Adversarial Attack for Deep Learning Based IoT Appliance Classification Techniques
Abhijit Singh, Biplab Sikdar · 2021
Simultaneous advancements in Internet of Things and Machine Learning have resulted in fascinating interdisciplinary applications, such as classification tasks based on smart device generated data for diverse applications such as resource allocation, security, and activity classification. However, such applications may be susceptible to attacks by adversarial examples. In this paper, we develop a white-box adversarial attack mechanism to generate adversarial examples for data obtained from smart-meters installed in residential houses, and demonstrate that their statistical properties are indistinguishable from those of the true datapoints. The attack mechanism focuses specifically on Deep Learning based models used to perform appliance classification in smart home environments. The statistical indistinguishability of the adversarial datapoints from the true datapoints indicates that non Machine Learning based solutions may not be able to tackle the challenge posed by adversarial examples. The effectiveness of the proposed techniques is demonstrated using the publicly available United Kingdom-Domestic Appliance-Level Electricity smart-meter dataset.