In the Internet of Vehicles (IoV), privacy preservation is a major challenge due to the mobility of vehicles and their resource-constrained nature. The limited resources of on-board units (OBUs) and embedded sensors lure the adversaries to launch various types of attacks. Thus, lightweight but reliable authentication schemes need to be designed to combat these attacks. Another major challenge is the scarcity of available bandwidth and excessive delay experienced by vehicles while they communicate with the servers located at the cloud. These servers execute various machine learning (ML) and deep learning (DL) algorithms to extract useful features from upstream traffic of IoV. To addresses these challenges, we propose an ML-based authentication scheme that trains and classifies the vehicles at the edge servers in a distributed manner, preserves the privacy of communicating entities and minimizes the bandwidth consumption and delay experienced by the vehicles. The ML-based approach extends the decision power of vehicles and edge servers to identify adversaries. Our scheme requires that each vehicle participates in an offline phase, where a trusted authority shares a list of MaskIDs and secret keys of legitimate vehicles and edge servers. A timestamp is embedded in the payload of each encrypted message to prune the proposed scheme against well-known adversarial attacks. The simulation results verify the exceptional performance of our scheme in terms of computational overhead, communication overhead, and storage overhead.

Abstract Although the Internet of Things (IoT) has been considered one of the most promising technologies to automate various daily life activities, that is, monitoring and prediction, it has become extremely useful for problem solving with the introduction and integration of artificial intelligence (AI)‐enabled smart learning methodologies. Therefore, due to their overwhelming characteristics, AI‐enabled IoTs have been used in different application environments, such as agriculture, where detection, prevention (if possible), and prediction of crop diseases, especially at the earliest possible stage, are desperately required. Bacterial stalk root is a common disease of tomatoes that severely affects its production and yield if necessary measures are not taken. In this article, AI and an IoT‐enabled decision support system (DSS) have been developed to predict the possible occurrence of bacterial stalk root diseases through a sophisticated technological infrastructure. For this purpose, Arduino agricultural boards, preferably with necessary embedded sensors, are deployed in the agricultural field of maize crops to capture valuable data at a certain time interval and send it to a centralized module where AI‐based DSS, which is trained on an equally similar data set, is implemented to thoroughly examine captured data values for the possible occurrence of the disease. Additionally, the proposed AI‐ and IoT‐enabled DSS has been tested on benchmark data sets, that is, freely available online, along with real‐time captured data sets. Both experimental and simulation results show that the proposed scheme has achieved the highest accuracy level in timely prediction of the underlined disease. Finally, maize crop plots with the proposed system have significantly increased the yield (production) ratio of crops.

Artificial Intelligence of Medical Things (AIoMT) is a hybrid of the Internet of Medical Things (IoMT) and artificial intelligence to materialize the acquisition of real-time data via the smart wearable devices. Due to a diverse geographical environment of IoMT, secure, and reliable communication among these devices is a challenging task that needs to be resolved on priority basis. For this purpose, numerous device-focused authentication approaches have been proposed in the literature, however, the problem still persists. This article introduces an advanced, secured, and efficient solution for the IoMT by leveraging a lightweight mutual authentication scheme as well as facilitating AI-enabled Big Data analytics and predictive modeling. The proposed approach is specifically designed to establish secured communication between wearable sensing devices and servers within IoMT by exploiting the desirable features of cloud–edge paradigm. In this approach, every device needs to verify whether the requesting wearable device is legitimate or not and this process needs to be carried out prior to the actual communication. Our proposed approach employs a hybrid of Advanced Encryption Standard, i.e., AES-128 bit and medium access control (MAC) for the establishment of secured communication sessions. In addition, the proposed approach utilizes real-time data collection from wearable devices, enabling predictive modeling for the early detection of health anomalies, thereby, enhancing the patient outcomes of a specific disease. This continuously adaptive approach excels in real-time decision making, promptly alerting healthcare professionals of potential risks. Simulation results have verified that the proposed approach serves an ideal solution for the resource-constrained devices by achieving the expected level of authenticity through minimum possible communication and processing overhead. Additionally, this scheme is prune against well-known security attacks in the AIoMT infrastructures.

Hyperspectral image (HSI) data has a wide range of spectral information that is valuable for numerous tasks. HSI data encounters some challenges, such as small training samples, data scarcity, and redundant information. Researchers present numerous investigations to address these challenges, with convolutional neural networks (CNNs) being extensively used in HSI classification because of their capacity to extract features from data. Moreover, vision transformers have demonstrated their ability in the remote sensing field. However, the training of these models required a significant amount of labeled training data. We proposed a vision-based transformer module that consists of a multiscale feature extractor to extract joint spectral-spatial low-level, shallow features. For high-level semantic feature extraction, we proposed a regional attention mechanism with a spatially gated module. We tested the proposed model on four publicly available HSI datasets: Pavia University, Salinas, Xuzhou, Loukia, and the Houston 2013 dataset. Using only 1%, 1%, 1%, 2%, and 2% of the training samples from the five datasets, we achieved the best classification in terms of overall accuracy (OA), average accuracy (AA), and Kappa coefficient.

Abstract With oceanic reverberation and a large amount of data being the main sources of interference for underwater acoustic target detection, it is difficult to obtain a more robust detection performance by relying on the traditional constant false alarm rate (CFAR) detection method. An adaptive sonar CFAR detection method based on a back propagation (BP) neural network is proposed. The method combines the artificial intelligence algorithm and the traditional detection algorithm, and uses the classification ability of the algorithm to select the detection algorithm, which can effectively improve the adaptation ability of the algorithm and the environment and the false alarm control ability. The method combines the artificial intelligence algorithm and the traditional detection algorithm, and uses the classification ability of the algorithm to select the detection algorithm, which can effectively improve the adaptation ability of the algorithm and the environment and the false alarm control ability. This method uses a BP neural network to train the target echo signal to complete the clutter background classification and establish the clutter background recognition classification set. According to the output result of each classification, the best CFAR detector is selected from four CA/SO/GO/OS‐CFAR detectors to detect the target. The simulation results show the detection performance of the proposed method in a uniform environment, a multi‐target environment, and a clutter edge environment. The results show that the environment adaptability is strong for different clutter backgrounds, which further improves the control ability of false alarms under a non‐uniform background.

Data integrity and authenticity are among the key challenges faced by the interacting devices of Internet of Things (IoT). The resource-constrained nature of sensor-embedded devices makes it even more difficult to design lightweight security schemes for these networks. In view of limited resources of the IoT devices, this article proposes a lightweight and trustworthy device-to-server mutual authentication scheme for edge-enabled IoT networks. Initially, a trusted authority generates and assigns identities (IDs) and mask them to servers and clients, also known as member devices, in an offline phase. These IDs are utilized to prevent possible infiltration of the adversary device(s). Next, every device ensures the authenticity of requesting devices using a sophisticated challenge, which is encrypted using a 128-b secret key,$\lambda _{i}$. Each device expects a reply from the intended destination device for resolving the encrypted challenge within the defined timeframe,$i.e., \bigtriangleup T$. Moreover, authenticity of the requesting device is verified through the stored IDs, which are shared in the offline phase. Simulation results have verified the exceptional performance of the proposed authentication scheme against field proven approaches in terms of computational and communication costs.

Abstract Automatic modulation classification is a task that is essentially required in many intelligent communication systems such as fibre‐optic, next‐generation 5G or 6G systems, cognitive radio as well as multimedia internet‐of‐things networks etc. Deep learning (DL) is a representation learning method that takes raw data and finds representations for different tasks such as classification and detection. DL techniques like Convolutional Neural Networks (CNNs) have a strong potential to process and analyse large chunks of data. In this work, we considered the problem of multiclass (eight classes) classification of modulated signals, which are, Binary Phase Shift Keying, Quadrature Phase Shift Keying, 16 and 64 Quadrature Amplitude Modulation corrupted by Additive White Gaussian Noise, Rician and Rayleigh fading channels using 3D‐CNN architectures in both frequency and spatial domains while deploying three approaches for data augmentation, which are, random zoomed in/out, random shift and random weak Gaussian blurring augmentation techniques with a cross‐validation (CV) based hyperparameter selection statistical approach. Simulation results testify the performance of 10‐fold CV without augmentation in the spatial domain to be the best while the worst performing method happens to be 10‐fold CV without augmentation in the frequency domain and we found learning in the spatial domain to be better than learning in the frequency domain.

Internet of Things (IoT) is producing an extraordinary volume of data daily, and it is possible that the data may become useless while on its way to the cloud, due to long distances. Fog/edge computing is a new model for analysing and acting on time-sensitive data, adjacent to where it is produced. Further, cloud services provided by large companies such as Google, can also be localised to improve response time and service agility. This is accomplished through deploying small-scale datacentres in various locations, where needed in proximity of users; and connected to a centralised cloud that establish a multi-access edge computing (MEC). The MEC setup involves three parties, i.e., service providers (IaaS), application providers (SaaS), network providers (NaaS); which might have different goals, therefore, making resource management difficult. Unlike existing literature, we consider resource management with respect to all parties; and suggest game-theoretic resource management techniques to minimise infrastructure energy consumption and costs while ensuring applications’ performance. Our empirical evaluation, using Google’s workload traces, suggests that our approach could reduce up to 11.95 percent energy consumption, and$\sim$17.86% user costs with negligible loss in performance. Moreover, IaaS can reduce up to 20.27 percent energy bills and NaaS can increase their costs-savings up to 18.52 percent as compared to other methods.

Cloud datacenters have become a backbone for today’s business and economy, which are the fastest-growing electricity consumers, globally. Numerous studies suggest that$\sim$30% of the US datacenters are comatose and the others are grossly less-utilized, which make it possible to save energy through resource consolidation techniques. However, consolidation comprises migrations that are expensive in terms of energy consumption and performance degradation, which is mostly not accounted for in many existing models, and, possibly, it could be more energy and performance efficient not to consolidate. In this paper, we investigate how migration decisions should be taken so that the migration cost is recovered, as only when migration cost has been recovered and performance is guaranteed, will energy start to be saved. We demonstrate through several experiments, using the Google workload data for 12,583 hosts and approximately one million tasks that belong to three different kinds of workload, how different allocation policies, combined with various migration approaches, will impact on datacenter’s energy and performance efficiencies. Using several plausible assumptions for containerised datacenter set-up, we suggest, that a combination of the proposed energy-performance-aware allocation (Epc-Fu) and migration (Cper) techniques, and migrating relatively long-running containers only, offers for ideal energy and performance efficiencies.

Industry 5.0 is the digitalization, automation and data exchange of industrial processes that involve artificial intelligence, Industrial Internet of Things (IIoT), and Industrial Cyber-Physical Systems (I-CPS). In healthcare, I-CPS enables the intelligent wearable devices to gather data from the real-world and transmit to the virtual world for decision-making. I-CPS makes our lives comfortable with the emergence of innovative healthcare applications. Similar to any other IIoT paradigm, I-CPS capable healthcare applications face numerous challenging issues. The resource-constrained nature of wearable devices and their inability to support complex security mechanisms provide an ideal platform to malevolent entities for launching attacks. To preserve the privacy of wearable devices and their data in an I-CPS environment, we propose a lightweight mutual authentication scheme. Our scheme is based on client-server interaction model that uses symmetric encryption for establishing secured sessions among the communicating entities. After mutual authentication, the privacy risk associated with a patient data is predicted using an AI-enabled Hidden Markov Model (HMM). We analyzed the robustness and security of our scheme using BurrowsAbadiNeedham (BAN) logic. This analysis shows that the use of lightweight security primitives for the exchange of session keys makes the proposed scheme highly resilient in terms of security, efficiency, and robustness. Finally, the proposed scheme incurs nominal overhead in terms of processing, communication and storage and is capable to combat a wide range of adversarial threats.

The wireless environment poses a significant challenge to the propagation of signals. Different effects such as multipath scattering, noise, degradation, distortion, attenuation, and fading affect the distribution of signals adversely. Deep learning techniques can be used to differentiate among different modulated signals for reliable detection in a communication system. This study aims at distinguishing COVID‐19 disease images that have been modulated by different digital modulation schemes and are then passed through different noise channels and classified using deep learning models. We proposed a comprehensive evaluation of different 2D Convolutional Neural Network (CNN) architectures for the task of multiclass (24‐classes) classification of modulated images in the presence of noise and fading. It is used to differentiate between images modulated through Binary Phase Shift Keying, Quadrature Phase Shift Keying, 16‐ and 64‐Quadrature Amplitude Modulation and passed through Additive White Gaussian Noise, Rayleigh, and Rician channels. We obtained mixed results under different settings such as data augmentation, disharmony between batch normalization (BN), and dropout (DO), as well as lack of BN in the network. In this study, we found that the best performing model is a 2D‐CNN model using disharmony between BN and DO techniques trained using 10‐fold cross‐validation (CV) with a small value of DO before softmax and after every convolution and fully connected layer along with BN layers in the presence of data augmentation, while the least performing model is the 2D‐CNN model trained using 5‐fold CV without augmentation.

Summary Software Defined Network (SDN) and Network Virtualization (NV) are emerged paradigms that simplified the control and management of the next generation networks, most importantly, Internet of Things (IoT), Cloud Computing, and Cyber‐Physical Systems. The Internet of Things (IoT) includes a diverse range of a vast collection of heterogeneous devices that require interoperable communication, scalable platforms, and security provisioning. Security provisioning to an SDN‐based IoT network poses a real security challenge leading to various serious security threats due to the connection of various heterogeneous devices having a wide range of access protocols. Furthermore, the logical centralized controlled intelligence of the SDN architecture represents a plethora of security challenges due to its single point of failure. It may throw the entire network into chaos and thus expose it to various known and unknown security threats and attacks. Security of SDN controlled IoT environment is still in infancy and thus remains the prime research agenda for both the industry and academia. This paper comprehensively reviews the current state‐of‐the‐art security threats, vulnerabilities, and issues at the control plane. Moreover, this paper contributes by presenting a detailed classification of various security attacks on the control layer. A comprehensive state‐of‐the‐art review of the latest mitigation techniques for various security breaches is also presented. Finally, this paper presents future research directions and challenges for further investigation down the line.