The Complexity–Performance Tradeoff in Resource Allocation for URLLC Exploiting Dynamic CSI

2021-09-01

IEEE Internet of Things Journal

AI Summary

URLLC-oriented IoT resource allocation faces a fundamental trade-off among ultra-low latency, high reliability, and computational complexity. To address this, we propose a dynamic CSI-driven lightweight graph neural network (GNN) framework for resource allocation. Our method introduces a novel dynamic pilot allocation mechanism to adaptively ensure CSI freshness and extends GNN-based modeling to time-varying channel environments by jointly incorporating CSI temporal correlation and greedy heuristic optimization. Evaluated in dense IoT networks, the proposed approach improves spectral efficiency by over 12% compared to conventional greedy algorithms; integrating dynamic pilot allocation yields an additional 3–5% gain. Moreover, it significantly enhances user fairness and system throughput while maintaining low computational overhead, strong scalability, and real-time feasibility—thereby achieving a balanced compromise among performance, timeliness, and practical deployability.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Graph-based Machine Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

The challenging applications envisioned for the future Internet of Things networks are making it urgent to develop fast and scalable resource allocation algorithms able to meet the stringent reliability and latency constraints typical of the ultra-reliable, low-latency communications (URLLCs). However, there is an inherent tradeoff between complexity and performance to be addressed: sophisticated resource allocation methods providing optimized spectrum utilization are challenged by the scale of applications and the concomitant stringent latency constraints. Whether nontrivial resource allocation approaches can be successfully applied in large-scale network instances is still an open question that this article aims to address. More specifically, we consider a scenario in which channel-state information (CSI) is used to improve spectrum allocation in a radio environment that experiences channel time correlation. Channel correlation allows the usage of CSI for longer time before an update, thus lowering the overhead burden. Following this intuition, we propose a dynamic pilot transmission allocation scheme in order to adaptively tune the CSI age. We systematically analyze the improvement of this approach applied to a sophisticated, recently introduced graph-based resource allocation method that we extend here to account for CSI. The results show that even in very dense networks and accounting for the higher computational time of the graph-based approach, this algorithm is able to improve spectrum efficiency by over 12% as compared to a greedy heuristic, and that dynamic pilot transmissions allocation can further boost its performance in terms of fairness, while concomitantly further increase spectrum efficiency of 3%–5%.

Problem

• Develop fast, scalable resource allocation for URLLC in IoT networks.
• Address complexity-performance tradeoff in spectrum utilization optimization.
• Enhance spectrum efficiency using dynamic CSI and pilot transmission.

Innovation

• Dynamic CSI for improved spectrum allocation
• Graph-based resource allocation method extension
• Dynamic pilot transmission to adapt CSI age