Kuai Xu | Xuanli Lin

Inferring User Activities from Internet of Things (IoT) Connected Device Events Using Machine Learning Based Algorithms

Thu, 05 Dec 2024 00:00:00 +0000

Abstract

An Internet-of-Things (IoT) learning framework (IoT learning framework) may train AI models to infer user activities from IoT connected device events using machine learning based algorithms. According to such an example, processing circuitry obtains a training dataset indicating IoT device events and extracts representative user activity patterns from the sequences of IoT device events. In such an example, processing circuitry trains the AI model to learn an optimal subset of the sequences of IoT device events corresponding to a smallest quantity of the sequences of IoT device events to predict user activities with accuracy that satisfies a threshold and outputs the AI model. According to such an example, processing circuitry may obtain new data indicating new sequences of IoT device events and generates output indicating one or more user activities predicted by the AI model to have occurred based on the new sequences of IoT device events.

Guoliang Xue, Yinxin Wan, Xuanli Lin, Kuai Xu, and Feng Wang. “Inferring User Activities from Internet of Things (IoT) Connected Device Events Using Machine Learning Based Algorithms.” U.S. Patent Application Publication US20240403648A1, published December 5, 2024.

This record is a published patent application. Application 18/734,240 was filed June 5, 2024, with a provisional application 63/506,316 filed June 5, 2023. The assignee listed on the publication is the Arizona Board of Regents on Behalf of Arizona State University.

Exploring Machine Learning Algorithms for User Activity Inference from IoT Network Traffic

Tue, 26 Sep 2023 00:00:00 +0000

Abstract

The availability of ubiquitous and heterogeneous Internet-of-Things (IoT) devices in smart homes and their interactions with users provide a unique opportunity to monitor, understand, recognize, learn, and infer user activities for safety monitoring, connected health, energy saving as well as other disruptive services. Our analysis on IoT network traffic from smart homes with a variety of IoT devices has discovered that user activities often trigger overlapping traffic waves from multiple IoT devices that are deployed near the activities. This insight leads us to adopt wavelet analysis to decompose IoT network traffic in smart homes into low, middle, and high frequency bands that distinguish IoT traffic waves triggered by user activities from background noises such as heartbeat signals between IoT devices and cloud servers. Subsequently, we extract a broad range of traffic features from these IoT traffic waves and explore supervised machine learning (ML) algorithms to classify various user activities with these features. Based on the labelled user activities and IoT network traffic data collected from real smart home environments, our experiments have demonstrated that the ML-based algorithms are able to use IoT network traffic to accurately infer various user activities in smart homes.

Best Paper Award Recipient

Extracting Spatial Information of IoT Device Events for Smart Home Safety Monitoring

Fri, 19 May 2023 00:00:00 +0000

Abstract

Smart home IoT devices have been widely deployed and connected to many home networks for various applications such as intelligent home automation, connected healthcare, and security surveillance. The network traffic traces generated by IoT devices have enabled recent research advances in smart home network measurement. However, due to the cloud-based communication model of smart home IoT devices and the lack of traffic data collected at the cloud end, little effort has been devoted to extracting the spatial information of IoT device events to determine where a device event is triggered. In this paper, we examine why extracting IoT device events’ spatial information is challenging by analyzing the communication model of the smart home IoT system. We propose a system named IoTDuet for determining whether a device event is triggered locally or remotely by utilizing the fact that the controlling devices such as smartphones and tablets always communicate with cloud servers with relatively stable domain name information when issuing commands from the home network. We further show the importance of extracting spatial information of IoT device events by exploring its applications in smart home safety monitoring.

IoT System Vulnerability Analysis and Network Hardening with Shortest Attack Trace in a Weighted Attack Graph

Tue, 09 May 2023 00:00:00 +0000

Abstract

In recent years, Internet of Things (IoT) devices have been extensively deployed in edge networks, including smart homes and offices. Despite the exciting opportunities afforded by the advancements in the IoT, it also introduces new attack vectors and vulnerabilities in the system. Existing studies have shown that the attack graph is an effective model for performing system-level analysis of IoT security. In this paper, we study IoT system vulnerability analysis and network hardening. We first extend the concept of attack graph to weighted attack graph and design a novel algorithm for computing a shortest attack trace in a weighted attack graph. We then formulate the network hardening problem. We prove that this problem is NP-hard, and then design an exact algorithm and a heuristic algorithm to solve it. Extensive experiments on 9 synthetic IoT systems and 2 real-world smart home IoT testbeds demonstrate that our shortest attack trace algorithm is robust and fast, and our heuristic network hardening algorithm is efficient in producing near optimal results compared to the exact algorithm.

Best Paper Award Recipient

Inferring User Activities from IoT Device Events in Smart Homes: Challenges and Opportunities

Mon, 05 Sep 2022 00:00:00 +0000

Abstract

The ubiquitous deployment of IoT devices in smart homes has led to growing research interests in studying the home network traffic for various applications such as network measurements, device profiling, and IoT device event inference. Recent studies have shown that user activities can be inferred from a home network using extracted device event logs. However, existing solutions for user activity inference such as IoTMosaic and E2AP have limitations when handling ambiguities caused by device malfunctions. In this paper, we first identify the challenges faced by the existing user activity inference algorithms and the root causes of their poor performances on certain types of inputs. We then show that useful information can still be obtained even in situations where device malfunctions introduce ambiguities in user activity patterns. We achieve so by designing an extension to the existing algorithms. We also apply our extension in a digital forensics application. Our extensive experimental evaluations demonstrate that our solutions can effectively provide insights to user activity inference despite the presence of indistinguishable user activity patterns.

An Effective Machine Learning Based Algorithm for Inferring User Activities From IoT Device Events

Wed, 20 Jul 2022 00:00:00 +0000

Abstract

The rapid and ubiquitous deployment of Internet of Things (IoT) in smart homes has created unprecedented opportunities to automatically extract environmental knowledge, awareness, and intelligence. Many existing studies have adopted either machine learning approaches or deterministic approaches to infer IoT device events and/or user activities from network traffic in smart homes. In this paper, we study the problem of inferring user activity patterns from a sequence of device events by first deterministically extracting a small number of representative user activity patterns from the sequence of device events, then applying unsupervised learning to compute an optimal subset of these user activity patterns to infer user activity patterns. Based on extensive experiments with sequences of device events triggered by 2,959 real user activities and up to 30,000 synthetic user activities, we demonstrate that our scheme is resilient to device malfunctions and transient failures/delays, and outperforms the state-of-the-art solution.