Internet of Things (IoT) has more security issues due to the data being shared in an open platform. Integrating blockchain into IoT for security is a new development in computational communication systems. However, attackers are adapting their methods and creating new vulnerabilities in blockchain-based IoT platforms. Furthermore, when the blockchain is integrated with IoT networks, vulnerabilities, privacy issues, and security threats are amplified due to malicious transactions and active attacks. This paper proposes BlockDLO, an approach to IoT security that combines blockchain technology with deep learning. A five-phase architecture is proposed for the edge computing blockchain. In its first phase, network localization is resolved with chaotic map-based identification and authentication. The second phase proposes page rank-based clustering for edge computing. Then, BlockDLO combines the shared-chain technique with a deep distributed file system to address issues with block creation and ledger distribution, and an ethereum smart contract to address data security concerns. The communication route optimization is done with page rank centrality search optimization in its next phase. Finally, the integration of deep learning model to detect malicious data in the IoT network is done using the authenticated received data. BlockDLO creates an efficient intrusion detection system by combining a deep convolution neural network with blockchain. The proposed system is trained using public data sources and tested using an in-house network testbed. The results demonstrate that the proposed system outperforms existing work in terms of energy usage, packet loss rate, end-to-end delay, routing overhead, network lifetime, accuracy, and security strength.
The Internet of Things (IoT) presents new challenges to traditional communication models, particularly in terms of security, which are exacerbated by the rapid evolution of cyberthreats. Traditional security methods, especially those using Machine Learning models, often struggle with the limited computational resources available, making it difficult to detect attacks across the entire network. Software-defined networks (SDN) offer a solution by centralizing security policies, enabling more effective implementation and enforcement. The study investigates the SDN architecture from a security perspective. This article proposes a Deep Learning-based SDN architecture for IoT security which can significantly enhance real-time cyberthreat detection. Specifically, a secure communication channel is first designed using a blockchain-based authentication to resist well-known intruders. Second, a deep learning convergence model using an adaptive threshold scoring method that stops all local model training and allows edge models to contribute to the cloud model until a specified accuracy is achieved. To achieve low CPU usage and provide real-time services, SDN is next used as a cloud-based system security administrator to protect IoT networks from zero-day attacks by sending requests from edge devices to a cloud SDN controller. The efficiency of the proposed framework is demonstrated using simulations with two different network datasets E-IIoT and ToN-IoT against various attacks and the results are compared with similar works. The proposed model effectively detects and mitigates cyberthreats such as DDoS, Black-Sink-Worm hole, MitM, and Ransomware. It achieves high performance with 99.15% accuracy, 99.31% precision, 98.97% recall, and a 99.14% F1 score, on an average while using less CPU power for real-time IoT network protection.
31 August 2023 in International Journal on Recent and Innovation Trends in Computing and Communication
Agriculture is one of the major sectors that influence the India economy due to the huge population and ever-growing food demand. Identification of diseases that affect the low yield in food crops plays a major role to improve the yield of a crop. India holds the world's second-largest share of tomato production. Unfortunately, tomato plants are vulnerable to various diseases due to factors such as climate change, heavy rainfall, soil conditions, pesticides, and animals. A significant number of studies have examined the potential of deep learning techniques to combat the leaf disease in tomatoes in the last decade. However, despite the range of applications, several gaps within tomato leaf disease detection are yet to be addressed to support the tomato leaf disease diagnosis. Thus, there is a need to create an information base of existing approaches and identify the challenges and opportunities to help advance the development of tools that address the needs of tomato farmers. The review is focussed on providing a detailed assessment and considerations for developing deep learning-based Convolutional Neural Networks (CNNs) architectures like Dense Net, ResNet, VGG Net, Google Net, Alex Net, and LeNet that are applied to detect the disease in tomato leaves to identify 10 classes of diseases affecting tomato plant leaves, with distinct trained disease datasets. The performance of architecture studies using the data from plantvillage dataset, which includes healthy and diseased classes, with the assistance of several different architectural designs. This paper helps to address the existing research gaps by guiding further development and application of tools to support tomato leaves disease diagnosis and provide disease management support to farmers in improving the crop.