Intelligent Radio Resource Slicing for 6G In-Body Subnetworks

TL;DR

This paper introduces an intelligent radio resource slicing approach for 6G In-body Subnetworks using deep reinforcement learning to improve coexistence with other cellular users and enhance XR experience.

Contribution

It proposes a SAC-based resource slicing strategy that explicitly considers XR delay requirements and coexistence challenges, a novel application of DRL in this context.

Findings

Enhanced user experience by 12-85% in simulations.

Effectively balances XR delay and eMBB service guarantees.

Demonstrates robustness under realistic network conditions.

Abstract

6G In-body Subnetworks (IBSs) represent a key enabler for supporting standalone eXtended Reality (XR) applications. IBSs are expected to operate as an underlay to existing cellular networks, giving rise to coexistence challenges when sharing radio resources with other cellular users, such as enhanced Mobile Broadband (eMBB) users. Such resource allocation problem is highly dynamic and inherently non-convex due to heterogeneous service demands and fluctuating channel conditions. In this paper, we propose an intelligent radio resource slicing strategy based on the Soft Actor-Critic (SAC) deep reinforcement learning algorithm. The proposed SAC-based slicing method addresses the coexistence challenge between IBSs and eMBB users by optimizing a refined reward function that explicitly incorporates XR cross-modal delay alignment to ensure immersive experience while preserving eMBB service guarantees. Extensive system-level simulations are performed under realistic network conditions and the results demonstrate that the proposed method can enhance user experience by 12-85% under different network densities compared to baseline methods while maintaining the target data rate for eMBB users.