针对高校计算机网络课程思政的教学实际,分析新工科背景下课程思政要解决的3个关键问题,提出"理论—方法—实践"3层建设框架,介绍计算机网络课程思政建设的具体实践方法,最后利用问卷调查法,通过最近学期的实际教学数据汇总,基于主观评价方法对课程思政的实践效果进行分析与评价.

Chinese text recognition has been one of the hot topics. Although there are many text recognition methods available, some problems are still not effectively solved, such as the appearance of Chinese characters that do not appear in the training process. Previous Chinese text recognition methods have not thoroughly solved this problem, and many of them are based on English text recognition methods. English words can be split into different character, and Chinese character can be similarly split Chinese characters into multiple stroke sequences, then recognize the stroke sequences, and finally connect the stroke sequences into sentences. Howerver, the same sequence of strokes can form different Chinese characters, so a sequence of strokes can also form different sentences. To solve this confusion problem, a text classification model is trained to rank all possible sentences and then find the one that best matches the normal semantics.

This paper introduces an aviation surveillance information fusion method based on multi neural networks. With the rapid growth of the number of civil and military aircraft, the air traffic is more and more busy. In order to ensure the flight safety of the aircraft, the application of the aviation surveillance information processing system is indispensable. The aviation surveillance information fusion technology in the system is the key to obtain the accurate information of the flight target position. As the traditional method of aviation surveillance information fusion method, the Kalman filter has the characteristics that it does not need to retain the past measurement data, but only needs to be recursive according to the state equation. But the Kalman filter has the problem of high requirements for model error and calculation accuracy, and it requires a large number of professionals to conduct time-consuming parameter adjustment, which consumes a lot of human and material resources. Thus, it is necessary to find an efficient and accurate information fusion method of aviation surveillance. The multi neural network proposed in this paper can overcome the above shortcomings of the Kalman filter. Furthermore, the multi neural network can achieve better than single neural network, which can predict the position of the aircraft more accurately.

At present, with the comprehensive development of economy and technology in our country, the air transportation industry has also ushered in a golden period of development. The air traffic volume increases year by year, and the air route traffic volume will increase, which will lead to the need to carry more aircraft on the limited channel, resulting in the congestion of the channel and the potential safety problems. In order to ensure that the aircraft can fly safely, it is necessary for traffic management personnel to maintain the order of the aircraft in the channel according to the aviation monitoring information. Therefore, the accuracy of aviation surveillance information is particularly important. As a traditional track fusion algorithm, the Kalman filtering has the problem of requiring accurate error estimation, insensitivity to noise, and long calculation time in the case of large data volume. In this paper, a method of air surveillance information fusion based on ensemble learning is proposed, which can predict and fuse multiple air surveillance sources, i.e. multiple radar surveillance information, so as to obtain more accurate position estimation of the monitored target and reduce the error caused by the accuracy of radar itself, geographical location and surrounding environment.

Since the measurement error of surveillance sensors such as radar differs each other in the detection of the same target, it's necessary to fuse the multi-source radar data to estimate the true location of target and reduce the measurement error of radar. The key is to establish nonlinear regression model since the uncertainty of measurement error. In this paper, the Support Vector Regression(SVR) methodology was adopted to estimate the true location of target based upon the measurement results of multi-source radar. We uniquely identify a region by a sequence of radar id which means a target can be detected in this area by radars with id listed in the sequence. Different regression model was established in different region which are independent of each other. Since the coordinate system used by radar data and ADSB data is different, we mapped all the data into the same two-dimensional Cartesian coordinate system. In the same region, two regression models were established to estimate the values of aircraft on the x-axis and the y-axis. After we predict the x and y coordinates of the target, we convert the coordinates back to the WGS84 format.

This paper introduces a multi-radar track fusion method based on random forest regression and provides an accurate and stable fusion track. The increasing number of aircraft will lead to congested routes, further leading to safety issues. Therefore, an effective track fusion method can accurately locate the aircraft, thereby ensuring the safety of the aircraft in the case of crowded routes. The basic idea of the method proposed in this paper is to select the radar data of a certain track of a certain day to train the model, and predict the position of the aircraft on the next day of the track through the trained model. As a traditional track fusion algorithm, the Kalman filtering has the problem of requiring accurate error estimation, insensitivity to noise, and long calculation time in the case of large data volume. The neural network method that compensates for these shortcomings also has the disadvantage of poor generalization ability in the case of a large amount of noise. The random forest regression model proposed in this paper can overcome the shortcomings of over-fitting in neural network, so it can achieve better prediction results. And through the real data test, the average error is 40m, compared with the neural network method, the result is increased by 50%.

Aviation surveillance information fusion is aimed at merging the detection data from multiple sources of the same target aircraft to obtain more accurate monitoring information, including aircraft position, heading, acceleration and other information.The traditional Kalman filter-based fusion technology has shortcomings, such as poor integration in the maneuvering state, and it takes a lot of manpower and material resources to repeat the adjustment.Therefore, this paper uses the recurrent neural network to conduct the experiment of aviation surveillance information fusion.Firstly, the recurrent neural network is used to identify the maneuver state of the aircraft, and the weighted least squares method is used to predict the position of the aircraft according to the maneuvering state, so as to obtain the monitoring information of each radar at the same time.After that, the recurrent neural network model is used to fuse the monitoring information of multiple radars.The experimental results show that the maneuvering state discriminant model based on recurrent neural network can effectively identify the maneuvering state.The least square method based on maneuvering state can accurately predict the position of the aircraft.The aeronautical surveillance information fusion model based on recurrent neural network can also obtain more accurate fusion results.The whole process includes four parts: preprocessing, maneuver status discrimination, position prediction and information fusion.The total time is about 500ms.

On the basis of Time Domain Analysis, this paper proposes a method for music beat tracking. Through this method, music beats are detected and tracked by timestamp and intensity value. Generally, the beat-areas of music signal converge more energy than other areas, therefore, the spots of beat can be filtered out by a tracking algorithm with a dynamic threshold value. In this paper, dynamic threshold value in tracking algorithm is modified by using two sliding windows, which are Prediction Window and Detection Window. Also, a new indicator which indicates the stationarity of the signal is proposed. This factor can distinguishes the music signal with rhythmic beats from which with lone-tone and noise in time-domain. The experimental result proves the simplicity, adaptability, and robustness of this method, and it is an efficient algorithm on music beat tracking.

Based on the theories of frequency domain and time domain signal processing, wavelet analysis, and singular value decomposition (SVD), an effective method for content based music feature extraction is proposed in this paper. Music feature can be divided into three parts by this method, which are frequency feature, auditory perceptual feature, and statistical characteristic of beat. The characteristic of each music can be well described by these features. The results of logistic regression classification model and linear support vector machine (SVM) classification model which is on a data set consists of several different styles of music and use the feature extraction method in this paper show the high precision of 95.33% in average, and also prove the effectiveness of the proposed method. Feature extraction is the foundation of content based recommendation, retrieval, classification, and cluster. Hence this method has good prospect in these area.

Cyberpedia is an online encyclopedia with numerous data that covers many areas.Cyberpedia search ser-vice is constructed by SolrCloud which has features of central configuration,automatic failover,real-time search and automatic load balancing.This paper introduces the constructure of this SolrCloud project,the design of the index structure,and the addition of the Chinese tokenizer which improves the result of word segments.Moreover,this paper discusses the search engine optimization.The key fields are segmented by a few patterns of different granularities when indexing and retrieving.A document is scored by the times it is cited,and the score effects the rank of the document.The experiment shows that the optimization takes good effect of the Cyberpedia search service.

Recommemder System is becoming more and more important for getting information in recent 20 years. But recommender system has the weakness of extreed large scale that makes it delayable for recommendation, which making it cannot offer real-time service. Business recommender system is general divided into two parts, the on-line recommend part and the off-line calculation part. It precomputes the off-line part to get quicker recommendation when needed. Pretended real-time recommendation is a compromise with the growing and changing system. We propose the better way to get better real-time service by processing the off-line calculation on GPU , which is a high-speed parallel processor , to speed up the first part of recommender system to get more real-time service. Our experiments show, the off-line part can speed up 19 times when using GPU, and the larger of the data scale, the better it can improve.

LDA (Latent Dirichlet Allocation) is a text modeling algorithm based on a generative probabilistic model. It is widely used to discover latent topics among a set of documents. Mahout has implemented LDA algorithm, however, the execution time of the LDA program is very long when processing a large amount of documents, because the documents are processed in sequence. This paper introduces a method to modify this program with CUDA toolkit provided by NVIDIA, in order that a group of documents could be processed in parallel on GPU. Using this method, the LDA program could be accelerated greatly.

Document clustering is one of the most important tasks in text mining. In clustering algorithms, high-dimensional vector is usually used to represent a document which causes that the algorithms are often computationally expensive. On the other hand, Graphic Processing Unit (GPU) is increasingly important in parallel computing due to its powerful parallel capacity and high bandwidth. This paper implements a GPU-based Harmony K-means Algorithm (HKA) with NVIDIA's Compute Unified Device Architecture (CUDA), and uses it for document clustering. In our experiment, our GPU-based program can acquire a maximum 20 times speedup in contrast with CPU-based program.

As the expansion of Internet, the recommender system is attracting the attention of many industry engineers and researcher, especially the collaborating filtering recommender system. However, there are still some challenges. For example, the sparse feature and large scale system degrades the recommendation accuracy and efficiency. In this paper, we propose implied-similarity and filled-default-value methods to improve the denseness of the preference matrix and use GPU to parallel the process. Our experiments show that the accuracy can improve 20% and efficiency can speed up 4 times.

Generating shorter testing sequence is an important issue in protocol conformance test.This paper proposes a method for generating BUIO sequence by using UIO sequence.Some UIO sequences switching to BUIO sequences can bring the cost reduce of the BUIO generation.The heuristic sequences generation algorithom based on UIO and BUIO is analyzed and improved to ensure the automatic generation of test sequence.Test sequence is obtained through applying the algorithm to protocol ECMA-203,which is compared to the test sequence generated by using Rural Chinese Postman algorithm and UIO sequences.

The invention provides a route test method of an aeronautical telecommunication network and a router virtual machine. The router virtual machine comprises a virtual machine platform module, a virtual machine protocol module and a network topology structure storage module, wherein the virtual machine protocol module is connected with the virtual machine platform module; the virtual machine protocol module is used for providing a route protocol to the virtual machine platform module; the network topology structure storage module is used for storing network topology structure information and virtual machine configuration and description information; and the virtual machine platform module is used for setting the position and the configuration parameter of the virtual machine in the network according to the network topology structure information and the virtual machine configuration and description information stored in the network topology structure storage module when the virtual machine is initialized and is also used for obtaining the route protocol in the virtual machine protocol module and testing. The invention is convenient to establish the network topology structure module, atester can conveniently establish the network topology module needed by the test, and a channel error rate is set in the module to simulate an authentic situation.

ATN is the next generation of aeronautical communication network. An important issue concerning ATN is the mobile routing problem. Generally, a kind of mobile routing mechanism similar to mobile IP can be used to solve this problem. But there are some defects with it when distributing routing information and carrying Out handoff. Using the predictability of the path information of an aircraft in ATN, this paper suggests a new mobile routing mechanism to solve the problems in traditional one. By the simulation experiments we make comparisons on delay, delay jitter, packet loss and throughput between the two mechanisms. The numerical results prove that the new mechanism solves the problems basically and achieves expectant effect.

为配合计算机网络课程的教学,我们设计了基于UDP隧道技术的数据链路层实验.该实验改进了Andrew S.Tanenbaum数据链路层实验的仿真方法,使得实验开展模式更灵活、多样化.实验内容增加了建链、拆链和状态转移过程,更全面地体现了数据链路协议的完整过程.

路由算法和协议在路由器中的地位至关重要,文章对路由算法IS-IS及路由算法的效率进行研究和改进,提高了路由器的处理能力和稳定性.

随着IP应用对组播可靠性的要求不断提高,国际上很多大学和研究机构在一定的应用背景下提出了可靠组播通信,并开发了各自的解决方案.本文系统地介绍了常见的实现方法并进行分析.