Optimal AI Model Splitting and Resource Allocation for Device-Edge Co-Inference in Multi-User Wireless Sensing Systems
Authors:
- 25 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 23 (9) , 11094-11108
Abstract
With recent advancements in artificial intelligence (AI), wireless sensing has recently been accepted as an attractive solution to enable accurate detection of human activities by analyzing the radio signal variations of sensor devices (SDs) using a well-trained AI model. However, due to the limited communication and computation resources at SDs, it is impractical to support energy- and delay-sensitive sensing services by solely processing the massive computation workload at local or offloading it to the edge server (ES) for edge inference. To address this problem, we consider in this paper device-edge co-inference in a wireless sensing system where multiple users collaboratively perform a common inference task. In particular, the AI model deployed at each SD can be split into two sequential parts. Each SD executes the former part of AI model at local, and leaves the remaining part computed at the ES. We aim to minimize the energy consumption of SDs subject to a prescribed inference latency requirement. To this end, we formulate a mixed integer non-linear programming (MINLP) to jointly optimize the model splitting point and system resource allocation, where the major difficulty lies in the tight couplings among splitting decisions of collaborative SDs. To solve the problem, we propose an integrated learning and optimization algorithm named LOP, which tackles the combinatorial model splitting by using a deep reinforcement learning (DRL)-based method, and deals with the remaining resource allocation problem via convex optimization. To gain some engineering insights, we study the optimal model splitting design in a practical wireless indoor crowd counting system, where the optimal splitting point exhibits a threshold-based structure related to the user channel gain. Simulation results demonstrate that the proposed LOP algorithm can achieve a near-optimal energy performance with on average 0.8% optimality gap while enjoying a hundredfold reduction in computation delay.Keywords
Funding Information
- National Natural Science Foundation of China (62271325)
- Basic and Applied Basic Research Foundation of Guangdong Province (2024B1515020089, 2022A1515010973, 2024A1515010198, 2023A1515110158)
- Shenzhen Science and Technology Program (20220810142637001, JCYJ20210324093011030)
This publication has 32 references indexed in Scilit:
- DASECount: Domain-Agnostic Sample-Efficient Wireless Indoor Crowd Counting via Few-Shot LearningIEEE Internet of Things Journal, 2022
- Edge-Assisted Real-Time Instance Segmentation for Resource-Limited IoT DevicesIEEE Internet of Things Journal, 2022
- Deep Learning for Radio-Based Human Sensing: Recent Advances and Future DirectionsIEEE Communications Surveys & Tutorials, 2021
- WS-WiFi: Wired Synchronization for CSI Extraction on COTS-WiFi-TransceiversIEEE Internet of Things Journal, 2021
- WiFi-based Human Activity Recognition using Raspberry Pi2020
- Joint Multiuser DNN Partitioning and Computational Resource Allocation for Collaborative Edge IntelligenceIEEE Internet of Things Journal, 2020
- MEC-Assisted Immersive VR Video Streaming Over Terahertz Wireless Networks: A Deep Reinforcement Learning ApproachIEEE Internet of Things Journal, 2020
- Efficient partition of integer optimization problems with one-hot encodingScientific Reports, 2019
- Edge Intelligence: Paving the Last Mile of Artificial Intelligence With Edge ComputingProceedings of the IEEE, 2019
- A Survey on Mobile Edge Computing: The Communication PerspectiveIEEE Communications Surveys & Tutorials, 2017