After the introduction of federated learning technology in intrusion detection scenarios， there is a problem that the traffic data between nodes is non-independent and identically distributed （non-iid）， which makes it difficult for models to aggregate and obtain a high recognition rate. To solve this problem， an efficient federated learning algorithm named H?E?Fed was constructed， and a network intrusion detection model based on this algorithm was proposed. Firstly， a global model for traffic data was designed by the coordinator and was sent to the intrusion detection nodes for model training. Then， by the coordinator， the local models were collected and the skewness of the covariance matrix of the local models between nodes was evaluated， so as to measure the correlation of models between nodes， thereby reassigning model aggregation parameters and generating a new global model. Finally， multiple rounds of interactions between the coordinator and the nodes were carried out until the global model converged. Experimental results show that compared with the models based on FedAvg （Federated Averaging） algorithm and FedProx algorithm， under data non-iid phenomenon between nodes， the proposed model has the communication consumption relatively low. And on KDDCup99 dataset and CICIDS2017 dataset， compared with baseline models， the proposed model has the accuracy improved by 10.39%， 8.14% and 4.40%， 5.98% respectively.

In order to improve the efficiency of dam safety operation and maintenance, the dam defect object detection models can help to assist inspectors in defect detection. There is variability of the geometric shapes of dam defects, and the Single Shot MultiBox Detector (SSD) model using traditional convolution methods for feature extraction cannot adapt to the geometric transformation of defects. Focusing on the above problem, a DeFormable convolution Single Shot multi-box Detector (DFSSD) was proposed. Firstly, in the backbone network of the original SSD:Visual Geometry Group (VGG16), the standard convolution was replaced by the deformable convolution, which was used to deal with the geometric transformation of defects, and the model's spatial information modeling ability was increased by learning the convolution offset. Secondly, according to the sizes of different features, the ratio of the prior bounding box was improved to prompt the detection accuracy of the model to the bar feature and the model's generalization ability. Finally, in order to solve the problem of unbalanced positive and negative samples in the training set, an improved Non-Maximum Suppression (NMS) algorithm was adopted to optimize the learning effect. Experimental results show that the average detection accuracy of DFSSD is improved by 5.98% compared to the benchmark model SSD on dam defect images. By comparing with Faster Region-based Convolutional Neural Network (Faster R-CNN) and SSD models, it can be seen that DFSSD model has a better effect in improving the detection accuracy of dam defect objects.

Glioma segmentation depends on multi-modal Magnetic Resonance Imaging (MRI) images. Convolutional Neural Network (CNN)-based segmentation algorithms are often trained and tested on fixed multi-modal images, which ignores the problem of missing or increasing of modal images. To solve this problem, a method mapping images of different modalities to the same feature space by CNN and using the features in the same feature space to segment tumors was proposed. Firstly, the features of different modalities were extracted through the same deep CNN. Then, the features of different modal images were concatenated, and passed through the fully connected layer to realize the feature fusion. Finally, the fused features were used to segment the brain tumor. The proposed model was trained and tested on the BRATS2015 dataset, and verified with the Dice coefficient. The experimental results show that, the proposed model can effectively alleviate the problem of data missing. At the same time, compared with multi-modal joint method, this model is more flexible, and can deal with the problem of modal data increasing.

In order to solve the problem of human aggressive behavior recognition, an aggressive behavior recognition method based on human joint points was proposed. Firstly, OpenPose was used to obtain the human joint point data of a single frame image, and nearest neighbor frame feature weighting method and piecewise polynomial regression were used to realize the completion of missing values caused by body self-occlusion and environmental factors. Then, the dynamic "safe distance" threshold was defined for each human body. If the true distance between the two people was less than the threshold, the behavior feature vector was constructed, including the human barycenter displacement between frames, the angular velocity of human joint rotation and the minimum attack distance during interaction. Finally, the improved LightGBM (Light Gradient Boosting Machine) algorithm, namely w-LightGBM (weight LightGBM), was used to realize the classification and recognition of aggressive behaviors. The public dataset UT-interaction was used to verify the proposed method, and the accuracy reached 95.45%. The results show that this method can effectively identify the aggressive behaviors from various angles.

How to ensure the survivability of static multicast business has become a widespread concern in the multi-domain optical network of pre-configured multicast business. Concerning the above problem, by adopting the global topological information and scheduling calculation model based on hierarchical Path Computation Element (PCE) architecture, a bimatrix game model was used to generate link-disjoint multicast trees and multicast protected trees, finally hierarchical PCE-based and bimatrix game-based multicast dedicated protection algorithm was put forward under static state, and concrete examples of the algorithm were given. Theoretical analysis and experimental results show that under certain redundancy allocation of network resources, the proposed algorithm has low time complexity, and it can obviously improve the multicast business survivability in multi-domain optical network under static state, with optimizing resources allocation structure of protection work in the optimal multicast working trees and multicast protected trees at the same time.

To solve the problems of small fault depth and complex manual deduction in previous fault attacks on SIMON, an Algebraic Fault Attack (AFA) method was proposed. Firstly, Correct equations of full-round SIMON encryption was established based on the algebraic representation of SIMON core operation ‘&’. Then faults were injected into the internal states and two models were provided for fault representation based on whether attackers knew the exact fault information or not. Finally, a CryptoMinisat-2.9.6 solver was used for round-keys recovery. The simulation results show that the fault-known and fault-unknown model need 5 and 6 faults to recover the entire key set with single-bit faults injected in the 26th round of SIMON32/64. As for SIMON128/128, two models both need only 2 faults to recover the entire key set with n-bit length faults injected in the 65th round. Moreover, it can be found that the influencing factor of average solving time will change from fault information to computation with fault number growing.

Aiming at the uncertainty of Single-Input-Single-Output (SISO) nonlinear systems, a novel adaptive control design based on extended Back Propagation (BP) neural network was proposed. Firstly, the weight vectors of BP neural network were trained via the offline data. Then, the scaling factor and estimation parameter of approximate accuracy were adjusted online to control the whole system by update law. In the design process of controller, with the Lyapunov stability analysis, the adaptive control scheme was proposed to guarantee that all the states of the closed-loop system were Uniformly Ultimately Bounded (UUB). Compared with the traditional adaptive control method of BP neural network, the proposed method can effectively decrease the parameter number of online adjustment and reduce the burden of computation. The simulation results show that the proposed method can make all the states of the closed-loop system tend to be zero, which means the system reaches the steady state.

Aiming at the problems of Artificial Bee Colony (ABC) algorithm, which are the single search mechanism and the high coupling between global search and local search, a Multi-Population ABC (MPABC) algorithm based on hybrid search was proposed. Firstly, the population was sorted according to the fitness value to get an ordered queue, which was divided into three sorted subgroups including random subgroup, core subgroup and balanced subgroup. Secondly, different difference vectors were constructed according to the corresponding individual selection mechanism and search strategy to different subgroups. Finally, in the process of group search, the effective control of individuals with different fitness functions was realized through three subgroups, thus improving the balance ability of global search and local search. The simulation results based on 16 benchmark functions show that compared with ABC algorithm with Variable Search Strategy (ABCVSS), Modified ABC algorithm based on selection probability (MABC), Particle Swarm-inspired Multi-Elitist ABC (PS-MEABC) algorithm, Multi-Search Strategy of the ABC (MSSABC) and Improved ABC algorithm for optimizing high-dimensional complex functions (IABC), MPABC achieves a better optimization effect; and on the solution of high dimensional (100 dimensions) problems, compared with ABC, MPABC has higher convergence speed which is increased by about 23% and better search accuracy.

Since the efficiency of data forwarding in Vehicular Ad Hoc Network (VANET) is low, a data forwarding mechanism in VANET based on Software-Defined Network (SDN) was proposed. Firstly, a hierarchical architecture of SDN based VANET was designed. This architecture was consist of local controller and vehicular, it could implement the separation of control and data forwarding, and also could achieve high scalability, reliability and efficiency. Secondly, a new data forwarding mechanism was proposed, which used dynamic programming and binary search. Finally, compared with Ad Hoc On-demand Distance Vector routing (AODV), Destination Sequenced Distance Vector routing (DSDV), Dynamic Source Routing (DSR) and Optimized Link State Routing (OLSR) algorithm, the proposed algorithm could improve packet delivery fraction and end-to-end delay. Therein, the average increase of packet delivery fraction was about 100%, while the average reduction of end-to-end delay was about 20%. The simulation results show that the data forwarding mechanism in software-defined VANET can effectively improve the packet delivery and reduce the end-to-end delay.

Concerning the problem that the existing Voice Activity Detection (VAD) algorithms based on Hidden Markov Model (HMM) were poor to track noise, a method using Baum-Welch algorithm was proposed to train the noise with different characteristics, and the corresponding noise model was generated to establish a library. When voice activity was detected, depending on the measured background noise of the speech, the voice was dynamically matched to a noise model in the library. Meanwhile, in order to meet real-time requirements of speech signal processing, reduce the complexity of the speech parameter extraction, the threshold was improved to ensure the inter-frame correlation of the speech signal. Under different noise environments, the improved algorithm performance was tested and compared with Adaptive Multi-Rate (AMR), G.729B of the International Telecommunications Union (ITU-T). The test results show that the improved algorithm can effectively improve the accuracy of detection and noise tracking ability in real-time voice signal processing.

When the network abnormal event occurs, the spatial-temporal correlation of the sensor nodes is very obvious. While existing methods generally separate time and space data properties, a decentralized algorithm of spatial-temporal abnormal detection based on Probabilistic Graphical Model (PGM) was proposed. Firstly the Connected Dominating Set (CDS) algorithm was used to select part of the sensor nodes to avoid monitoring all the sensor nodes, and then Markov Chain (MC) was used to predict time exception event, at last Bayesian Network (BN) was utilized in modelling the spatial dependency of sensors, combining spatio-temporal events to predict whether the abnormal events would or would not occur. Compared with the simple threshold algorithm and BN algorithm, the experimental results demonstrate that the proposed algorithm has higher detection precision, and low delay rate, greatly reducing the communication overhead and improving the response speed.

A general construction method of optimal frequency hopping sequence sets with no-hit zone was proposed in this paper, which included several known constructions as special cases. The general method was obtained by performing the column permutation on the signal matrix. In the proposed construction, the length, the number of the sequences and the length of the no-hit zone could be changed flexibly. Furthermore, the available concrete construction methods were abundant. Some properties of the frequency hopping sequence sets were influenced by the concrete construction methods and the parameters. The parameters of frequency hopping sequence sets obtained by this method reach the theoretical bound; hence they are classes of optimal frequency hopping sequence sets with no-hit zone.

A real-valued extended propagator method algorithm based on virtual interpolated array named VIA-EPM real-valued algorithm was proposed to solve the utilization of Virtual Interpolated Array (VIA) in the non-circular Direction Of Arrival (DOA) algorithm. The real array output was virtually transformed by utilizing transformation matrix which was obtained through the real array manifold and virtual array manifold. The real part and imaginary part of the transformed array output were obtained, which could be reconstructed in series to extend dimensions according to the characteristic that the signal sources are real-valued, and then a Propagator Method (PM) DOA estimation algorithm was obtained after splitting the extended array output matrix. The simulation results show that, if sensor position errors exist, the performance of the new algorithm is similar to the calibrated extended propagator method real-valued algorithm (EPM real-valued algorithm) by using VIA for the calibrated sensor position data, and this algorithm also keeps the performance in array extension, high accuracy and high resolution, and the performance of the new algorithm is obviously better than the non-calibrated EPM real-valued algorithm in the case of two-dimension sensor position errors. Analysis of the computational complexity of VIA-EPM real-valued algorithm concludes that the new algorithm has the advantage of virtual interpolated array and non-circular characteristic, and its computational complexity is much lower than complex-valued algorithm.

The error concealment is an important technique in the video transmission, and it can ensure the reconstruction video quality and efficiently recover the data loss and the data errors in the transmission process caused by severe transmission environments. In order to enhance the error resilience of AVS-P2, the paper proposed a new adaptive temporal-spatial error concealment method based on the redundancy motion vectors. To conceal a lost block, the paper used the spatial error concealment for the I-frame macroblocks, and used the temporal error concealment for the non-I-frame macroblocks. At the same time, according to the motion intensity of the macroblocks, it used the default error concealment of AVS-P2 and error concealment method based on redundancy motion vectors, respectively. Lastly, the proposed algorithm was realized based on the platform of the AVS-P2 RM52_20080721. The simulation results show that the proposed method is significantly better than the existing techniques in terms of both objective and subjective quality of reconstruction video.

The Ethnic Group Evolution Algorithm (EGEA) has used ethnic group mechanism, a kind of population-structured technology, to control the evolution tendency of population; meanwhile, it has used the binary code similarity among individuals to be the ethnic group clustering criterion. Because the hamming cliff problem of nature binary code was likely to affect the accuracy of ethnic group clustering, we proposed to make use of gray code to improve the evolution efficiency of EGEA. The simulations of numerical optimization show the EGEA based on gray code can improve the searching speed and the solution precision greatly.

In FIRA MiroSot robot-soccer game,vision system was a unique way throught which the whole system obtained the global information. The speed and precision of the recognition of the vision system directly affected the victory or defeat of the game. According to the disadvantage of the traditional vision system in the robot-soccer game, a new design of robot-soccer vision system was put forward based on multi-resolution analyse and FCM algorithm. Experiment results show that the design can improve the speed and precise of recognition in the game,and has well adaptability.