Energy-Efficient Edge Inference in Integrated Sensing, Communication, and Computation Networks

TL;DR

This work tackles energy-efficient edge inference in task-oriented ISCC networks by jointly optimizing sensing, communication, and computation. It introduces a flexible ISCC framework with split inference, adaptive pruning, and stochastic feature quantization, and develops a low-complexity alternating-optimization algorithm with a golden-section search and KKT-based steps to minimize total energy under accuracy and latency constraints, including an explicit accuracy bound. The authors derive an efficient approximation for inference accuracy and demonstrate substantial energy savings—up to 40% in low-latency regimes—while revealing the critical role of sensing quality in achieving high classification performance. These findings enable practical, energy-conscious edge AI in industrial cyber-physical systems and motivate future extensions to multi-device collaborative inference under ISCC.

Abstract

Task-oriented integrated sensing, communication, and computation (ISCC) is a key technology for achieving low-latency edge inference and enabling efficient implementation of artificial intelligence (AI) in industrial cyber-physical systems (ICPS). However, the constrained energy supply at edge devices has emerged as a critical bottleneck. In this paper, we propose a novel energy-efficient ISCC framework for AI inference at resource-constrained edge devices, where adjustable split inference, model pruning, and feature quantization are jointly designed to adapt to diverse task requirements. A joint resource allocation design problem for the proposed ISCC framework is formulated to minimize the energy consumption under stringent inference accuracy and latency constraints. To address the challenge of characterizing inference accuracy, we derive an explicit approximation for it by analyzing the impact of sensing, communication, and computation processes on the inference performance. Building upon the analytical results, we propose an iterative algorithm employing alternating optimization to solve the resource allocation problem. In each subproblem, the optimal solutions are available by respectively applying a golden section search method and checking the Karush-Kuhn-Tucker (KKT) conditions, thereby ensuring the convergence to a local optimum of the original problem. Numerical results demonstrate the effectiveness of the proposed ISCC design, showing a significant reduction in energy consumption of up to 40% compared to existing methods, particularly in low-latency scenarios.

Figures (11)

• Figure 1: Model of ISCC based edge inference system.
• Figure 2: Comparison between the classification results obtained from the proposed approach and the ideal case.
• Figure 3: Geometry of classification margin under a classification problem.
• Figure 4: Classification accuracy versus $P_{\text{S}}$.
• Figure 5: Logical diagram of the proposed algorithm for solving problem (\ref{['pro2']}).