Various deep learning techniques, including blockchain-based approaches, have been explored to unlock the potential of edge data processing and resultant intelligence. However, existing studies often overlook the resource requirements of blockchain consensus processing in typical Internet of Things (IoT) edge network settings. This paper presents our FLCoin approach. Specifically, we propose a novel committee-based method for consensus processing in which committee members are elected via the FL process. Additionally, we employed a two-layer blockchain architecture for federated learning (FL) processing to facilitate the seamless integration of blockchain and FL techniques. Our analysis reveals that the communication overhead remains stable as the network size increases, ensuring the scalability of our blockchain-based FL system. To assess the performance of the proposed method, experiments were conducted using the MNIST dataset to train a standard five-layer CNN model. Our evaluation demonstrated the efficiency of FLCoin. With an increasing number of nodes participating in the model training, the consensus latency remained below 3 s, resulting in a low total training time. Notably, compared with a blockchain-based FL system utilizing PBFT as the consensus protocol, our approach achieved a 90% improvement in communication overhead and a 35% reduction in training time cost. Our approach ensures an efficient and scalable solution, enabling the integration of blockchain and FL into IoT edge networks. The proposed architecture provides a solid foundation for building intelligent IoT services.

In the recent era, the advancement of communication technologies provides a valuable interaction source between people of different regions. Nowadays, many organizations adopt the latest approaches, i.e., sentiment analysis and aspect-oriented sentiment classification, to evaluate user reviews to improve the quality of their products. The processing of multi-lingual user reviews is a key challenge in Natural Language Processing (NLP). This paper proposes a multi-layer network with divided attention to perform aspect-based sentiment classification for cross-lingual data. It extracts the Part-of-Speech (POS) tagging information of the given reviews, preprocesses them, and converts them into tokens. Furthermore, bi-lingual dictionaries are leveraged to map the converted tokens from one language to another. Given the preprocessed and mapped reviews, vectors are generated by leveraging the multi-lingual BERT and passed to the proposed deep learning classifier. The 10351 restaurant reviews from SemEval-2016 Task 5 dataset are exploited for the prediction of aspect-based sentiment. The results of cross-lingual validation suggest that the proposed approach significantly outperforms the state-of-the-art approaches and improves the precision, recall, and F1 by more than 23%, 20%, and 22%, respectively.

Kubernetes, a container orchestration tool for automatically installing and managing Docker containers, has recently begun to support a federation function of multiple Docker container clusters. This technology, called Kubernetes Federation, allows developers to increase the responsiveness and reliability of their applications by distributing and federating container clusters to multiple service areas of cloud service providers. However, it is still a daunting task to manually manage federated container clusters across all the service areas or to maintain the entire topology of cloud applications at a glance. This research work proposes a method to automatically form and monitor Kubernetes Federation, given application topology descriptions in TOSCA (Topology and Orchestration Specification for Cloud Applications), by extending the orchestration tool that automatizes the modeling and instantiation of cloud applications. It also demonstrates the successful federation of the clusters according to the TOSCA specifications and verifies the auto-scaling capability of the configured system through a scenario in which the servers of a sample application are deployed and federated.

We argue that the success of the Internet of Things (IoT) vision will greatly depend on how its main ingredient—the “thing”—is architected and prepared to engage. The IoT’s fragmented and widevarying nature introduces the need for additional effort to homogenize these things so they may blend together with the surrounding space to create opportunities for powerful and unprecedented IoT applications. We introduce the IoT Device Description Language (IoT-DDL), a machine- and human-readable descriptive language for things, seeking to achieve such integration and homogenization. IoT-DDL explicitly tools things to self-discover and securely share their own capabilities, entities, and services, including the various cloudbased accessories that may be attached to them. We also present the Atlas thing architecture—a lightweight architecture for things that fully exploits IoT-DDL and its specifications. Our architecture provides new OS layers, services, and capabilities we believe a thing must have in order to be prepared to engage in IoT scenarios and applications. The architecture and IoT-DDL enable things to generate their offered services and self-formulate APIs for such services, on the fly, at power-on or whenever a thing description changes. The architecture takes advantage of widely used device management, micro-services, security, and communication standards and protocols. We present details of IoT-DDL and corresponding parts of the thing architecture. We demonstrate some features of IoT-DDL and the architecture through proof-of-concept implementations. Finally, we present a benchmarking study to measure and assess time performance and energy consumption characteristics of our architecture and IoT-DDL on real hardware platforms.

In recent years, the Internet of Things (IoT) has gained tremendous attention and exponential growth in every domain of life. However, these devices face many challenges due to the limited resources in terms of storage and computation. Collaborative Edge Computing (CEC) is an emerging paradigm that solves these issues where multiple edge devices share computational resources to collaborate and satisfy user requirements. The fundamental issues in CEC are to make an offloading decision while considering the flow scheduling and balancing the load over multiple edge devices. Moreover, the efficient resource allocation of edge nodes is challenging, particularly when IoT devices are more vulnerable and become resource-hungry. The combination of Software Defined Networking (SDN) and Blockchain (BC) can play a vital role in solving the issues mentioned above. This paper presents a token-based resource management mechanism as SDBlockEdge by integrating these two technologies. The programming abstractions and global view of the SDN can help offload decision and flow scheduling, whereas the smart contract mechanism of BC can help control the abnormal behavior of IoT devices. We design the Resource Management Controller (RMC), which collaborates with the SDN controller to keep the record of available resources. It helps for offloading decisions and balancing the load over edge servers, whereas the SDN controller helps consider the less loaded path, while offloading reduces the task completion time. Moreover, the resources are allocated against tokens, and smart contracts are used to pay the cost of these resources. The proposed approach is implemented in Mininet_WiFi and Containernet, where docker hosts act as edge nodes with different IoT devices. The results after the extensive simulation show the effectiveness of the proposed approach.

In recent years, there is an exponential explosion of data generation, collection, and processing in computer networks. With this expansion of data, network attacks have also become a congenital problem in complex networks. The resource utilization, complexity, and false alarm rates are major challenges in current Network Intrusion Detection Systems (NIDS). The data fusion technique is an emerging technology that merges data from multiple sources to form more certain, precise, informative, and accurate data. Moreover, most of the earlier intrusion detection models suffer from overfitting problems and lack optimal detection of intrusions. In this paper, we propose a multi-source data fusion scheme for intrusion detection in networks (MIND) , where data fusion is performed by the horizontal emergence of two datasets. For this purpose, the Hadoop MapReduce tool such as, Hive is used. In addition, a machine learning ensemble classifier is used for the fused dataset with fewer parameters. Finally, the proposed model is evaluated with a 10-fold-cross validation technique. The experiments show that the average accuracy, detection rate, false positive rate, true positive rate, and F-measure are 99.80%, 99.80%, 0.29%, 99.85%, and 99.82% respectively. Moreover, the results indicate that the proposed model is significantly effective in intrusion detection compared to other state-of-the-art methods.