Boosted by mobile communication technologies, Human Activity Recognition (HAR) based on smartphones has attracted more and more attentions of researchers. One of the main challenges is the classification time and accuracy in processing long-time dependent sequence samples with noisy or missed data. In this paper, a 1-D Convolution Neural Network (CNN)-based bi-directional Long Short-Term Memory (LSTM) parallel model with attention mechanism (ConvBLSTM-PMwA) is proposed. The original features of sensors are segmented into sub-segments by well-designed equal time step sliding window, and fed into 1-D CNN-based bi-directional LSTM parallel layer to accelerate feature extraction with noisy and missed data. The weights of extracted features are redistributed by attention mechanism and integrated into complete features. At last, the final classification results are obtained with the full connection layer. The performance is evaluated on public UCI and WISDM HAR datasets. The results show that the ConvBLSTM-PMwA model performs better than the existing CNN and RNN models in both classification accuracy (96.71%) and computational time complexity (1.1 times faster at least), even if facing HAR data with noise.

Existing intrusion detection and defense models for CPSS (Cyber-Physical-Social Systems) are based on analyzing the static intrusion characteristics, which cannot effectively detect large-scale Low-Rate Denial-of-Service (LR-DDoS) attacks, especially in the edge environment. In this paper, we firstly explore and enhance Mirai botnet to a sophisticated multi-targets low-rate TCP attack network, which makes edge LR-DDoS more powerful and obfuscates their activity. And then, we develop a novel intrusion detection and defense hybrid method for above CPSS LR-DDoS scenario in edge environment, which takes advantage of locality sensitive features extraction and Deep Convolution Neural Network (DCNN) to auto learn the optimal features of the original data distribution and employs deep reinforcement learning Q-network as the powerful decision maker to defend attacks. The experimental results in detection phase prove the proposed method can distinguish abnormal network attack flows with higher detection accuracy and faster response time than kinds of Support Vector Machines (SVM), K-means and Surface Learning Neural Network etc. Even more, it has a certain detection rate for unknown new attacks, which means the method is effective and suitable for the actual network environment. The experimental results in defense phase prove it can defense LR-DDoS attacks smoothly.

We present an innovative approach for a Cybersecurity Solution based on the Intrusion Detection System to detect malicious activity targeting the Distributed Network Protocol (DNP3) layers in the Supervisory Control and Data Acquisition (SCADA) systems. As Information and Communication Technology is connected to the grid, it is subjected to both physical and cyber-attacks because of the interaction between industrial control systems and the outside Internet environment using IoT technology. Often, cyber-attacks lead to multiple risks that affect infrastructure and business continuity; furthermore, in some cases, human beings are also affected. Because of the traditional peculiarities of process systems, such as insecure real-time protocols, end-to-end general-purpose ICT security mechanisms are not able to fully secure communication in SCADA systems. In this paper, we present a novel method based on the DNP3 vulnerability assessment and attack model in different layers, with feature selection using Machine Learning from parsed DNP3 protocol with additional data including malware samples. Moreover, we developed a cyber-attack algorithm that included a classification and visualization process. Finally, the results of the experimental implementation show that our proposed Cybersecurity Solution based on IDS was able to detect attacks in real time in an IoT-based Smart Grid communication environment.

Low-rate Distributed Denial of Service (LDDoS) attacks have been one of the most notorious network security threats, which use periodic slight multi-variate time series pulse flows to degrade network quality. Limited by the poor data in a single client, a powerful and satisfactory LDDoS attack detection model is hard to be trained. Federated Learning (FL) is a promising paradigm offering joint learning through multiple clients. We propose an asynchronous federated learning arbitration framework based on bidirectional LSTM (bi-LSTM) and attention mechanism (AsyncFL-bLAM). In the AsyncFL-bLAM, the leader node election algorithm is proposed for constructing the framework of asynchronous federated learning. The proposed bLAM model composed of feature extracter and arbitrator takes on the responsibility of LDDoS detection locally. Furthermore, the novel AsyncFL framework helps to upload and aggregate the bLAM models’ parameters asynchronously between leader node and client nodes. Experimental results show that the AsyncFL-bLAM outperforms the state-of-the-art models in accuracy, and reduces the overall communication rounds.

Low-rate Distributed DoS (LDDoS) attack is a complex large-scale attack behavior with strong time-domain characteristics in blockchain-based wireless network. Blockchain with Machine learning-based models, as promising ways, are taken to detect them and secure wireless network. However, researchers focused on how to improve models' detection performance and work out new blockchain-based protection technologies during the past decades. Due to lack of evolving data, these models and technologies may have poor stability in the face of confrontational samples. To cope with the problem, this paper proposes a novel LSTM-CGAN method to generate high-quality LDDoS adversarial samples for blockchain-based wireless network detection models. In this method, we give a brief feature analysis about LDDoS attack in blockchain-based wireless network and work out its corresponding time series model firstly. And then, we take use of Long Short-Term Memory Networks (LSTM) to learn relationships among sequenced network packages in the same flow. At last, we establish a Condition Generative Adversarial Networks (CGAN) model to use above relationships as specific conditions for generating mimicking behaviors of LDDoS attacks in blockchain-based wireless network. The experimental results show that these generated adversarial samples based on both public and private datasets can cheat the machine learning detection models, and have the similar attack characteristics with the real samples. Consequently, they can be used as blockchain-based wireless network dataset of machine learning classifiers for training to enhance models' stability.

In the age of the Internet of Things, connected devices are changing the delivery system in the healthcare communication environment. With the integration of IoT in healthcare, there is a huge potential for improvement of the quality, safety, and efficiency of health care in addition to promising technological, economical, and social prospects. Nevertheless, this integration comes with security risks such as data breach that might be caused by credential-stealing malware. In addition, the patient valuable data can be disclosed when the perspective devices are compromised since they are connected to the internet. Hence, security has become an essential part of today’s computing world regarding the ubiquitous nature of the IoT entities in general and IoT-based healthcare in particular. In this paper, research on the algorithm for anonymizing sensitive information about health data set exchanged in the IoT environment using a wireless communication system has been presented. To preserve the security and privacy, during the data session from the users interacting online, the algorithm defines records that cannot be revealed by providing protection to user’s privacy. Moreover, the proposed algorithm includes a secure encryption process that enables health data anonymity. Furthermore, we have provided an analysis using mathematical functions to valid the algorithm’s anonymity function. The results show that the anonymization algorithm guarantees safety features for the considered IoT system applied in context of the healthcare communication systems.