Abstract:In recent years, more and more large-scale English tests begin to use the automatic scoring system.Therefore, the research of this system is of great value.In this paper, we first extract a lot of features according to English writing guide.Then we use autoencoder and discretization algorithm to learn a different representation of features.Finally, we use a hierarchical multinomial model to output the final scores of articles.Experimental results indicate that this method not only achieves great performance for those essays of the same topic, but also shows good robustness when predicts essays of different topics.Compared with the traditional automatic score method, our approach achieves higher than 9.7% in term of Pearson Correlation Coefficient, with good practical values.

Abstract:The investigation of node similarity is an important component in link prediction and community detection.In this paper, four kinds of algorithms including common neighbor (CN), resource allocation (RA), Adamic-Adar (AA) and Sorenson are introduced into various kinds of real networks and two kinds of simulation networks comprised of small world network and scale free network.The Area Under the Curve (AUC) is computed to compare their predictive accuracy.It's found that RA performs much better than the other three kinds of algorithms.Then four algorithms are adopted in functional connectivity networks that characterize electroencephalograph (EEG) recordings from eight patients with generalized epilepsy.It's demonstrated that RA performs best from the point of prediction accuracy.According to RA technique, clusters could be determined from nodes that own maximum similarity which provides an objective index for quantifying brain condition, and this might be applied for clinical auxiliary diagnosis in the future.

Abstract:Imaging in the atmosphere presents the phenomenon of low contrast, low saturation and hue offset due to atmospheric particles such as haze and fog.In this paper, a fast method is proposed to remove haze from a single image based on modified bilateral filtering.The similar Gaussian kernel that is firstly proposed in this paper is applied to structure the modified bilateral filtering.The modified bilateral filtering is a linear edge preserving filter and is used to refine medium transmission;meanwhile, it can effectively enhance the speed of data processing.The dark channel prior and fog image are used to accurately estimate the value of global atmospheric light by an interval through weighted fusion.Compared to some state-of-the-art methods, the proposed method can achieve a faster processing speed, effectively improve the visibility and contrast of the restored images, and obtain good color effect.

Abstract:In recent years, the rapid development of social networks has significantly changed the styles of people's work and life.The way people acquiring knowledge shows a clear trend of the network.When people acquire information through the Internet, it also leaves personal traces.Taking the high cost of obtaining personal information in reality into account, it's a good idea to catch people's leaving traces in the network and study its "mapping" in the network society.User portrait as a virtual representative of real users is based on a series of real data on the user model.Based on dynamic modeling of user portrait, the user is defined as the basic attributes, social attributes, interests, and abilities by digging the "ZhiHu" site.Then we make an empirical analysis of the 1303 users who follow the topic of PM 2.5.

Abstract:Numerous modules or third-party data access and multifarious upper application require system data transmission with flexible, universal and extensible performance.Therefore, a middle layer data transmission design becomes a key task of developers.Add front layer in data access, which uses the structure of communication link layer, protocol processing layer, management layer, and front real-time data processing library.Considering the clear principle of cross-platform and data type distinction, the message bus data transmission with ICE middleware and JSON string as the core is constructed and the interaction interface between upper application and front module is designed.The bus transfer mechanism separates the application software layer from the pre-layer, and implements the information exchange through the bus according to the agreed message types.It shields the system's various upper-layer application requirements and the pre-developed communication protocol characteristics, and improves the developer's work efficiency and the system's robustness, scalability, and also reduces the follow-up maintenance personnel to maintain the difficulty.

Abstract:The switchgear is one of the most critical and complex equipment in a power system, and plays an important role in ensuring the safe and reliable operation of the grid system.In order to better assess the health status of the switch cabinet, according to expert experience and the actual situation of monitoring system, this paper establishes a fuzzy comprehensive evaluation model based on a dynamic weighted mechanism.The evaluation model establishes a comprehensive evaluation index system and a membership degree distribution function based on non-electric characteristic parameters, combined with fuzzy comprehensive evaluation method and the dynamic weight correction model, considering the different effects of each index of health condition of the switchgear, and emphasizing the abnormal state of the equipment.In addition, the simulation analysis is carried out by using the evaluation model.The experimental results show that the evaluation model is effective and reasonable, and conforms to the actual standard and demand of the electric power industry.

Abstract:By analyzing and summarizing the existing test methods for isolation of cloud IaaS, a large-scale distributed testing system based on message middleware is designed and implemented for validating the isolation of a cloud IaaS.The system is a message-based distributed architecture;the control nodes and the testing nodes are separated, which helps to reduce the coupling of the system and enhance scalability.By testing the isolation of cloud IaaS in operation, the feasibility of the testing system is verified on the OpenStack platform.The testing system is suitable for testing other security abilities of cloud services.

Abstract:A controller of virtual studio three-dimensional real-time rendering engine is proposed in this paper, and the research focuses on the architecture, communication method and control protocol.The virtual studio three-dimensional real-time rendering engine is the core technology of virtual studio.It is different with the general three-dimensional rendering engines in the rendering function and especially in the design of controller for virtual studio application.The architecture design of the controller meets the demands of virtual studio, which is a control command transfer from the controller to the rendering module.The communication method is designed to be a protocol driven TCP communication method.It not only ensures the real-time feature of rendering, but also guarantees the scalability of control protocol, which will facilitate the follow-up research and the application development.

Abstract:There is a need for procedure reorganization and business collaboration in cement enterprises.Aiming at this demand, the current operating circumstances and informatization status are analyzed in this paper.Based on the analysis, we apply service oriented methods in the integration for the cement enterprise application with comprehensive utilization of the advantage of hybrid cloud.A service partition algorithm for cement enterprise is designed.In addition, mobile services and equipment diagnosing services are illustrated in details.We build a cement enterprise platform for application integration based on the architecture which makes full use of the cloud and enterprise resources.Integration and sharing of the applications are realized.The remote access to the platform is achieved.The platform satisfies enterprise scientific management, process optimization, and business collaboration needs.

Abstract:There are some defects such as inconvenience and maintenance trouble for network management in the traditional C/S (client/server) model of network management system.Therefore, we put forward a network management system that based on B/S (browser/server) model, in which the original C/S model of network management system is turned to B/S model, and the structure of system is optimized which helped the customers to conveniently and reliably monitor and manage the network resource.This system is based on the interface that the original network management system server exposed, uses the RMI (remote method invocation) technology call server interface to realize data communication with Web server, and the whole system adopts the load balancing strategy to balance the access requests from multiple browser to each Web server for processing.Meanwhile, the system adopts the hot standby process on the single load balancer, thus greatly improves the overall performance and reliability of the network management system.

Abstract:Oriented at the actual application demands of the remote control with the information security equipment, this paper proposes an information security equipment structure supporting remote control, and presents the workflow and functional structure of the equipment module.The structure is compatible with the early equipment effectively, and has characteristics of credible to ensure the safe operation of the equipment.The structure adopts module design.Each functional module can be configured flexibly according to application requirements.At the same time, the equipment supports destroying the sensitive information automatically in emergency cases, which can effectively improve the level of safety management of equipment.

Abstract:In this paper, based on the KETTLE which is an open source data analysis tool, an ETL process model is constructed to extract, transform and load the existing basic big data of the cigarette sales.Through the ETL model, the basic data is converted to analytical data, and the data integration can be implemented quickly and efficiently.In this paper, based on the integrated data, combining with the platform construction and visualization of GIS, the visualization and analysis system of the cigarette sales data is designed and implemented to provide data display and analysis platform for the enterprise data analysis and policy support.

Abstract:A wireless retinal imaging system is designed and implemented, which can dissolve the disadvantages in traditional ophthalmoscope and fundus camera eye inspection equipment such as the complexity of structure, poor mobility, and the disability to real-time observing and other defects.The retinal image acquisition and WiFi wireless local area network technology are studied.In this system, the WiFi wireless transmission technology, HTTP network programming, and the remote control of retinal images with the PC machine equipped with a network adapter and the image, are transmitted, displayed and stored in real time.The PC machine with the network card is connected to the fundus camera through WiFi, which is convenient to realize the wireless collection and observation of the retinal images and provides the objective basis for the diagnosis and treatment of diseases in the department of ophthalmology.

Abstract:After analyzing the shortages of current wireless self-organized network, this paper presents a radio self-organized network scheme based on cross-layer design, particularly describes the scheme contents of MAC and Network layers.Then, the semi-physical simulation environment for prototype aeroplane is developed.The simulation results show that this scheme can obviously improve the ability of network throughput comparing with the traditional self-organized network mode.

Abstract:IT systems' structural relationship visualization system is a web system, which utilizes three-dimensional visualization technology to build the business to IT visual management model and presents the relationship among the system's business, systems, processes and physical devices supported by three-dimensional, simulation and real-time interaction[1].It can visually show business relationship in different levels of application nodes, owner, relationship between the application and the relevant components, correspondence relationship between the logical and physical components of the entity, flow relationship in business applications.Moreover, it can achieve system monitoring information, display topology according to the condition and so on.In the meantime, it supports for a failure's node position, influences' analysis and fault's backtracking function.Visualization system data is from the configuration of the management module including its backend using Struct2+Hibernate4+Spring3 technology and its front visualization module using Unity3D engine.The whole visualization system has the characteristics of easy to use, good expansibility, service-oriented, configuration support and intuitive data presentation.

Abstract:In this paper, we design an image compression and encryption GUI system based on compressive sensing with Matlab platform, to improve the image storage space utilization and to increase the security of the image data.The system adopts the wavelet transform to decompose the data of images into sparse coefficients, uses Gaussian random matrix to sample the coefficients and stores in the server, to improve server space utilization;meanwhile the system takes the measurement matrix as a key to encryption, increasing security of the image information.The key matrix and reconstruction algorithm are used to reconstruct the original image when needed.The GUI can help to illustrate the process of the image compression and encryption system based on compressive sensing.Additionally, we compile the code into independent executable file, in order to run directly on the image compression, encryption, access and reconstruction.

Abstract:For environmental monitoring, the existing online analysis system is difficult to deal with massive atmospheric particle data.In this paper, we propose an atmospheric particle online analysis system based on real-time technologies, which aims to achieve atmospheric particle statistics, concentration change and the source analysis.The system adopts real-time databases to realize real-time capturing, stores massive atmospheric particle data, and solves the massive data problem in environmental monitoring.Besides, to accelerate data analysis and reduce data scale, the system adopts the ART-2a neural network algorithm and logistic regression model.The experiment results prove that the online analysis system could get accurate analysis result within a reasonable time.Besides, the experiment demonstrates the practical significance of our system.

Abstract:In order to ensure the real-time data processing of remote device, a real-time processing system based on RTX is proposed, which is composed of RTX real-time process and man-machine interaction.The RTX real-time process gets data from remote device and processes the real-time data.Then, it displays and storages data by transferring the processed data to a Win32 user interface layer by sharing memory and synchronizing event object.Simultaneously, the user could alter remote device upon commands from man-machine interaction.Experiments have shown that the system can reduce the latency of data processing, ensure the real-time and stability of equipment control, and satisfy the real-time control requirements for remote device.

Abstract:Due to the vast of urban shelter forest, water shortage, and dry climate, it is difficult to achieve timely and accurate data collection, and it is difficult for the forestry management department to supervise.Combining the GSM wireless communication technology and the embedded technology, the monitoring system of temperature and humidity of urban shelterbelt in arid area of Xinjiang based on Internet of Things is designed.The system uses the temperature and humidity sensor, combined with the development and control level of the existing embedded microprocessors, to carry out long-distance wireless transmission based on GSM data research, to achieve temperature and humidity monitoring and management purposes based on mobile phones and computers.Experiments show that the system has good practicality, reliability and scalability.

Abstract:Similar App recommendation is useful for helping users to discover their interested Apps.Different from existing similarity learning algorithms, the similar App recommendation focuses on presenting a ranking list of similar Apps for each App.In this paper, we put emphasis on how to learn a similarity function in a ranking scenario.Previous studies model the relative similarity in the form of triplets.Instead of triplets, we model the ranking list as a whole in the loss function, and propose a listwise multi-kernel similarity learning method, referred as SimListMKL.Experimental results on real world data set show that our proposed method SimListMKL outperforms the baselines approaches based on triplets.

Abstract:In order to solve the low recognition rate of template matching method in license plate character recognition, especially the similar characters, this paper proposes a method of license plate recognition based on template matching combined with local HOG features.First, we use the template matching method for preliminary identification of all the characters of a license plate.Second, the HOG features corresponding to the most discriminative characters among the similar characters of the plate and the template are extracted, and then the feature vector is constructed.Last, according to the Euclidean distance among feature vectors to measure the similarity of the license plate character and the template character, and then complete the second recognition.The experimental results show that this method is effective to solve the problem of false recognition of similar character and the recognition rate is significantly improved.

Abstract:An image set consists of a large number of different images, and these images represent the same person.In the real life, the dates of image sets are nonlinear due to the variation from viewpoint, emotion and illumination.Inspired by the manifold theory knowledge, we assume that modeling each image set as a manifold will be more efficient compared with the traditional method of modeling an image set as a subspace intrinsically.Because the images in an image set are different from each other, it is unreasonable to model an image set as a linear subspace without ignoring the data structure in the set, which may decrease the recognition rate ultimately.In the paper, we introduce a method of improved multi-manifold based face recognition for image sets, and propose a new method for computing the distance between two subspaces.In addition, in order to get the minimal manifold distance, we choose the mean value of closest pair of subspaces as the manifold distance.We call this new method, an improved multi-manifold (IMM).Experimental results on the public available face databases, CMU PIE, demonstrate that the new method outperforms the competing methods

Abstract:Aiming at the poor recommendation quality due to the data sparsity problem of traditional collaborative filtering recommendation, this paper puts forward an improved collaborative filtering algorithm.The improved algorithm proposes a collaborative filtering algorithm based on the similarity integration of item categories and user interests to make optimization on the similarity calculation.The algorithm does not simply concentrate on similarity calculation, but divides it into two aspects:users-item category interest similarity and users-item category rating similarity, which will finally be integrated with appropriate weights to get the final similarity.After a series of verification and comparison carried out on the MovieLens public data set, it is concluded that the improved algorithm based on data sparsity of collaborative filtering indeed plays a positive role in reducing the influence caused by data sparsity and improves the accuracy of recommendation.

Abstract:Due to the interference effect of the existence of the isolated points and prone points, SVM-based segmentation algorithm cannot obtain an ideal segmentation effect.An image segmentation method based on visual attention and fuzzy SVM (MFSVM) with improved membership degree function is proposed.In order to avoid the interference from non-vital training samples and noises, the segmentation result is coincided with the characteristics of human vision.The new membership degree function not only considers the distance of samples to center, but also divides the sample points into two types according to the distance of sample points to the center.It has enhanced the effect of support vectors and can effectively improve the segmentation accuracy.Multiple sets of color images are selected to verify the effectiveness of the proposed method.Result shows that, comparing with the standard SVM and FSVM methods, the proposed method shows an effective segmentation result as well as good noise immunity ability.

Abstract:Traditional collaborative filtering recommendation algorithm only bases on matrix.Due to the sparsity of matrix, the quality of recommendation is not high.This paper proposes a hybrid recommendation algorithm whose similarity is combined with the properties of projects.This algorithm improves the data sparseness in collaborative filtering through the change of the weighted factor, controlling the proportion of two kinds of similarity that one is the similarity of attribute between projects and the other is the similarity of item-based collaborative filtering algorithm.And the comprehensive prediction score and user-based collaborative filtering prediction score are combined to improve the quality of recommendation.Finally, the recommendation is given according to the comprehensive scores.Experiments show that the algorithm has better recommendation quality.

Abstract:The user's interest model is the core component in a personalized services system.Considering the impact of context information on user interests, this paper deeply studies the user behavior log data based on context information, and proposes a user interest modeling method based on context information.First, we get the user's context set by calculating the context similarity, and reduce the dimension of the "user-interest item-context" 3D model through the method of context pre-filtering.Second, user browsing content forms interest topic, and web page content forms interest keyword.Then a hierarchical vector space model is set up based on the user profile.The experimental result shows that the prediction error of user interest degree is controlled within 9%, which is effective.

Abstract:This paper establishes a mining algorithm based on the location and weight transaction item sets.It needs to scan the transaction database once.The algorithm adopts the vertical Boolean matrix to store transaction data, and the logic "and" operation and weight to calculate the support.Then it prunes through the searching location and combining the same columns of transaction matrix.Thereby the storage space and computing time used by the algorithm in the mining process can be reduced.The experimental results show that the improved algorithm performance has been significantly improved.

Abstract:In order to make the readers get the most informative and representative opinions efficiently among the news comments, this paper proposes a novel news article comments summarization algorithm and then designs an article summarization system, which combines the clustering algorithm with the ranking algorithm.First, it groups comments using the modified BorderFlow clustering algorithm.Second, for each group, it uses the similar PageRank algorithm to score and rank comments, and selects top comments in each cluster as representation.At last, it ranks the selected comments by MMR algorithm and displays the top-K comments as the comments summarization.According to the experimental statics of NDCG and MAP data, the proposed method meets the intuitive sense of readers more.Meanwhile, it shows the better effectiveness and accuracy theoretically.

Abstract:In view of the difficulty of obtaining the ideal effect by the single feature in image matching, an improved image matching algorithm based on deep convolution network is proposed.First of all, the algorithm expands the convolution layers, and efficiently computes the matrix multiplication by using the BLAS (Basic Linear Algebra Subprograms) libraries.The algorithm can accelerate the running speed.Then, a screening method of matching points based on the POEM (Pattern of Oriented Edge Magnitudes) feature similarity of feature points is used as well.The method can remove some wrong matching points, make the estimated fundamental matrix more robust and improve the repeating texture and rotational image.The accuracy and instantaneity of the algorithm are proved by the experimental results.

Abstract:At present, most of the researches of sentiment classification are carried out from the writer's perspective with quite few analyses from readers.This paper is to study the sentiment analysis from the news readers.A model of multi-label correlation sentiment classification based on completion matrix and LDA is proposed to extract the topic.The original news text is represented with the generated text-subject features, which are taken as the input to a subsequent classifier.Furthermore, the paper constructs a model of enhanced completion label matrix (CM-LDA) by appending the label correlation matrix to the original label matrix.Results show that the accuracy of this approach achieves 85.72% in the multi-label classification task, which outperforms the traditional LDA methods significantly.

Abstract:To resolve the problem of quality detection in the manufacturing process of smart card, an on-line visual inspection scheme based on SIM is proposed.We present a localization algorithm based on geometric feature matching combining the lighting scheme and algorithm design.It is implemented through the design of image's light source collection and research of the positioning and detection algorithm of module.Some key technologies, such as edge detection, feature extraction, opening and closing operation and object recognition are used in this algorithm too.It is useful for detecting if there is the chips, and other problems like dirts, rusts, banding severe bending fracture, decorative pattern, crease, silica gel on the back of migration, silica gel thick, scrapped modules silicone abnormal bulge, module stripe location hole on the back of damage, et al.Test results show that the system is an effective implementation of the quality detection of SIM card.

Abstract:EDI (electronic data interchange), as a kind of information technology, has been widely used in the field of shipping.However, since various countries and various shipping companies use different EDI packet formats and systems, the ability of data circulation is poor.In order to solve the problem of semantic heterogeneity in shipping field information sharing, this paper introduces the concept of ontology into the field of shipping and proposes an ontology construction method based on the XSLT and XPath technology.It realizes the semi-automatization conversion from the XML Schema document to the OWL syntax ontology document and builds the maritime domain ontology.The experimental results prove that this method improves the efficiency of ontology construction, and ensures the correctness of the ontology.

Abstract:When using optimization algorithms to solve optimization problems, keeping the diversity of population and accelerating the convergence rate of the population can improve the performance of an algorithm.To overcome the main drawbacks of the shuffled frog leaping algorithm which may be easy to get stuck and premature convergence in a local optimal solution, this paper proposes a novel differential shuffled frog leaping algorithm.The algorithm is based on the idea of mutation crossover in differential evolution.In the earlier, it uses beneficial information of the other individuals in sub-group to update the worst individual, which increases the local disturbance and the diversity of population;in the later, the algorithm uses the best individual information to conduct the mutation and cross operation for speeding up the convergence rate of the population.Moreover, this paper uses the archive to keep the diversity of population.The experimental results show that the proposed algorithm is superior to the basic frog leaping algorithm and the average frog leaping algorithm in solving optimization problems.

Abstract:Because of the existence of image noises, classical edge detection operators often fail to detect the edge exactly in many cases.In view of this, we analyze the geometry characters of edge under the condition of noises, and propose an enhanced algorithm of Canny operator, based on geometry edge enhancing.The algorithm firstly applies Canny operator to compute initial edges, and then excludes weak edges according to an automated threshold.In the following, it computes the directed gradients of remaining edges to improve the edges, and computes the double thresholds according to the improved edges.Finally, this algorithm detects filters and connects the edges by the double-thresholds method.This algorithm improves the accuracy of the edge detection, and can handle the false edges caused by noises or textures efficiently.

Abstract:The accurate identification of college student microblogs is helpful to location the sources of college student rumors and university public opinion events early, which can gain time for the university tutors and related authorities to take remedial measures and deal with emergencies promptly.It matches the college student microblogs information with the student background information from the university as much as possible, so that the student microblog accounts can be identified in certain area of university.The experiments are performed on three different threshold values, and the results prove the effectiveness of the identification by this loop matching method.

Abstract:In order to provide seamless network services for mobile nodes at any time, any place, the handover authentication technology is particularly important.We propose an ID-based handover protocol to achieve nodes privacy, which does not require third party involvement during the authentication.The proposed protocol only requires two handshakes without the pairing operation.Comparing with other protocols, the proposed protocol not only enhances the security but also performs better in authentication efficiency and supporting batch authentication.

Abstract:Finding features of users' trajectories in a period of time is one of the key point to realize user's personalized recommendation service.In this paper, how to find the interests in a period from the large amount of user's trajectories is presented with a filter-refinement strategy.In the filter step, the user's trajectories in the same period for several certain days are clustered based on density to obtain the user's stops;in the refinement step, the stops are clustered to obtain the user's interests.Finally, experiments show the effectiveness of this work.

Abstract:Since Android system components cannot fully meet the needs of developers, this paper proposes a custom general visual component, introduces the UI design principles based on Android and the realization methods of tables, and mainly discusses the asynchronous task mechanism used in the custom component.This component includes a circular diagram, a histogram and a line chart, and the circular diagram can display dynamically.Experimental results show that the component is simple and easy to use, and also has a strong commonality.

Abstract:This paper proposes a hotel reviews sentiment analysis model based on text features.The way of the model includes the building of hotel reviews emotional dictionary and the analyzing of hotel reviews sentence patterns and grammar patterns.By this way, we have solved the problems of incomplete emotional matches and the inaccuracy of emotional values counting.Results of this test show that this model has better classification results especially for hotel reviews sentiment analysis.

Abstract:In smart home applications, ZigBee and WiFi devices are two common wireless communication technologies.The problem is that the smart home devices in the ZigBee self-organizing networks can not directly communicate with the intelligent terminals in the WiFi networks.In order to solve this problem, a home-oriented application environment ZigBee-WiFi wireless gateway is designed.The gateway uses S3C6410 as the control core, runs on the Linux operating system, connects to a wireless network card with USB and connects CC2530 module with the DMA channel, which resolves the issue of mutual interference between ZigBee and WiFi signals by scanning and switching the transmission channels.Key is distributed to the new devices that join the ZigBee network by miniUSB connections.Experiments show that the wireless gateway has stable signal and less packet loss.It is stable and reliable in application of smart home system.

Abstract:The problem of excavating knowledge from historical parking data and forecasting the number of parking spaces in a short period is studied.By analyzing the factors that affect parking space, we establish a BP neural network in which the network input variables are defined through the combination of time series.Then, a self-adaptive studying rate is used in different stage of training and the momentum terms are added to improve the convergence of the network.According to the real data collected from a large underground parking in town, the simulation and analysis are executed based on Matlab, which results in well-accepted prediction effect.The conclusion shows that the proposed method can improve the prediction accuracy compared with the traditional time series prediction method.

Abstract:Resource public key infrastructure (RPKI) is a kind of technology which is used to protect the authenticity of Internet code number resources allocation and a kind of system of supporting inter-domain routing security which solves the problem of the lack of validation of route origination in BGP.However, it may result in the lack of authenticity and validity of ROA information due to the current data synchronism mechanism between the relying party of RPKI system and BGP routers.Meanwhile, it will bring a lot of performance load of BGP routes that query the cache lists continuingly.In this paper, we propose an improved method for route origination authentication.The sender routers real-timely apply for ROA certificates from RP and transmit them to the peer routers with the update message.Then the peer routers can apply for the public key to verify the certificates and verify the authenticity of the route originate.The verification mechanism is changed from updating the cache list periodically to real-time application for certification.It can effectively solve the problem that the ROA of the RP and the router data synchronization may be wrong, and reduce the running load of routes caused by querying the cache lists effectively.It is proved that the feasibility of the scheme using the simulation tool of Quagga and we make the detailed analysis for the applicable situation of two mechanisms.

Abstract:This paper analyzes and evaluates the motor vector control model of ARM platform, and makes up the blank of the evaluation in current servo control system.The analysis of the vector control model for motor and the environment of software and hardware need to realize the model based on ARM, which lays in the foundation of the analysis of vector control for motor based on ARM and performance evaluation.We use one method by idle task computing CPU load, and use this method to compute CPU load for performance evaluation experiments.Applications of vector control strategies under different conditions are analyzed.It has a guiding significance for vector control strategies choose based on ARM.

Abstract:This paper firstly describes the format and features of heterogeneous protocol packet.Then it describes the extended protocol packets by using the XML language, and modularizes the parser function to design a model of dynamic heterogeneous protocol parser based on the method of simple protocol analysis.Realizations applying this model can be achieved without recompiling, and only a small amount of code changes is needed to complete the model automatic matching of heterogeneous protocol packets, and then complete the analysis of packet data.We apply the model into two simple protocols, and the results show that the method of using XML to describe the protocol message and the reconfigurable parser can describe and analyze the heterogeneous protocol message effectively, and obtain reliable analysis result.

Abstract:Taxi service strategy is a group of taxi drivers' wisdom embodied, hidden in the large-scale taxi GPS data.Mining GPS traces using big data analysis tools to find and understand the service strategies of skilled taxi drivers to guide other drivers can increase their salaries and improve the efficiency of taxi operation.Passenger searching strategy is the main content of taxi service strategies, which loads GPS traces data into HDFS after the data pre-processing;splits the data to get each driver's personal GPS traces;calculates driver's salary;models taxi driver's service strategies;and then studies how the service strategies influence the driver's salary.A case study indicates that, the differences between the taxi drivers who have better salaries and the drivers who have ordinary salaries are significant in terms of passenger searching strategy.