Ensuring Bidirectional Privacy on Wireless Split Inference Systems

• Published in: IEEE wireless communications Vol. 31; no. 5; pp. 134 - 141
• Main Authors: Sa, Chia-Che, Cheng, Li-Chen, Chung, Hsing-Huan, Chiu, Te-Chuan, Wang, Chih-Yu, Pang, Ai-Chun, Chen, Shang-Tse
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.10.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Artificial neural networks; Cloud computing; Computational modeling; Containers; Data models; Data privacy; Edge computing; Image segmentation; Inference; Machine learning; Network latency; Partitioning; Privacy; Semantic segmentation; Servers; User experience; Wireless communication; Wireless communications
• DOI: 10.1109/MWC.014.2300400
• ISSN: 1536-1284

With the advances of machine learning, edge computing, and wireless communications, split inference has tracked more and more attention as a versatile inference paradigm. Split inference is essential to accelerate large-scale deep neural network (DNN) inference on resource-limited edge devices through partitioning a DNN between the edge device and the cloud server with advanced wireless communications such as B5G/6G and WiFi 6. We investigate the U-shape partitioning inference system, where both the input raw data and output inference results are kept on the edge device. We use image semantic segmentation as an exemplary application in our experiments. The experiment results showed that an honest-but-curious (HbC) server can launch the bidirectional privacy attack to reconstruct the raw data and steal the inference results, even when only the middle-end partition of the model is visible. To ensure bidirectional privacy and user experience on the U-shape partitioning inference system, a privacy and latency-aware partitioning strategy is needed to balance the trade-off between service latency and data privacy. We compared our proposed framework to other inference paradigms, including conventional split inference and inferencing entirely on the edge device or the server. We analyzed their inference latencies in various wireless technologies and quantitatively measured their level of privacy protection. The experiment results show that the U-shape partitioning inference system is advantageous over inference entirely on the edge device or the server.