Lightweight and Unobtrusive Data Obfuscation at IoT Edge for Remote Inference

TL;DR

This paper proposes a lightweight, unobtrusive data obfuscation method for IoT edge devices that protects data privacy during remote neural network inference without compromising accuracy.

Contribution

It introduces a novel obfuscation approach that requires minimal computation and does not reveal obfuscation status, enhancing privacy in edge-to-cloud inference workflows.

Findings

Effective data confidentiality protection demonstrated across three case studies.

Maintains high inference accuracy despite obfuscation.

Low computational overhead suitable for IoT edge devices.

Abstract

Executing deep neural networks for inference on the server-class or cloud backend based on data generated at the edge of Internet of Things is desirable due primarily to the limited compute power of edge devices and the need to protect the confidentiality of the inference neural networks. However, such a remote inference scheme incurs concerns regarding the privacy of the inference data transmitted by the edge devices to the curious backend. This paper presents a lightweight and unobtrusive approach to obfuscate the inference data at the edge devices. It is lightweight in that the edge device only needs to execute a small-scale neural network; it is unobtrusive in that the edge device does not need to indicate whether obfuscation is applied. Extensive evaluation by three case studies of free spoken digit recognition, handwritten digit recognition, and American sign language recognition shows that our approach effectively protects the confidentiality of the raw forms of the inference data while effectively preserving the backend's inference accuracy.