at the cost of performance degradation on tracking weak signal for dynamic receivers. On the other hand, to track weak signal for high dynamic receivers, many integrated INS/GNSS systems are proposed, especially ultra-tight integration emerging recently. Interstate Electronics Corporation and Draper Laboratories firstly proposed the concept of ultra-tight integration. P. Li proposes an INS-aiding architecture based on ultra-tight integrated INS/GNSS system to reacquire GNSS signal from short-time undesired blockage. P. Li’s INS-aiding schemes estimate carrier Doppler by Kalman filter or other estimation techniques, and compensate Doppler when GNSS signal is in sight again. But such kind of INS aided tracking loop usually consumes more power. In this study, we propose a low-power, high sensitivity and high dynamic adaptive intermittent tracking scheme, which is aided by an Inertial Measurement Unit (IMU). The new tracking scheme consumes less power compared to conventional integrated INS/GNSS system based on the observation that the power consumption of an IMU, especially MEMS-based IMU, is much less than that of GNSS receiver (roughly 10%-20%). Furthermore, the proposed scheme possesses higher performance compared to conventional intermittent tracking schemes. The proposed tracking unit works in mainly two modes, i.e. wireless-IMU mode and IMU-only mode. The durations of these two modes are determined by wireless signal quality, dynamic of the receiver and the required power reduction. During wireless-IMU mode, both the tracking loop based on conventional PLL and IMU work, they are combined by a Kalman filter to provide better control of the tracking circuits. The mode is similar to ultra-tight integrated INS/GNSS system for high dynamic receivers; the difference is that our method focuses on reduction of power consumption; therefore a simplified state equation is applied to reduce computation complexity. During IMU-only mode, the RF circuits and tracking loop is shut down while IMU is still on. In order to avoid signal loss when the receiver comes back to wireless-IMU mode, we have developed an estimator to recover the code phase, carrier phase and Doppler frequency for the tracking loop One of the key issues for the proposed scheme is the parameter estimation of tracking loop when receiver switches from IMU-only mode to wireless-IMU mode. When the duty cycle and period is fixed, the estimation errors consist of systematic errors and stochastic errors. The systematic errors can be eliminated by Kalman filter during wireless-IMU mode. Based on the assumption that systematic errors are constant during IMU-only mode, we compensate systematic errors when tracking loop is turned on again. To get better estimation, we use a two order auto regressive (AR) model to model stochastic errors during IMU-only mode from IMU measurements. Simulation shows that the simple AR model gives a good balance between computation complexity and estimation accuracy compared to other models. Finally, to achieve optimal performance, selection of the duty cycle is discussed. The thermal noise is higher when the carrier to noise ratio (C/N0) of received signal is higher, which impose an upper bond on estimation errors to accommodate 3-sigmal errors that a carrier tracking loop can tolerate. That means in low C/N0 environment, we should increase the duty cycle of the proposed tracking scheme, which would consumes more power. Another performance degradation is positioning accuracy when C/N0 is low. To achieve best positioning performance under various C/N0, trade-off between power, dynamic and accuracy should be made. In this study, we simulate and test receiver’s performance under various combinations of C/N0 and duty cycle based on proposed tracking scheme. Finally, feasible operation regions under different C/N0 are provided given certain constraints on positioning performance. In this study, a novel intermittent tracking scheme with the aid from IMU is proposed. Both analysis and simulation show that the proposed tracking scheme achieve high positioning performance, including accuracy and dynamic tolerance, under a certain constraint of power consumption. The proposed scheme reduces the power consumption compared to INS/GNSS system and achieves high positioning performance compared to conventional intermittent tracking loop. To better estimate the loop parameters when receiver switch from IMU-only mode to wireless-IMU mode again, we use an AR process to model the change of code phase carrier phase and Doppler. Furthermore, we investigate positioning performance under different C/N0, and provide feasible operation regions (mainly selected by duty cycle) for the proposed tracking under certain performance constraint, including dynamic tolerance, accuracy and power consumption.
Published in:	Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) September 16 - 20, 2013 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	1065 - 1070
Cite this article:	Xu, L., Wang, M., Ying, R., Liu, P., Yu, W., "Low Power Tracking of GNSS Signal by IMU Aided Intermittent Tracking Algorithm," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 1065-1070.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In