Search

Network intrusion detection (NID) is an important method for network system administrators to detect various security holes. The performance of traditional NID methods can be affected when unknown or new attacks are detected. Compared with other machine learning methods, the intrusion detection method based on convolutional neural network (CNN) can significantly improve the accuracy of classification, but the convergence speed and generalization ability of CNN are not ideal in model training process resulting in a low true rate and a high false alarm rate. To solve the above problems, this paper proposes a deep multi-scale convolutional neural network (DMCNN) for network intrusion detection. Different levels of features in a large number of high-dimensional unlabeled original data are extracted by different scales convolution kernel. And the learning rate of network structure is optimized by batch normalization method to obtain the optimal feature representation of the raw data. We use NSL-KDD dataset as the benchmark thus we can compare the performance of our proposed method with other existing works. This dataset includes two testing sets which are the first one is KDDTest+ while the second one is $$\text {KDDTest}^{-21}$$ which is more difficult to be classified. The experimental results reveal that the AC and TPR are higher through our DMCNN model. Especially, in terms of DOS, the AC appropriately reaches to 98%. DMCNN has a high intrusion detection accuracy and a low false alarm rate, which overcomes the limitations of using the traditional detection methods and makes the new approach an attractive one for practical intrusion detection.

Currently, medical image segmentation has attracted more attention from researchers, which can assist in medical diagnosis. However, in the process of traditional medical image segmentation, it is sensitive to the initial contour and noise, which is difficult to deal with the weak edge image, complex iterative process. In this paper, we propose a new medical image segmentation method, which adopts density-oriented BIRCH (balanced iterative reducing and clustering using hierarchies) clustering method to modify active contour model and improve the robustness of noise. The BIRCH is a multi-stage clustering method using clustering feature tree. The improved model can effectively deal with the gray non-uniformity of real medical images. And we also introduce a new energy function in active contour model to make the contour curve approach to the edge, and finally stay at the edge of the image to complete the object segmentation. Experimental results show that this new model can overcome the influence of complex background on medical image segmentation and improve the speed and accuracy of medical segmentation results.

In the transmission of medical images, if the image is not processed, it is very likely to leak data and personal privacy, resulting in unpredictable consequences. Traditional encryption algorithms have limited ability to deal with complex data. The chaotic system is characterized by randomness and ergodicity, which has advantages over traditional encryption algorithms in image encryption processing. A novel V-net convolutional neural network (CNN) based on four-dimensional hyperchaotic system for medical image encryption is presented in this study. Firstly, the plaintext medical images are processed into 4D hyperchaotic sequence images, including image segmentation, chaotic system processing, and pseudorandom sequence generation. Then, V-net CNN is used to train chaotic sequences to eliminate the periodicity of chaotic sequences. Finally, the chaotic sequence image is diffused to change the raw image pixel to realize the encryption processing. Simulation test analysis demonstrates that the proposed algorithm has better effect, robustness, and plaintext sensitivity.

The original BP neural network has some disadvantages, such as slow convergence speed, low precision, which is easy to fall into local minimum value. So this paper proposes an improved particle swarm optimization (PSO) algorithm to optimize BP neural network. In this new algorithm, PSO uses improved adaptive acceleration factor and improved adaptive inertia weight to improve the initial weight value and threshold value of BP neural network. And we give the detailed improved process. At the end, simulation results show that the new algorithm can improve convergence rate and precision of prediction of BP neural network, which reduces the error of prediction. At the end, we use multimedia evaluation model to verify the new method’s performance.

Currently, big data is a new and hot issue. Particularly, the rapid growth of the Internet of Things causes a sharp growth of data. Enormous amounts of networking sensors are continuously collecting and transmitting data to be stored and processed in the cloud, including remote sensing data, environmental data, and geographical data. And region is regarded as the very important object in remote sensing data, which is mainly researched in this article. Region search is a crucial task in remote sensing process, especially for military area and civilian fields. It is difficult to fast search region accurately and achieve generalizability of the regions’ features due to the complex background information, as well as the smaller size. Especially, when processing region search in large-scale remote sensing image, detailed information as the feature can be extracted in inner region. To overcome the above difficulty region search task, we propose an accurate and fast region search in optical remote sensing images under cloud computing environment, which is based on hybrid convolutional neural network. The proposed region search method partitioned into four processes. First, fully convolutional network is adopted to produce all the candidate regions that contain the possible object regions. This process avoids exhaustive search for input images. Then, the features of all candidate regions are extracted by a fast region-based convolutional neural network structure. Third, we design a new difficult sample mining method for the training process. At the end, in order to improve the region search precision, we use an iterative bounding box regression algorithm to normalize the detected bounding boxes, in which the regions contain candidate objects. The proposed algorithm is evaluated on optical remote sensing images acquired from Google Earth. Finally, we conduct the experiments, and the obtained results show that the proposed region search method constantly achieves better results regardless of the type of images tested. Compared with traditional region search methods, such as region-based convolutional neural network and newest feature extraction frameworks, our proposed methods show better robustness with complex context semantic information and backgrounds.

Searchable asymmetric encryption (SAE) can also be called Public Key Encryption with Keyword Search (PEKS), which allows us to search the keyword on the data of having been encrypted. The essence of Asymmetric searchable encryption is that users exchange the data of encryption, one party sends a ciphertext with key encryption, the other party with another key receives the ciphertext. Encryption key is not the same as the decryption key, and cannot deduce another key from any one of the key, thus it greatly enhances the information protection, and can prevent leakage the user's search criteria—Search Pattern. Secure schemes of SAE are practical, sometimes, however the speed of encryption/decryption in Public-key encryption is slower than private key. In order to get higher efficiency and security in information retrieval, in this paper we introduce the concept of distributed SAE, which is useful for security and can enable search operations on encrypted data. Moreover, we give the proof of security.

In this paper, we perfect the mutual coupling of compact microstrip array antenna by designing a new defected ground structure. When the resonant frequency is 2.45GHz, array element spacing is 0.1 times of free space wavelength, we introduce new defected ground structure into antenna array. Then we use HFSS to make simulation and compare the changing of antenna's parameters before and after adding defected ground structure. The results demonstrate that the parameters representing mutual coupling in new model can reduce by 30dB, which effectively perfects the mutual coupling of compact microstrip array antenna.

As we all know, traditional electromagnetism mechanism (EM) algorithm has the disadvantage with low solution precision, lack of mining ability and easily falling into precocity. This paper proposes a new chaos electromagnetism mechanism algorithm combining chaotic mapping with limited storage Quasi-Newton Method (EM-CMLSQN). Its main idea is that it adopts limit quasi-Newton operator to replace the local optimization operator in EM algorithm for local searching in the late of algorithm. In the process of algorithm, the chaos mapping is introduced into optimization processes, and it generates new individuals to jump out of local to maintain the population diversity according to characteristics of chaos mapping random traversal. Finally, the experiments show that the new algorithm can effectively jump out of local optimal solution through comparing three continuous space test functions. The new algorithm has obvious advantages in terms of convergence speed compared to traditional EM algorithm, in addition, it is more accuracy than particle swarm optimization (PSO) algorithm. We compare the new chaos electromagnetism mechanism algorithm with ant colony optimization (ACO) algorithm, PSO algorithm, the results represent that new scheme can obtain the optimal path in the path optimization process, which shows that the new method has better applicability in the discrete domain problem.

Maneuvering target tracking is a target motion estimation problem, which can describe the irregular target maneuvering motion. It has been widely used in the field of military and civilian applications. In the maneuvering target tracking, the performance of Kalman filter(KF) and its improved algorithms depend on the accuracy of process noise statistical properties. If there exists deviation between process noise model and the actual process, it will generate the phenomenon of estimation error increasing. Unbiased finite impulse response(UFIR) filter does not need priori knowledge of noise statistical properties in the filtering process. The existing UFIR filters have the problem that generalized noise power gain(GNPG) does not change with measurement of innovation. We propose an improved UFIR filter based on measurement of innovation with ratio dynamic adaptive adjustment at adjacent time. It perfects the maneuvering detect-ability. The simulation results show that the improved UFIR filter has the best filtering effect than KF when process noise is not accurate.

In this paper, we express arc furnace electrode regulator system as a time-variant nonlinear-linear-nonlinear model. On this basis, we propose an online identification method based on nonlinear-linear-nonlinear model system. This new scheme solves the problem of model variation and prediction precision decline causing by time-varying of arc characteristic. In order to dispose the difficulty of parameters separation in the online identification process, this new method adopts the mind of update the parameters of linear parts and nonlinear parts respectively. It realizes the parameters separation of system effectively. Simulation results show that this method can track the changes of arc characteristics effectively. That it achieves the aim of real-time monitoring and controlling system parameters.

As we all know, the parameter optimization of Mamdani model has a defect of easily falling into local optimum. To solve this problem, we propose a new algorithm by constructing Mamdani Fuzzy neural networks. This new scheme uses fuzzy clustering based on particle swarm optimization(PSO) algorithm to determine initial parameter of Mamdani Fuzzy neural networks. Then it adopts PSO algorithm to optimize model's parameters. At the end, we use gradient descent method to make a further optimization for parameters. Therefore, we can realize the automatic adjustment, modification and perfection under the fuzzy rule. The experimental results show that the new algorithm improves the approximation ability of Mamdani Fuzzy neural networks.