Abstract:This paper studies a new multi-device edge artificial-intelligent (AI) system, which jointly exploits the AI model split inference and integrated sensing and communication (ISAC) to enable low-latency intelligent services at the network edge. In this system, multiple ISAC devices perform radar sensing to obtain multi-view data, and then offload the quantized version of extracted features to a centralized edge server, which conducts model inference based on the cascaded feature vectors. Under this setup and by considering classification tasks, we measure the inference accuracy by adopting an approximate but tractable metric, namely discriminant gain, which is defined as the distance of two classes in the Euclidean feature space under normalized covariance. To maximize the discriminant gain, we first quantify the influence of the sensing, computation, and communication processes on it with a derived closed-form expression. Then, an end-to-end task-oriented resource management approach is developed by integrating the three processes into a joint design. This integrated sensing, computation, and communication (ISCC) design approach, however, leads to a challenging non-convex optimization problem, due to the complicated form of discriminant gain and the device heterogeneity in terms of channel gain, quantization level, and generated feature subsets. Remarkably, the considered non-convex problem can be optimally solved based on the sum-of-ratios method. This gives the optimal ISCC scheme, that jointly determines the transmit power and time allocation at multiple devices for sensing and communication, as well as their quantization bits allocation for computation distortion control. By using human motions recognition as a concrete AI inference task, extensive experiments are conducted to verify the performance of our derived optimal ISCC scheme.

Performance Optimization in Integrated Sensing, Computation, and Communication Systems.

Integrated Sensing, Computation, and Communication: System Framework and Performance Optimization

Integrated Sensing and Communication Enabled Multi-Device Multi-Target Cooperative Sensing: A Fairness-aware Design

Device Scheduling for Privacy-Aware Integrated Sensing, Computation, and Communication Systems

Experimental Study on ISAC Performance with Different Sensing Sequences

Reconfigurable Intelligent Surface Assisted Integrated Sensing, Communication and Computation Systems

Radio Resource Allocation for Integrated Sensing, Communication, and Computation Networks

Task-Oriented Sensing, Computation, and Communication Integration for Multi-Device Edge AI

Performance Optimization of Energy Efficient Semantic Communications over Wireless Networks

Efficient Design for NOMA Enabled Integrated Sensing and Semantic Communication

Robust Offloading for Edge Computing-Assisted Sensing and Communication Systems: A Deep Reinforcement Learning Approach

Integrated Sensing and Communication in IRS-assisted High-Mobility Systems: Design, Analysis and Optimization

Joint Device Scheduling and Resource Allocation for ISCC-Based Multi-View-Multi-Task Inference

Integrated Distributed Semantic Communication and Over-the-air Computation for Cooperative Spectrum Sensing

A Framework of RIS-assisted ICSC User-centric Based Systems: Latency Optimization and Design

Integrated Sensing and Communication from Learning Perspective: An SDP3 Approach

Integrated Sensing-Communication-Computation for Over-the-Air Edge AI Inference

Task-Oriented Integrated Sensing, Computation and Communication for Wireless Edge AI

Multi-Functional Beamforming Design for Integrated Sensing, Communication, and Computation

A Survey on Integrated Sensing, Communication, and Computation

Rethinking the fundamental performance limits of integrated sensing and communication systems