Self-Configurable Cyber-Physical Intrusion Detection for Smart Homes Using Reinforcement Learning

Abstract

The modern Internet of Things (IoT)-based smart 1 home is a challenging environment to secure: devices change, 2 new vulnerabilities are discovered and often remain unpatched, 3 and different users interact with their devices differently and 4 have different cyber risk attitudes. A security breach's impact is 5 not limited to cyberspace, as it can also affect or be facilitated 6 in physical space, for example, via voice. In this environment, 7 intrusion detection cannot rely solely on static models that 8 remain the same over time and are the same for all users.9We present MAGPIE, the first smart home intrusion detection 10 system that is able to autonomously adjust the decision function 11 of its underlying anomaly classification models to a smart home's 12 changing conditions (e.g., new devices, new automation rules and 13 user interaction with them). The method achieves this goal by 14 applying a novel probabilistic cluster-based reward mechanism 15 to non-stationary multi-armed bandit reinforcement learning. 16 MAGPIE rewards the sets of hyperparameters of its underlying 17 isolation forest unsupervised anomaly classifiers based on the 18 cluster silhouette scores of their output. 19 Experimental evaluation in a real household shows that MAG-20 PIE exhibits high accuracy because of two further innovations: 21 it takes into account both cyber and physical sources of data; 22 and it detects human presence to utilise models that exhibit the 23 highest accuracy in each case. MAGPIE is available in open-24 source format, together with its evaluation datasets, so it can 25 benefit from future advances in unsupervised and reinforcement 26 learning and be able to be enriched with further sources of data 27 as smart home environments and attacks evolve.

A Survey on Metaverse: Fundamentals, Security, and Privacy

“…To be further adaptive to wearable devices with extreme size and energy constraints, Heartfield et al [100] propose a multilayered lightweight anomaly detection method by exploiting radio-frequency wireless communications to/from them to identify potentially malicious transactions. In [101], reinforcement learning methods are employed for intrusion detection in smallscale applications such as smart homes. The above defense approaches can provide some lessons to resist unknown/new threats in the metaverse.…”

The Applicability of Reinforcement Learning Methods in the Development of Industry 4.0 Applications

“…We can highlight state space design [12,25,33,107,144,179,193,208,217,220,222,224,227,266,267] and action space design [109,220,246,268], reward construction [14,76,110,199,220,226,246,[269][270][271][272][273], and exploration strategy planning [86,274] which can be determinants from the whole application point of view. [11,13,17,20,21,24,38,43,61,62,66,69,82,89,93], Allocation, assignment, resource management [20,22,…”

“…Referred publications Markov decision process [12,23,24,37,64,70,75,84,96,100,101,104,127,130,133,138,144,153,165,167,170,177,188,191,199], [203, 207, 211, 212, 214, 217, 220, 231, 252, 256-259, 263, 264, 272, 274, 281, 291, 309, 313, 320, 340, 343, 346], [369][370][371][372][373][374][375][376] Multiarmed bandit [61,66,102,198,351,377,378] Dynamic programming [16,19,27,52,68,70,84,90,93,…”

An Integrated Reinforcement DQNN Algorithm to Detect Crime Anomaly Objects in Smart Cities

“…The problems that may rise due to the connectivity of electronic devices within an intellectual city are as similar as those that arise in an intelligent home rather on a smaller scale, but the risks that may occur to the associated accessories with security and for handling of the information [5]. The consequences of a security flaw are not restricted to online; it could also influence or be assisted in spatial context, for as by speech.…”