Graph-Enhanced LLM for SWAN-ISAC

TL;DR

This paper introduces a learning framework that uses graph representations and large language models to optimize antenna deployment and beamforming in integrated sensing and communication systems.

Contribution

It presents a novel approach combining graph learning and LLMs with LoRA for joint optimization in ISAC systems.

Findings

Achieves a good tradeoff between communication rate and sensing accuracy.

Utilizes a CSI-induced self-graph to model interactions among users and targets.

Employs a LLM backbone with task-specific heads for prediction.

Abstract

Segmented pinching antenna assisted integrated sensing and communication (ISAC) systems enable flexible spatial resource utilization by allowing different waveguide segments to be dynamically configured for transmission and reception. However, the resulting design requires the joint optimization of antenna deployment, segment partitioning, and beamforming under coupled communication and sensing constraints. In this paper, we propose a general learning framework for segmented pinching antenna assisted ISAC systems. Specifically, a channel state information (CSI)-induced self-graph is constructed to capture the scenario-dependent interactions among communication users and sensing targets. Based on the learned graph representation, a large language model (LLM) backbone with low-rank adaptation (LoRA) is employed, followed by two task-specific output heads for antenna deployment and beamforming prediction, respectively. Simulation results show that the proposed framework achieves a favorable tradeoff between communication rate and sensing accuracy