Joint Detection and Tracking of Unknown Beacons for Navigation with 5G Signals and Beyond

Abstract

A receiver architecture is proposed to jointly detect and track unknown beacons to extract navigation observables from fifth generation (5G) new radio (NR) signals of opportunity and beyond. Unlike conventional opportunistic receivers which require knowledge of the signal structure, particularly the reference signals (RSs), the proposed receiver requires knowledge of only the RS period and carrier frequency of the signal. The transmitted RSs for private networks are unknown for an opportunistic receiver. Moreover, to use the spectrum more efficiently, some of these RSs are only transmitted on demand in 5G NR, which limits the existing opportunistic navigation frameworks to signals which are on always-on; hence, limiting the exploitable RS bandwidth. To exploit the full available bandwidth and improve ranging accuracy, the proposed receiver is designed to estimate all the RSs contained in the transmitted signals corresponding to multiple unknown sources. Navigation observables (pseudorange and carrier phase) are subsequently derived from the estimated RSs. The proposed receiver operates in two stages: (i) detection of unknown signals and (ii) tracking. The detection of unknown signals is modeled as a sequential detection problem where the number of sources and their corresponding RSs and Doppler frequencies are unknown. The generalized likelihood ratio (GLR) test for sequentially detecting active gNBs is used to estimate the number of sources and their RSs. In order for the receiver to refine and maintain the Doppler and RS estimates provided by the acquisition stage, tracking loops are used. The output of the tracking loops, namely carrier phase and code phase, are then used to estimate the receiver’s position. Experimental results are presented demonstrating the capabilities of the proposed receiver with real 5G signals on ground and aerial platforms, with an experiment showing the navigation results with real 5G signals on an unmanned aerial vehicle (UAV) navigating using the proposed receiver over a 416 m trajectory with a position root mean-squared error (RMSE) of 4.35 m.