[Paper Review] Joint Pilot Optimization, Target Detection and Channel Estimation for Integrated Sensing and Communication Systems

Zhe Huang, Kexuan Wang+4 more|arXiv (Cornell University)|Feb 5, 2022

Radar Systems and Signal ProcessingEngineering43 references70 citations

TL;DR

Proposes a two-stage joint pilot optimization, target detection, and channel estimation scheme (J-PoTdCe) for ISAC systems, leveraging Turbo Sparse Bayesian inference and a rank-1 pilot optimization to exploit joint burst sparsity between radar and communication channels.

ABSTRACT

Radar sensing will be integrated into the 6G communication system to support various applications. In this integrated sensing and communication system, a radar target may also be a communication channel scatterer. In this case, the radar and communication channels exhibit certain joint burst sparsity. We propose a two-stage joint pilot optimization, target detection and channel estimation scheme to exploit such joint burst sparsity and pilot beamforming gain to enhance detection/estimation performance. In Stage 1, the base station (BS) sends downlink pilots (DP) for initial target search, and the user sends uplink pilots (UP) for channel estimation. Then the BS performs joint target detection and channel estimation based on the reflected DP and received UP signals. In Stage 2, the BS exploits the prior information obtained in Stage 1 to optimize the DP signal to achieve beamforming gain and further refine the performance. A Turbo Sparse Bayesian inference algorithm is proposed for joint target detection and channel estimation in both stages. The pilot optimization problem in Stage 2 is a semi-definite programming with rank-1 constraints. By replacing the rank-1 constraint with a tight and smooth approximation, we propose an efficient pilot optimization algorithm based on the majorization-minimization method. Simulations verify the advantages of the proposed scheme.

Motivation & Objective

• Motivate integration of radar sensing and communication in 6G ISAC systems and exploit joint burst sparsity between radar targets and communication scatterers.
• Develop a two-stage framework to perform joint target detection and channel estimation using optimized pilots.
• Leverage sparse Bayesian inference to jointly detect targets and estimate channels under a dynamic AoA grid.
• Improve low-SNR performance through pilot beamforming gains and information sharing across stages.

Proposed method

• Two-stage J-PoTdCe framework: Stage 1 uses omnidirectional downlink pilots and uplink pilots for initial search and estimation; Stage 2 uses prior information to optimize downlink pilots toward targets/scatterers for refinement.
• Turbo Sparse Bayesian Inference (Turbo-SBI) with a hidden Markov model to capture joint burst sparsity of radar and communication channels.
• Dynamic angular grid for sparse angular domain representation to mitigate grid mismatch while preserving resolution.
• A joint measurement model linking reflected downlink pilots and uplink pilots; Bayesian inference yields posteriors for x^r, x^c, s^r, s^c and grid theta.
• Pilot optimization in Stage 2 formulated as SDP with rank-1 constraints; replaced by a tight smooth approximation and solved via majorization-minimization (MM).
• Comparison to Turbo-OAMP and SDR approaches, with simulations under CDL channel models to verify gains.

Experimental results

Research questions

• RQ1 How can joint burst sparsity between radar targets and communication scatterers be exploited to improve target detection and channel estimation in ISAC systems?
• RQ2 Can a two-stage pilot design and Turbo SBI framework enhance detection/estimation performance, especially in low SNR regimes?
• RQ3 How to design Stage-2 pilots to maximize beamforming gain and minimize CRB for target parameters using a rank-1 constrained optimization?
• RQ4 What is the impact of dynamic AoA grids on estimation accuracy and computational complexity in massive MIMO ISAC setups?

Key findings

• A two-stage J-PoTdCe scheme with Turbo-SBI achieves joint target detection and channel estimation by exploiting joint burst sparsity.
• Dynamic grid AoA representation with a large grid improves robustness to grid mismatch and enables super-resolution estimation.
• Stage-2 pilot optimization via MM-based approach with rank-1 relaxation provides efficient beamforming gain and CRB reduction.
• Turbo-SBI effectively computes marginal posteriors for x^r, x^c, s^r, s^c under the joint HMM prior, achieving improved performance in simulations.
• The framework is validated through simulations using the 3GPP R15 CDL channel model, showing advantages over traditional methods.