Journal of China Universities of Posts and Telecommunications最新文献

This paper cascades crash analysis and NS2 simulation analysis to study the network system hidden. It is combined with the theory of complex systems from the nodes and links in the network system vulnerability analysis. Firstly, we abstract complex network systems to the node and link network topology. And then we remove key nodes and link vulnerability analysis center between lines metrics to analyze the effect of the network functionality caused by node crashes or link crashes. Secondly, we construct a cascade collapse model, using NS2 simulation to study outside deliberate attack and simulation system failure analysis; and give the embodiment of cascading breakdown on the network function.

The virtual community has emerged as a state-of-the-art technology, grouping the user with the common interest and providing a feasible solution for large-scale deployment of resource sharing services. Efficient sharing for video resources and low maintenance cost of overlay topology are key factors in MP2P video on demand systems. In this paper, a novel virtual community-based video resource sharing (VCVRS) solution in MANETs is proposed. VCVRS design a virtual community construction strategy to group the nodes with the common playback behavior and a flexible community maintenance mechanism so as to obtain the fast supplier discovery and low maintenance overhead. Simulation results show how VCVRS achieves higher lookup success rate, lower server stress, lower node load and less overlay maintenance overhead, in comparison with another state of the art solution.

Traffic guidance is a promise approach of traffic congestion alleviation, and the travel time is one of the most important basic data for the reasonable and effective route planning which is the core of traffic guidance. The traffic intersection is one of the chief components of the whole traffic road networks, so the estimation of travel time of the intersection plays an important role in traffic guidance. This paper pays more attention to the estimation of travel time for left-turning lane connected to an intersection, introduces the features for travel time estimation, and designs an estimator based on the learning vector quantity (LVQ) neural network. A suite of reasonable test shows that the method can effectively estimate the travel time of vehicles at left-turning lane with lower error to the real data.

The traditional short-text classification's accuracy usually highly relies on statistical feature selection. Owing to the fact that short-text has inherent defects such as short length, weak signal and less features. It is hard to avoid noise words when doing feature extension which will highly influence the accuracy of classification. In order to solve the above problem, this paper proposes a semantic dictionary method for short-text classification. The method builds a set of domain dictionary by analyzing the specific characteristics in certain field. As each word's weight in the dictionary is designed according to the correlation between the word and the category, classification accuracy has improved to some extent. Then, in order to enhance dictionary vocabulary coverage, association rules are utilized to automatically extend semantic dictionary. Finally, an experiment based on micro-blog data is conducted which shows that the method has a good effect.

An new color image fusion method is presented based on dual-tree complex wavelet transform (DT-CWT) and Lαβ space for visual and infrared night image fusion. The color transfer technology is conducted to colorize the gray-scale visual image based on Lαβ space and the three component values of L, α and β can be gained. The appropriate dynamic range of gray image in Lα β space is proposed, and the reason of oversaturated colors is analyzed. The DT-CWT is applied in gray image fusion processing because of it's properties: shift invariance, directional selectivity, in order to obtaining exact position and clear image presentation. The different fusion rule is used aiming at high and low frequency components. The weighted average method is employed to low frequency part, and the high frequency parts are of selection greater local energy. The component L of colorized visual image is replaced by the gray-scale fused image. And the color fusion image is obtained by from Lα β to RGB(R: red, G: green, B:blue). The experiment results indicate that the proposed algorithm can achieve three requirements: detectability of the target, clear details and natural colors.

This paper integrated the quality of service (QoS) requirements of Internet of things (IoT) composite services, and put forward effective decomposition and optimization methods of the QoS metrics. The authors break down the complex QoS calculation model into four basic models, and each model is given a computational method. Combined with the QoS technology of the composite service, the authors adopt the algorithm to find the sub-optimal service with an acceptable cost under the QoS constraints. The proposed algorithm can quickly provide QoS computing services than other two algorithms for large-scale IoT compose services.

This paper presents a real-time communication framework for massive multiplayer online games (MMOG), which is to ensure the efficiency of communication and software readability at the same time. Firstly, we compare the current mainstream serialization technologies and demonstrate the significance of custom protocol through analysis of those technical limitations. Then, the paper puts forward the concept of protocol-bean innovatively. By this means, we implement an object-oriented custom protocol method, which has higher performance and more readability. Finally, through the comparison of data bandwidth and real-time response time, experimental data shows this framework could ensure the lowest data bandwidth without destroying software structure. And compared with real-time server Red5, response time is improved by 40 times.

Enterprise service bus (ESB) plays the part as intermediate hubs in enterprise service integration frameworks, and it is to implement the communications of shared contents in different systems. These contents are heterogeneous and message conversion protocols make rule of the transformation among them. This article studies about designing on enterprise service bus message conversion protocol based on extensible stylesheet language transformations (XSLT). It includes three parts of design of the ESB: message conversion frame, message conversion optimization and heterogeneous message integration. The first part introduces the process of service request and response using simple object access protocol (SOAP) in ESB and makes analysis about XSLT file elements. The second part sets up multiple-request message pattern by XSLT processor and ESB pre-defined request message content format based on Schema. The third part optimizes intelligent message conversion in ESB by automatic Schema mapping rule. All the functions help to develop the intelligent message conversion in ESB and improve the efficiency of the conversion.

Physical layer security is attracting more and more attention due to its inherent channel properties, while the increasing computing ability is not the obstacle for the traditional encryption any longer. In this paper, we study the base station (BS) and relay station (RS) placement problem in a cooperative secure communication system. Moreover, the system energy consumption problem also has been considered, and an energy-aware infrastructure placement for secure communications (EIPSC) scheme is proposed. Based on the analysis results of different candidate position of the security performance, location of the BS is determined and some imperative RSs are placed to guarantee the eavesdropped subscribers' secure communication. To decrease the system energy consumption, we propose a conception of sharing set of RS in order to place the RS as few as possible. During the BS and RS placing as well as adjusting procedure, renewable energy is also been considered to reduce dirty-energy consumption. Through computational experiments, we show our proposed algorithm can get better performance than the traditional placement algorithm not only at the system security guarantee but also at the system energy saving.

Driver fatigue severely affects driver's alertness and ability to drive safely. There are vital problems related to drivers fatigue on driving of trains, vehicles and airplanes. Therefore, the driver fatigue research is important. In this paper, we first study the impact of eye locations on face recognition accuracy, with Haar-like feature and AdaBoost classifier, face and eye area can be detected quickly and accurately. In the part of eye tracking, cam-shift based mean-shift algorithm is used to track the eyes. This method could automatically adjust the size of tracking window according to the different posture of driver. The performance of our eye detection method is validated by using image database with more than 6000 pictures. In addition, our real-time eye tracking system has been tested on railway line segment (China). There are 5 train drivers involved in the experiment. The validation shows that our eye detector has an overall 93% eye detection rate.