The rapid development of the power Internet of Things (IoT) has greatly enhanced the level of security, quality and efficiency in energy production, energy consumption, and related fields. However, it also puts forward higher requirements for the security and privacy of data. Ciphertext-policy attribute-based encryption (CP-ABE) is considered a suitable method to solve this issue and can implement fine-grained access control. However, its internal bilinear pairing operation is too expensive, which is not suitable for power IoT with limited computing resources. Hence, in this paper, a novel CP-ABE scheme based on elliptic curve cryptography (ECC) is proposed, which replaces the bilinear pairing operation with simple scalar multiplication and outsources most of the decryption work to edge devices. In addition, time and location attributes are combined in the proposed scheme, allowing the data users to access only within the range of time and locations set by the data owners to achieve a more fine-grained access control function. Simultaneously, the scheme uses multiple authorities to manage attributes, thereby solving the performance bottleneck of having a single authority. A performance analysis demonstrates that the proposed scheme is effective and suitable for power IoT.

With the reform of the power system, the prediction of power market pricing has become one of the key problems that needs to be solved in time. Power price prediction plays an important role in maximizing the profits of the participants in the power market and making full use of power energy. In order to improve the prediction accuracy of the power price, this paper proposes a power price prediction method based on PSO optimization of the XGBoost model, which optimizes eight main parameters of the XGBoost model through particle swarm optimization to improve the prediction accuracy of the XGBoost model. Using the electricity price data of Australia from January to December 2019, the proposed model is compared with the XGBoost model. The experimental results show that PSO can effectively improve the performance of the model. In addition, the prediction results of PSO-XGBoost are compared with those of SVM, LSTM, ARIMA, RW and XGBoost, and the average relative error and root mean square error of different power price prediction models are calculated. The experimental results show that the prediction accuracy of the PSO-XGBoost model is higher and more in line with the actual trend of power price change.

The research on malware detection enabled by deep learning has become a hot issue in the field of network security. The existing malware detection methods based on deep learning suffer from some issues, such as weak ability of deep feature extraction, relatively complex model, and insufficient ability of model generalization. Traditional deep learning architectures, such as convolutional neural networks (CNNs) variants, do not consider the spatial hierarchies between features, and lose some information on the precise position of a feature within the feature region, which is crucial for a malware file which has specific sections. In this paper, we draw on the idea of image classification in the field of computer vision and propose a novel malware detection method based on capsule network architecture with hyper-parameter optimized convolutional layers (MalCaps), which overcomes CNNs limitations by removing the need for a pooling layer and introduces capsule layers. Firstly, the malware is transformed into a grayscale image. Then, the dynamic routing-based capsule network is used to detect and classify the image. Without advanced feature extraction and with only a small number of labeled samples, the presented method is tested on an unbalanced Microsoft Malware Classification Challenge (MMCC) dataset and experimental results produce testing accuracy of 99.34%, improving on a number of traditional deep learning models posited in recent malware classification literature.

Power Internet of Things (IoT) is the application of IoT technology in the field of power grid, which can better control all kinds of power equipment, power personnel and operating environment. However, access to mass terminals brings higher requirements for terminal authentication and key management for the power IoT. And the traditional public key infrastructure (PKI) and identity-based public key cryptography (IB-PKC) exist the problems of certificate management and key escrow. Therefore, the paper proposes a novel authenticated key agreement scheme based on the certificateless public key cryptography (CL-PKC) mechanism. In addition, the proposed scheme is proven with the improved extended Canetti-Krawczyk (eCK) security model. Finally, the implementation of the authenticated key agreement protocol is given based on the actual application requirement of the power IoT, and the analysis and comparison of the simulation demonstrates that the proposed scheme has higher efficiency and would be suitable for the power IoT.

As a novel network security scheme, the zero trust model can effectively improve the traditional role-based access control model in the public network by continuous trust evaluation and dynamic authorization. As a private network for collecting the health status data of the power equipment, the terminals in the power Internet of Things (IoT) have the characteristics of fixed and controllable hardware, software, and users, and the main network security threat of the power IoT comes from the risk of terminals being damaged or forged. Therefore, based on the analysis of the key connotation of the zero trust and the security characteristics of the power IoT, this paper proposes a novel zero trust model and framework to synthetically evaluate the security status of the power IoT based on the behaviour and characteristics of the terminals. The analysis based on the stochastic Petri net and simulation results show that the proposed model can effectively address network security problems and significantly improve the level of security protection of the power IoT.

The echo state network (ESN) employs a huge reservoir with sparsely and randomly connected internal nodes and only trains the output weights, which avoids the suboptimal problem, exploding and vanishing gradients, high complexity and other disadvantages faced by traditional recurrent neural network (RNN) training. In light of the outstanding adaption to nonlinear dynamical systems, ESN has been applied into a wide range of applications. However, in the era of Big Data, with an enormous amount of data being generated continuously every day, the data are often distributed and stored in real applications, and thus the centralized ESN training process is prone to being technologically unsuitable. In order to achieve the requirement of Big Data applications in the real world, in this study we propose an algorithm and its implementation for distributed ESN training. The mentioned algorithm is based on the parallel particle swarm optimization (P-PSO) technique and the implementation uses Spark, a famous large-scale data processing framework. Four extremely large-scale datasets, including artificial benchmarks, real-world data and image data, are adopted to verify our framework on a stretchable platform. Experimental results indicate that the proposed work is accurate in the era of Big Data, regarding speed, accuracy and generalization capabilities.