Beam Scheduling for Cross-Layer ISAC: A Deep Reinforcement Learning Approach

2026-04-27

AI Summary

This work addresses the joint optimization challenge of low-latency communication and high-precision sensing in integrated sensing and communication (ISAC) systems under dynamic multi-user scenarios, where cross-layer coupling between data traffic and queue states further complicates resource allocation. The paper introduces, for the first time, deep reinforcement learning (DRL) into cross-layer ISAC beam scheduling, leveraging sensing observations as a surrogate for explicit channel state information to enable buffer-aware and environment-adaptive resource management without prior knowledge of angle-of-departure (AoD). The proposed approach significantly reduces signaling overhead while closely approaching the performance of an ideal benchmark with perfect AoD side information, achieving substantial gains in system throughput with only marginal latency increase.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search
• Multiagent Systems: Agent Communication

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Virtualization and resource management in Web systems and infrastructures
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Resource allocation in integrated sensing and communication (ISAC) systems needs to be optimized to balance the requirements of the communication and sensing modules considering complicated cross-layer data traffic and queue status in dynamic multi-user environments. This paper studies the beam allocation for cross-layer ISAC that achieves low-latency communication and minimizes sensing parameters estimation error. To handle the complex coupling between practical data buffer dynamics and varying wireless channels, we propose a deep reinforcement learning (DRL)-assisted approach. Rather than relying on explicit channel state information, the DRL-assisted beam allocation reduces feedback overhead by leveraging sensing observations. Simulation results verify that the DRL framework effectively takes buffer status into account and adapts to the wireless environment while allocating resources. The proposed multi-beam scheme improves overall throughput with only modest delay increases. Finally, the DRL-assisted beam management achieves both communication and sensing performance close to that of the genie-aided benchmark with perfect angle-of-departure (AoD) knowledge. These contributions advance the state-of-the-art intelligent resource management for ISAC systems.

Problem

• beam scheduling
• integrated sensing and communication
• cross-layer optimization
• resource allocation
• multi-user ISAC

Innovation

• Deep Reinforcement Learning
• Integrated Sensing and Communication
• Beam Scheduling
• Cross-Layer Optimization
• Buffer-Aware Resource Allocation