IoT Network Security from the Perspective of Adversarial Deep Learning

Abstract

Machine learning finds rich applications in Internet of Things (IoT) networks such as information retrieval, traffic management, spectrum sensing, and signal authentication. While there is a surge of interest to understand the security issues of machine learning, their implications have not been understood yet for wireless applications such as those in IoT systems that are susceptible to various attacks due the open and broadcast nature of wireless communications. To support IoT systems with heterogeneous devices of different priorities, we present new techniques built upon adversarial machine learning and apply them to three types of over-the-air (OTA) wireless attacks, namely jamming, spectrum poisoning, and priority violation attacks. By observing the spectrum, the adversary starts with an exploratory attack to infer the channel access algorithm of an IoT transmitter by building a deep neural network classifier that predicts the transmission outcomes. Based on these prediction results, the wireless attack continues to either jam data transmissions or manipulate sensing results over the air (by transmitting during the sensing phase) to fool the transmitter into making wrong transmit decisions in the test phase (corresponding to an evasion attack). When the IoT transmitter collects sensing results as training data to retrain its channel access algorithm, the adversary launches a causative attack to manipulate the input data to the transmitter over the air. We show that these attacks with different levels of energy consumption and stealthiness lead to significant loss in throughput and success ratio in wireless communications for IoT systems. Then we introduce a defense mechanism that systematically increases the uncertainty of the adversary at the inference stage and improves the performance. Results provide new insights on how to attack and defend IoT networks using deep learning.

Security Threats and Artificial Intelligence Based Countermeasures for Internet of Things Networks: A Comprehensive Survey

“…Poisoning attacks occur when an intruder injects false training sample to an ML algorithm for taking a wrong decision. Sagduyu et al [138] found that poisoning attacks would minimize DNN-based IDS predictability. Feedforward neural network was used as defensive mechanism which was implemented to systematically increase the adversary's confusion at the inference stage and enhance efficiency.…”

“…Sagduyu et al [138] considered various defense models against attacks on ML techniques such as poisoning attack and evasion attack. The work introduced a Stackelberg game approach to maximize the performance of the defense proce-This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.…”

Gradient Masking of Label Smoothing in Adversarial Robustness

“…Research on IoT system security has progressed from the perspective of deep learning applications [29]- [31]. In [29], they used a machine learning based classification model to identify the types of devices that attempt to connect to an IoT network to secure devices which have potential security vulnerabilities.…”

“…In [29], they used a machine learning based classification model to identify the types of devices that attempt to connect to an IoT network to secure devices which have potential security vulnerabilities. IoT security with DNNs' vulnerabilities to several attacks such as inference attack, poisoning attacks, and evasion attacks (adversarial attack) have been addressed [31]. An adversary can use an adversarial attack to transform the input data to cause the target device or model to misclassify.…”

Real-Time and Embedded Deep Learning on FPGA for RF Signal Classification

“…Deep learning finds rich application in wireless communications. Examples include spectrum sensing [10], MIMO detection [11], channel estimation and signal detection [12], physical layer communications [13], jammer detection [14], stealth jamming [15], [16], power control [17], signal spoofing [18], and transmitter-receiver scheduling [19]. RF signal classification can support different applications such as radio fingerprinting [28] that can be ultimately used in cognitive radio systems [29] subject to dynamic and unknown interference and jamming effects [30].…”