Heterogeneous IoT Research Articles

Dependable Intrusion Detection System for IoT: A Deep Transfer Learning Based Approach

Security concerns for IoT applications have been alarming because of their widespread use in different enterprise systems. The potential threats to these applications are constantly emerging and changing, and therefore, sophisticated and dependable defense solutions are necessary against such threats. With the rapid development of IoT networks and evolving threat types, the traditional machine learning-based IDS must update to cope with the security requirements of the current sustainable IoT environment. In recent years, deep learning, and deep transfer learning have progressed and experienced great success in different fields and have emerged as a potential solution for dependable network intrusion detection. However, new and emerging challenges have arisen related to the accuracy, efficiency, scalability, and dependability of the traditional IDS in a heterogeneous IoT setup. This manuscript proposes a deep transfer learning-based dependable IDS model that outperforms several existing approaches. The unique contributions include effective attribute selection, which is best suited to identify normal and attack scenarios for a small amount of labeled data, designing a dependable deep transfer learning-based ResNet model, and evaluating considering real-world data. To this end, a comprehensive experimental performance evaluation has been conducted. Extensive analysis and performance evaluation show that the proposed model is robust, more efficient, and has demonstrated better performance, ensuring dependability.

Enabling IoT Service Classification: A Machine Learning-Based Approach for Handling Classification Issues in Heterogeneous IoT Services

Enabling IoT Service Classification: A Machine Learning-Based Approach for Handling Classification Issues in Heterogeneous IoT Services

IoT trust aggregation using hybrid outlier detection and consensus

Trust modelling and management strategy used identify and mitigate threats by malicious devices rely on peer recommendations to compute trustworthiness. Aggregating opinions from independent devices is crucial in such recommendation-based systems to arrive at a consensus for decision making. Existing aggregation techniques like arithmetic mean, geometric mean (weighted/non-weighted), and maximum/minimum functions ignore the risk of biased and uncertain recommendations. To encounter such vulnerabilities, we propose a novel trust assessment model, outlier and uncertain recommendation-based trust management (OUR-Trust). It uses an outlier elimination and similarity-based scheme to evaluate the recommender's credibility before aggregation for consensus and decision making. The model employs revised Dempster-Shaffer combination rule for aggregation, which considers the uncertainty factor. Effectiveness of proposed approach is analysed for a dynamic and heterogeneous IoT network of dynamic and heterogeneous devices. OUR-Trust is validated for storage, power-efficiency, and scalability in terms of convergence time for more extensive IoT networks that employ recommender systems.

Real-Time Investigation of Cross-Technology Interference in Heterogeneous IoT Networks

Real-Time Investigation of Cross-Technology Interference in Heterogeneous IoT Networks

Modeling, Critical Threshold, and Lowest-Cost Patching Strategy of Malware Propagation in Heterogeneous IoT Networks

Modeling, Critical Threshold, and Lowest-Cost Patching Strategy of Malware Propagation in Heterogeneous IoT Networks

Noninvasive real-time traffic and congestion control algorithm based on policy

The Internet of Things has a large number of terminal devices and a wide range of deployment. This feature has higher requirements for reliable transmission and parallel connection, and is more prone to network congestion resulting in the loss of power information flow and excessive delay. The traditional congestion control algorithm is not suitable for the high concurrent mass heterogeneous iot terminal access control and mass information flow storage control. In this paper, the Transformer BBR algorithm is proposed, which is a congestion control algorithm based on BBR deep reinforcement learning combined with Transformer’s excellent long-term prediction ability. In the BBR algorithm bandwidth detection stage, transformer model is used as an agent to detect the network state and map it to the congestion window, so as to find the network changing state in time and make the corresponding decision action. Firstly, the congestion control strategy is learned in the simulation network environment, and then the efficiency is verified in the simulation environment. Finally, the experiment shows that the algorithm can reduce the delay while ensuring good bandwidth performance, and is superior to the main application algorithms in network throughput.

Semi-Supervised Federated Learning Over Heterogeneous Wireless IoT Edge Networks: Framework and Algorithms

Federated learning (FL) is a promising paradigm for future sixth-generation wireless systems to underpin network edge intelligence for smart cities applications. However, most of the data collected by the Internet of Things devices in such applications is unlabeled, necessitating the use of semi-supervised learning. Existing studies have introduced solutions to run semi-supervised FL; however, they overlooked the inherent critical impacts of the wireless characteristics at the network edge. We fill this gap by proposing novel solutions to run semi-supervised FL over wireless network edge, considering the limited computation and communication resources and deadline constraints and realizing that unlabeled data can be automatically labeled during the training rounds to improve the performance of the global model. The problem is first formulated as an optimization problem followed by a two-phase solution. In the first phase, we propose a bisection-based algorithm to find the transmit power and local processing speed that optimally fit the new injected labeled data. In the second phase, we propose three algorithms to control the local updates and injected samples that meet the deadline constraint. We analyze the performance of each algorithm concerning the tradeoffs between learning performance, training time, and total energy consumption. Targeting two applications in smart cities, human activity recognition and object detection, we conduct extensive simulations using realistic federated data sets under nonindependent and identically distributed settings. Numerical results show that the proposed algorithms effectively utilize unlabeled samples while accounting for the characteristics of wireless edge networks in smart cities.

Combining Device Behavioral Models and Building Schema for Cybersecurity of Large-Scale IoT Infrastructure

Modern buildings are increasingly getting connected by adopting a range of IoT devices and applications from video surveillance and lighting to people counting and access control. It has been shown that rich connectivity can make building networks more exposed to cyberattacks and, hence, difficult to manage. Currently, there is no systematic approach for evaluating or enforcing cybersecurity of building systems with a large number of heterogeneous IoT devices. In this article, we aim to enhance cybersecurity of a large-scale IoT infrastructure by formally capturing the expected behavior of the system using the static profile of devices’ intended usage, buildings information, and network configurations (predeployment) along with dynamic diagnosis (post-deployment) of network activity using machine-learning models. Our contributions are threefold: 1) we develop a tool that automatically generates a formal ontology of network communications for a connected infrastructure by taking a description of buildings (in the form of Brick schema), device network behavior (in the form of manufacturer usage description (MUD) specifications, MUD profile), and network configurations (address, port, and VLAN) as inputs. We contribute our tool as opensource, and apply it to a subset of our university smart campus testbed, covering 20 IoT devices of three types deployed in seven different buildings. We translate the formal model into network flow rules and enforce them to the network at runtime using programmable networking techniques; 2) we, then, measure the network activity of device-specific flow rules and diagnose their health using a set of trained anomaly detection models (one-class classifiers) each corresponding to a particular type of device and specific building location, and demonstrate how our method detects attacks with reasonable accuracy of 92.5%; and (3) finally, we demonstrate three types of location-defined network policies (deployment, administrative, and organizational) that can be verified by this formal model.

Communication-efficient semi-synchronous hierarchical federated learning with balanced training in heterogeneous IoT edge environments

Federated Learning (FL) aims to train a globally shared model by employing local data samples generated by data sources. The inherent heterogeneity of IoT environments, in terms of hardware, network technology, network condition, and power supply, imposes challenges to original FL associated with model convergence speed, model accuracy, balanced training, and communication costs. In this paper, we propose a Hierarchical FL (HFL) framework based on semi-synchronous communications to address heterogeneity issues of original FL in intelligent IoT Edge environments. In particular, the proposed approach runs the aggregation process of local models in a hierarchal manner to generate a global model and uses semi-synchronous communications to reduce communication costs. Moreover, a weighted aggregation method is introduced to overcome the unbalanced training problem of original FL in heterogeneous environments. To evaluate the performance of the proposed framework, we implement a heterogeneous IoT use case related to anomaly detection in a milling machining and design a prediction-based anomaly detection method based on Convolutional Neural Network (CNN), which is trained based on proposed Semi-synchronous HFL (SHFL). The results of implementations demonstrate the effectiveness of the proposed approach in comparison to fully synchronous and fully asynchronous FL approaches.

Towards SDN-based smart contract solution for IoT access control

Access control is essential for the IoT environment to ensure that only approved and trusted parties are able to configure devices, access sensor information, and command actuators to execute activities. The IoT ecosystem is subject to various access control complications due to the limited latency between IoT devices and the Internet, low energy requirements of IoT devices, the distributed framework, ad-hoc networks, and an exceptionally large number of heterogeneous IoT devices that need to be managed. The motivation for this proposed work is to resolve the incurring challenges of IoT associated with management and access control security. Each IoT domain implementation has particular features and needs separate access control policies to be considered in order to design a secure solution. This research work aims to resolve the intricacy of policies management, forged policies, dissemination, tracking of access control policies, automation, and central management of IoT nodes and provides a trackable and auditable access control policy management system that prevents forged policy dissemination by applying Software Defined Network (SDN) and blockchain technology in an IoT environment. Integration of SDN and blockchain provides a robust solution for IoT environment security. Recently, smart contracts have become one of blockchain technology’s most promising applications. The integration of smart contracts with blockchain technology provides the capability of designing tamper-proof and independently verifiable policies. In this paper, we propose a novel, scalable solution for implementing immutable, verifiable, adaptive, and automated access control policies for IoT devices together with a successful proof of concept that demonstrates the scalability of the proposed solution. The performance of the proposed solution is evaluated in terms of throughput and resource access delay between the blockchain component and the controller as well as from node to node. The number of nodes in the IoT network and the number of resource access requests were independently and systematically increased during the evaluations. The results illustrate that the resource access delay and throughput were affected neither linearly nor exponentially; hence, the proposed solution shows no significant degradation in performance with an increase in the number of nodes and/or requests.

Energy-Efficient Beamforming for Heterogeneous Industrial IoT Networks With Phase and Distortion Noises

The industrial Internet of Things (IIoT) is one of the key applications in 5G heterogeneous networks. To support high energy efficiency (EE) and reliability of IIoT equipment, it is important to design an efficient resource allocation algorithm in dynamic and complex environments. However, most of the studies on 5G heterogeneous IIoT networks did not address the transceiver hardware impairment (HWI) issues (e.g., phase noises, amplifier nonlinearities, and quantization errors) and the corresponding algorithms may not be applicable in practice. To this end, in this article, we investigate a realistic beamforming algorithm in a multicell downlink multiple-input single-output heterogeneous IIoT network by incorporating HWIs in our design. In particular, a beamforming design problem is formulated as a nonconvex optimization problem for maximizing the total EE of all equipment subject to the quality of service constraints of the IIoT equipment in both the macrocell and femtocells and the maximum transmit power constraints of base stations. In light of the intractability of the considered problem, we develop an EE-based iterative beamforming algorithm to tackle the formulated problem by employing the semidefinite relaxation method, Dinkelbach’s method, and the successive convex approximation method. Simulation results show that the proposed algorithm can achieve higher EE and bring less interference power to the macrocell IIoT equipment by comparing it with baseline algorithms.

Improving Situation Awareness via a Situation Model-Based Intersection of IoT Sensor and Social Media Information Spaces.

Existing techniques for distilling situation awareness currently focus on information harvested from either IoT sensors or social media. While the benefits of fusing information from these two distinct information spaces for achieving enhanced situation awareness are well understood, existing techniques and related systems for fusing the IoT sensors and social media information spaces are currently embryonic. Key challenges in intersecting, combining, and fusing these information spaces to distil high-value situation awareness include devising situation models and related techniques for filtering, integrating, and fusing sparse and heterogeneous IoT sensor data and social media postings to provide a richer and more accurate situation awareness. This paper proposes novel, semantically based techniques fusing social media and IoT sensor information spaces and a comprehensive, fully implemented system that utilizes these to provide enhanced situation awareness. More specifically, this paper proposes the design of semantic-based situation models for fusing sensor and social media information spaces and presents techniques for finding similarities across these information spaces and fusing social media posting and IoT sensor data to generate an enhanced situation awareness. Furthermore, the paper presents the design and implementation of a complete system that uses the proposed models and techniques and uses that in an experimental evaluation that illustrates improvements in situation awareness from fusing the IoT sensor and social media information spaces.

Detecting Attacks and Locating Malicious Devices Using Unmanned Air Vehicles and Machine Learning

Internet access in both private and public environments allows users to broadly access their data what makes possible the deployment of new services based on Internet of Things. This fact created Smart Environments (SEs) that are composed of a huge amount of heterogeneous devices, for example, personal devices (smartphones, notebooks, tablets, etc) and IoT devices (sensors, actuators, and others). However, these environments can facilitate the action of malicious agents interested in promoting Distributed Denial of Service (DDoS) attacks to the network, and, when they are public places, it is challenging to locate these attackers. In this way, it is necessary to deploy solutions that can detect DDoS in SEs and to determine the physical location of the attacker, which is essential to prevent future attacks. Within this context, this article presents an Intelligent System for detection of DDoS and physical location of devices in SEs, applying Machine Learning (ML) and trilateration techniques. The experiments performed, using real network traffic and simulation, suggest that the proposed system is capable of detecting attacks and finding malicious devices.

Towards dynamic and heterogeneous social IoT environments

Nowadays, there are millions of smart devices connected to the Internet. The purpose of these devices is to make people’s lives easier. Thanks to the collaboration among them, the possibilities that the Internet of Things brings can grow exponentially. However, many manufacturers develop closed protocols and devices to protect their market share, limiting in many ways this collaboration. This paper presents a conceptual architecture that improves the proactive collaboration between IoT devices regardless of the protocols developed by their manufacturers. This architecture aims to identify entities in smart environments, describe their features and interfaces, and identify strategies fostering their collaborations. As a result, devices from different manufacturers can communicate to create a collaborative environment in a simple, efficient, and affordable way. This architecture has been evaluated in a real and a simulated environment, to validate its feasibility and efficiency.

Novel Machine Learning (ML) Algorithms to Classify IPv6 Network Traffic in Resource-Limited Systems

Providing machine learning (ML) based security in heterogeneous IoT networks including resource-constrained devices is a challenge because of the fact that conventional ML algorithms require heavy computations. Therefore, in this paper, lightweight ProtoNN, CMSIS-NN, and Bonsai tree ML algorithms were evaluated by using performance metrics such as testing accuracy, precision, F1 score and recall to test their classification ability on the IPv6 network dataset generated on resource-scarce embedded devices. The Bonsai tree algorithm provided the best performance results in all metrics (98.8 in accuracy, 98.9% in F1 score, 99.2% in precision, and 98.8% in recall) compared to the ProtoNN, and CMSIS-NN algorithms.

The Rise of “Internet of Things”: Review and Open Research Issues Related to Detection and Prevention of IoT-Based Security Attacks

This paper provides an extensive and complete survey on the process of detecting and preventing various types of IoT-based security attacks. It is designed for software developers, researchers, and practitioners in the Internet of Things field who aim to understand the process of detecting and preventing these attacks. For each entry identified from the list, a brief description is provided along with references where more information can be found. However, We surveyed the current state-of-the-art IoT security solutions and focused on four main aspects: (1) handpicking representative attacks, (2) identifying potential solutions, (3) performing a threat analysis for each attack and solution, and (4) ranking solutions according to the threats they overcome. By adopting this framework, we identified five main categories of defense mechanisms: distributed denial of service detection/prevention, default password protection, encryption mechanisms, intrusion detection/prevention, and anomaly detection. These solutions are relatively mature in terms of utility and usability. However, the security analysis is conducted only concerning specific attacks, which may or may not be relevant to real-world deployment. Appropriate IoT security solutions should incorporate threat modeling while considering other factors such as resource consumption and implementation effort. Overall, evaluation of IoT security solutions is arduous due to the complexity of IoT OSes, heterogeneous IoT devices (e.g., various hardware platforms), limited availability of open-source codebases, and restrictive policies towards intellectual property disclosure. In addition, we note that there remains a lack of studies that perform a systematic evaluation of the state-of-the-art in terms of both frameworks/methodologies and mechanisms proposed.

IoTRepair: Flexible Fault Handling in Diverse IoT Deployments

IoT devices can be used to complete a wide array of physical tasks, but due to factors such as low computational resources and distributed physical deployment, they are susceptible to a wide array of faulty behaviors. Many devices deployed in homes, vehicles, industrial sites, and hospitals carry a great risk of damage to property, harm to a person, or breach of security if they behave faultily. We propose a general fault handling system named IoTRepair, which shows promising results for effectiveness with limited latency and power overhead in an IoT environment. IoTRepair dynamically organizes and customizes fault-handling techniques to address the unique problems associated with heterogeneous IoT deployments. We evaluate IoTRepair by creating a physical implementation mirroring a typical home environment to motivate the effectiveness of this system. Our evaluation showed that each of our fault-handling functions could be completed within 100 milliseconds after fault identification, which is a fraction of the time that state-of-the-art fault-identification methods take (measured in minutes). The power overhead is equally small, with the computation and device action consuming less than 30 milliwatts. This evaluation shows that IoTRepair not only can be deployed in a physical system, but offers significant benefits at a low overhead.

DSMLB: Dynamic switch-migration based load balancing for software-defined IoT network

The deployment of multiple controllers in Software-Defined Internet of Things (SD-IoT) networks enhance the distributed control plane scalability and reliability. With the rapid advancement of IoT, millions of smart devices generate enormous amounts of IoT traffic, causing significant challenges for the network to acquire the better Quality-of-Service (QoS) to the user. The control plane load management is crucial due to the dynamic nature of IoT networks and the heterogeneity of IoT devices in terms of flow generation rate. The load distribution among SD-IoT controllers relies on their available capacity to ensure optimal load balancing performance. The dynamic switch migration method was proposed in prior findings to handle load balancing distribution among controllers. However, the migration efficiency of conventional approaches is limited due to inefficient load balancing performance of the control plane after the migration of switch, leading to additional migration overhead on the network. Taking this into account, we propose a dynamic switch migration-based load balancing approach (DSMLB) to prevent control plane overhead and distribute traffic efficiently while considering heterogeneous IoT devices. The proposed DSMLB scheme measures the real-time load ratio to the mean load ratio of each controller by describing various load metrics and selecting the target controller by optimizing the migration efficiency with residual controller resource utilization. When choosing the candidate switches from the associated overloaded controller, DSMLB considers the migration efficiency and load balancing degree simultaneously for the fastest load reduction and enhances the load balancing performance. The simulation results show that DSMLB outperforms conventional approaches for efficient load balancing performance on a distributed control plane by reducing the controller response time up to 38.6% and migration cost about 45.5% on average. Moreover, DSMLB also improves controller load balancing rate and reduces communication overhead.

A Fault-Tolerant Workflow Composition and Deployment Automation IoT Framework in a Multicloud Edge Environment

With the increasing popularity of IoT application, e.g., smart home, smart manufacturing, the importance of underlying system availability, safety, reliability, and maintainability become crucial in the development processes. IoT applications are expected to continuously provide reliable services and features, which put the fault-tolerance mechanism as a priority. Current IoT fault-tolerant systems are designed to overcome any faults caused by human activities and physical errors to preserve the correct IoT workflow execution. However, addressing fault-tolerant interaction in multicloud edge environment and failed service deployment automation remains challenging. This article proposes a novel fault-tolerant model that offers the self-detection and automatic recovery of faults to increase IoT applications’ reliability and to address the infrastructure level failure in the heterogeneous IoT environments. Based on the proposed model, we implement a fault-tolerant workflow composition and deployment automation system, which is empowered by a layered architecture and a time-dependent failure model. The efficiency and effectiveness of the proposed system are validated and evaluated with a real-world IoT application.

Hardware–Software Co-Design Framework for Data Encryption in Image Processing Systems for the Internet of Things Environment

Data protection is a severe constraint in the heterogeneous IoT era. This article presents a Hardware-Software Co-Simulation of AES-128 bit encryption and decryption for IoT Edge devices using the Xilinx System Generator (XSG). VHDL implementation of AES-128 bit algorithm is done with ECB and CTR mode using loop unrolled and FSM-based architecture. It is found that AES-CTR and FSM architecture performance is better than loop unrolled architecture with lesser power consumption and area. For performing the Hardware-Software Co-Simulation on Zedboard and Kintex-Ultra scale KCU105 Evaluation Platform, Xilinx Vivado 2016.2 and MATLAB 2015b is used. Hardware emulation is done for grey images successfully. To give a practical example of the usage of proposed framework, we have applied it for Biomedical Images (CTScan Image) as a case study. Security analysis in terms of the histogram, correlation, information entropy analysis, and keyspace analysis using exhaustive search and key sensitivity tests is also done to encrypt and decrypt images successfully.