The Internet of things (IoT) has been widely used since its high efficiency and real-time collaboration. Wireless sensor network as the core technology to support the operation of the IoT, the security problem is becoming more and more serious. Aiming at the problem that the existing malicious node detection methods in wireless sensor networks can not be guarantee by fairness and traceability of detection process, in this paper, it presents a blockchain trust model (BTM) for malicious node detection in wireless sensor networks. Firstly, it gives the whole framework of the trust model. Then it constructs the blockchain data structure which is used to detect malicious nodes. Finally, it realizes the detection of malicious nodes in 3D space by using the blockchain smart contract and WSNs quadrilateral measurement localization method, and the voting consensus results are distributed recorded in the blockchain. The simulation results show that the model can effectively detect malicious nodes in WSNs and it can also ensure the traceability of the detection process.

Urban road traffic is a typical network. The analysis and study of the topology structure is the basis of the traffic state evaluation and the traffic organization optimization. This paper redefines the urban road traffic weighted network model with considering the functional properties of urban road network and presents the traffic efficiency concept of the road section in the urban road traffic network. According to the different edge weights which are determined by the different attribute of the edges, we analyses the statistical characteristic of the urban road traffic network based on the three different network models which include the length weighted network model, the traffic capacity weighted network model and the traffic efficiency weighted network model. We take Beijing road traffic network as an example to illustrate the effectiveness of the analysis methods for the structural characteristics

The relative low level of smart home system (SHS) device information security may threaten the privacy of users. In this paper, we propose a homomorphic consortium blockchain for SHS sensitive data privacy preserving (HCB-SDPP), which is based on the traditional smart home system. We add verification services, which are composed of verification nodes, to our model to verify working nodes and transactions in SHS. In order to record SHS device information transaction, we propose a new block data structure based on Homomorphic encryption (HEBDS). Using HCB-SDPP model, we design an encrypted algorithm based on Paillier encrypted for privacy protection. To verify the validity of the HCB-SDPP model, we firstly encrypt sensitive data of all gateway peers, and upload them to the consortium blockchain. Then we validate the security of sensitive data after homomorphic encryption processing. In the experiment, we also design attack experiments to attack different types of peers on the consortium blockchain in HCB-SDPP model. If these nodes are insecure, the influence on the whole model will be analyzed. Simulation result shows that HCB-SDPP model can protect customer privacy more effectively than SHS.

Human activity recognition (HAR) has been a very popular field in both real practice and theoretical research. Over the years, a number of many-vs-one Long Short-Term Memory (LSTM) models have been proposed for the sensor-based HAR problem. However, how to utilize sequence outputs of them to improve the HAR performance has not been studied seriously. To solve this problem, we present a novel loss function named harmonic loss, which is utilized to improve the overall classification performance of HAR based on baseline LSTM networks. First, label replication method is presented to duplicate true labels at each sequence step in many-vs-one LSTM networks, thus each sequence step can generate a local error and a local output. Then, considering the importance of different local errors and inspired by the Ebbinghaus memory curve, the harmonic loss is proposed to give unequal weights to different local errors based on harmonic series equation. Additionally, to improve the overall classification performance of HAR, integrated methods are utilized to exploit the sequence outputs of LSTM models based on harmonic loss and ensemble learning strategy. Finally, based on the LSTM model construction and hyper-parameter setting, extensive experiments are conducted. A series of experimental results demonstrate that our harmonic loss significantly achieves higher macro-F1 and accuracy than strong baselines on two public HAR benchmarks. Compared with previous state-of-art methods, our proposed methods can achieve competitive classification performance.

Long-term traffic forecasting has become a basic and critical work in the research on road traffic congestion. It plays an important role in alleviating road traffic congestion and improving traffic management quality. According to the problem that long-term traffic forecasting is short of systematic and effective methods, a long-term traffic situation forecasting model is proposed in this article based on functional nonparametric regression. In the functional nonparametric regression framework, autocorrelation analysis (ACF) is introduced to analyze the autocorrelation coefficient of traffic flow for selecting the state vector, and the functional principal component analysis is also used as distance function for computing proximities between different traffic flow time series. The experiments based on the traffic flow data in Beijing expressway prove that the functional nonparametric regression model outperforms forecast methods in accuracy and effectiveness.