Abstract
The integration of satellite and terrestrial 5G networks aims to provide ubiquitous coverage, improve service reliability, and enable the network scalability. However, the inherent characteristics of satellite channels bring challenges on the air interface design of integrated terrestrial-satellite networks. For example, for low earth orbits (LEO) mobile satellite communication (SatCom) system, it is unclear so far whether the 5G new radio (5G-NR) synchronization signals could meet the requirement of timing and frequency offset estimation in the presence of large Doppler shifts. In this paper, we investigate time and frequency synchronization for the downlink transmission of 5G-NR signals over LEO satellite channels. Starting from the maximum log-likelihood criterion for timing offset estimation given the observation of the received primary synchronization signals (PSS), we derive an upper bound of the objective function for simplicity. With a priori information that the maximum Doppler shift of LEO satellite-ground link is within a specific range, we construct the local synchronization sequence via using the modulated discrete prolate spheroidal sequences (DPSS) vectors. Then the timing offset estimation can be recast into a one-dimensional peak search problem. Moreover, the cyclic prefix (CP) structure of orthogonal frequency division multiplexing (OFDM) can be utilized to improve the estimation performance further. Once the timing and frequency offset are captured in the above initial synchronization phase, the tracking synchronization can be much simplified as the variation of both the timing and frequency offset is very small between two adjacent synchronization blocks. Simulation results show that by using the proposed algorithms, the 5G-NR signals can achieve near optimal downlink time and frequency synchronization in typical LEO SatCom systems.
Highlights
An integrated satellite-terrestrial communications network can take the advantages of both the satellite and the land mobile communications and can provide genuine ubiquitous coverage [1]–[4]
Based on the Doppler shift characteristics and synchronization signal model given our research focuses on time-frequency synchronization problems under large frequency offset in the following
Cross-correlation bound algorithm: The method using (21) to estimate the timing offset with no frequency offset and using (38) to estimate the frequency offset with no timing offset
Summary
An integrated satellite-terrestrial communications network can take the advantages of both the satellite and the land mobile communications and can provide genuine ubiquitous coverage [1]–[4]. In current terrestrial cellular communication systems, e.g., long term evolution (LTE) and 5G NR, the cross-correlation based estimation algorithms are used in the downlink timing and CFO estimation, with the help of defined primary synchronization signals (PSS) [26]–[29] These existing algorithms are designed for the ground wireless channels with relatively low mobility, and their performance may degrade in the presence of large CFO in LEO satellite-to-ground links. Notice that the maximum values of the Doppler shift, the variation of the Doppler shift, and the variation of the propagation delay within the visibility window are determined by the parameters of the satellite orbit and UE locations In this work, such characteristics will be utilized to develop the timing and frequency synchronization algorithm. Based on the Doppler shift characteristics and synchronization signal model given our research focuses on time-frequency synchronization problems under large frequency offset in the following. Based on the proposed PSS algorithm, we consider utilizing the CP as auxiliary information to improve the estimation performance and avoid the involved one-dimensional exhaustive search in the frequency offset estimation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
