With increasingly severe network security threats and increasingly complex security defense means， zero trust network is a new evaluation and review of traditional boundary security architecture. Zero trust emphasizes never always trusting anything and verifying things continuously. Zero trust network emphasizes that the identity is not identified by location， all access controls strictly execute minimum permissions， and all access processes are tracked in real time and evaluated dynamically. Firstly， the basic definition of zero trust network was given， the main problems of traditional perimeter security were pointed out， and the zero trust network model was described. Secondly， the key technologies of zero trust network， such as Software Defined Perimeter （SDP）， identity and access management， micro segmentation and Automated Configuration Management System （ACMS）， were analyzed. Finally， zero trust network was summarized and its future development was prospected.

Short-term traffic flow prediction is not only related to historical data， but also affected by the traffic of adjacent areas. Since the trend and spatial correlation of traffic flow are ignored by traditional Time Series Decomposition （TSD） models， a time series processing model based on the combination of Time Series Decomposition and Spatio-Temporal features （TSD-ST） was proposed. Firstly， the trend component and periodic component were obtained by using Empirical Mode Decomposition （EMD） and Discrete Fourier Transform （DFT）， the Spatio-Temporal （ST） correlation of the fluctuation component was mined by Mutual Information algorithm （MI）， and the state vector was reconstructed on the basis of the above. Then， the fluctuation component was predicted by using the state vector through Long Short-Term Memory （LSTM） network. Finally， the final predicted value was obtained by reconstructing the prediction results of the three parts of the sequence. The validity of the model was verified on the real data of Interstate I090 in Washington State， USA. Experimental results show that the Root Mean Square Error （RMSE） of the proposed model TSD-ST-LSTM is reduced by 16.5%， 34.0%， and 36.6% compared with that of Support Vector Regression （SVR）， Gradient Boosting Regression Tree （GBRT） and LSTM， respectively. It can be seen that the proposed model is very effective in improving prediction accuracy.

Bioinformaticly, it is useful to study the change process of biology, such as aging and canceration, by mining differential co-expression bicluster. The definition in the past only measured from the perspective of all set of genes, thus containing a lot of noise. Therefore, a new definition named MiSupport was put forward to measure the difference on gene level, and on the basis of MiSupport, an algorithm named MiCluster was proposed to mine the maximal differential co-expression bicluster in two real gene chips. Firstly, MiCluster constructed a differential weighted undirected sample-sample relational graph in two real-valued gene expression datasets. Secondly, the maximal differential biclusters was produced in the above differential weighted undirected sample-sample relational graph with efficiently pruning techniques and accurately generating candidates method by sample-growth and level-growth. The experimental results show that MiCluster is more efficient than the existing methods. Furthermore, the performance is evaluated by Mean Square Error (MSE) score and Gene Ontology (GO). The results show that this algorithm can find better statistical and biological significance.

Concerning the unbalanced energy and "hotspot" energy hole problems caused by uneven distributed node in Wireless Sensor Network （WSN） clustering routing protocols, an energybalanced uneven distributed node routing protocol was proposed. In the routing protocol, the three parameters, i.e., the "degree" of the node, the distance of nodes to Sink, and the ratio between the average residual energy of clusters nodes and the node itself residual energy, were considered when primary(deputy)cluster head was chosen，and each cluster communicated with Sink through routing tree. The simulation results show that the new protocol can reduce the cluster head rotation frequency when cluster is in "hotspot" or clusters destiny is high, delay the first nodes death time, and make the energy consumption more balanced and the network life cycle longer.

A classifiers ensemble approach based on Principal Component Analysis (PCA) was proposed. Lots of original classifiers were got from Random Subspace Method (RSM). According to their classification performance, their preservation scores were given, so the preferential ranks for classifiers preservation were ordered, by which a set of classifiers was selected from original classifiers. Theoretic analysis and experimental results in face database ORL show that this pattern classification method based on ensemble PCA is efficient for pattern recognition.