Cooperative Sensing-Assisted Predictive Beam Tracking for MIMO-OFDM Networked ISAC Systems

2025-08-18

AI Summary

This paper addresses the beam tracking problem for mobile target tracking in MIMO-OFDM networked ISAC systems, where multiple base stations (BSs) operate on non-overlapping frequency bands. To tackle this, we propose a sensing-assisted predictive beamforming framework that jointly exploits multi-BS cooperative sensing and target motion prediction. Communication rate and tracking robustness are co-optimized under a hard sensing accuracy constraint imposed by the Posterior Cramér–Rao Lower Bound (PC-CRLB). Target localization is performed via 2D-DFT-based local estimation followed by inter-BS measurement fusion using the Extended Kalman Filter (EKF). The resulting non-convex beamformer design problem is efficiently solved via semidefinite relaxation (SDR) combined with a penalty method. Compared to conventional approaches, the proposed framework significantly improves the achievable communication rate at the next time slot while guaranteeing sensing performance, offering high estimation accuracy, low computational complexity, and strong practicality.

Technology Category

• Search and Optimization: Mixed Discrete/Continuous Search
• Computer Vision: Motion & Tracking
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Security and Privacy: Tracking, profiling, and countermeasures against them

Abstract

This paper studies a multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) networked integrated sensing and communication (ISAC) system, in which multiple base stations (BSs) perform beam tracking to communicate with a mobile device. In particular, we focus on the beam tracking over a number of tracking time slots (TTSs) and suppose that these BSs operate at non-overlapping frequency bands to avoid the severe inter-cell interference. Under this setup, we propose a new cooperative sensing-assisted predictive beam tracking design. In each TTS, the BSs use echo signals to cooperatively track the mobile device as a sensing target, and continuously adjust the beam directions to follow the device for enhancing the performance for both communication and sensing. First, we propose a cooperative sensing design to track the device, in which the BSs first employ the two-dimensional discrete Fourier transform (2D-DFT) technique to perform local target estimation, and then use the extended Kalman filter (EKF) method to fuse their individual measurement results for predicting the target parameters. Next, based on the predicted results, we obtain the achievable rate for communication and the predicted conditional Cramér-Rao lower bound (PC-CRLB) for target parameters estimation in the next TTS, as a function of the beamforming vectors. Accordingly, we formulate the predictive beamforming design problem, with the objective of maximizing the achievable communication rate in the following TTS, while satisfying the PC-CRLB requirement for sensing. To address the resulting non-convex problem, we first propose a semi-definite relaxation (SDR)-based algorithm to obtain the optimal solution, and then develop an alternative penalty-based algorithm to get a high-quality low-complexity solution.

Problem

• Cooperative sensing-assisted predictive beam tracking for MIMO-OFDM ISAC systems
• Enhancing communication and sensing performance through BS cooperation
• Maximizing communication rate while satisfying sensing accuracy requirements

Innovation

• Cooperative sensing-assisted predictive beam tracking design
• Extended Kalman filter fusion for target parameter prediction
• Semi-definite relaxation algorithm for non-convex optimization