Importance-Aware Resource Allocation for Collaborative Task-Oriented Semantic Communication

2026-06-27

AI Summary

This work addresses the challenge of efficiently allocating limited wireless resources to support task-oriented semantic communication under fast-fading channels and stringent end-to-end latency constraints. The authors propose iCoTASC, a novel framework that integrates semantic importance assessment and diminishing utility modeling into multi-device collaborative resource allocation. Specifically, attribution analysis is employed to quantify the importance of semantic features, guiding the selection of embedding dimensions, while a data-driven utility model captures the diminishing returns of semantic gains. The framework combines offline precomputed utility lookup tables with low-complexity online scheduling, enabling channel-adaptive, real-time resource allocation without requiring model retraining. This approach significantly reduces online computational overhead while maintaining superior task inference performance.

Abstract

Task-oriented semantic communication must allocate scarce radio resources to semantic features under fast fading wireless conditions and strict end-to-end latency budgets. Existing solutions are either optimization-heavy, leading to prohibitive computational overhead during online operation, or rely on end-to-end retraining procedures together with slowly varying channel assumptions. We propose iCoTASC (importance-aware Collaborative Task-Oriented Semantic Communication), a hybrid offline-online framework designed for collaborative multi-device semantic communication systems. iCoTASC leverages attribution-based importance to guide per-dimension embedding selection as a practical communication control signal, models diminishing semantic returns of quantization through a data-driven utility function, and precomputes per-transmitter utility lookup tables offline, which together enable lightweight online scheduling via table lookup and low-complexity refinement under time-varying channels. The proposed framework supports real-time, channel-adaptive semantic resource allocation in distributed systems without requiring retraining of the underlying task inference model.

Problem

• semantic communication
• resource allocation
• task-oriented
• fast fading channels
• latency constraints

Innovation

• semantic communication
• importance-aware resource allocation
• collaborative multi-device
• utility-driven quantization
• hybrid offline-online framework