The LED(Light Emitting Diode)light communication protocol is currently limited to point to point short-distance communication,in order to improve the communication distance between indoor points, a light communication protocol based on the on-demand routing algorithm is designed, setting up a self-organizing and selfhealing wireless LAN(Local Area Network) indoor system. Simulation experiment and system testing is performed in the communication process between two indoor nodes after twice routings. Transceiver unit uses a pointtopoint halfduplex reliable communication. In the experiment,the communication rate is up to 30 bit/s and communication distance is about as three times as the direct communication, we find that the communication system based on this protocol can achieve the reliable routing path selection and adaptive transmission of sensor data in the space taht the light can illuminate.

In order to achieve fast tracking,orientation and test of dynamic target, wer designde a dynamic target image test system based on LabVIEW, realized the software programming using LabVIEW and IMAQ Vision. Through gathering, recognizing, tracking, playing back and processing data to the actual dynamic target image, the experiment result proves that the LabVIEW can process deviation target errors as well as the error statistical property quickly and accurately.  

When signal is coherent, the MUSIC(MUltiple SIgnal Classification) canwot estimates the directionofarrival accurately. To conquer this problem,  a new modifyMUSIC algorithmic is put forward. The principle of this new algorithmic is to calculate the cross covariance. Moreover, the new algorithmic has no loss of array aperture. The new algorithmic and the MUSIC algorithmic are simulated, then compare performances of the MUSIC algorithmic to that of the new algorrithmic. the result indicates that the new modifyMUSIC algorithmic is good at estimating the directionofarrival accurately when signals are coherent.

In order to solve the inability of personalized recommendation system on the semantic information processing, the semantic recommendation system model was built, which can descript the semantic sence in users interest information. A kind of recommendation algorithm was also proposed to calculate the relationship between users and resource. Description logic was used to apply the model in a special domain, two rules was introduced to achieve the ability of interest transmission between different level of semantic concepts both in user profiles and resource profiles. The experiment shows that semantic recommendation system model could produce more related results for particular user than classic methods, it can discovery more new interest which is implicated in the interest concepts.

In Mobile Network, a new data scheduling scheme named twodimension data scheduling scheme was propsoed to be used in wireless mobile P2P(PeertoPeer) streaming to address the scheduling issue in layered streaming. In the scheme, each data in the streaming was marked with twodimensional coordinates (blcok coordinate and layer coordinate), and the scheduling order of the data is determined by its twodimensional coordinates. The main contribution of our scheme comes in threefold:  it can assure the fluency of vedios playback on the mobile terminal;  making best use of the wireless bandwidth with minishing the useless packet ratio; achieving a low subscription ratio which brings the quality of the streamings stability.

 A large number of BT（BitTorrent） download on network, resulting in a lot of network bandwidth occupied, caused by network congestion and data transmission delay, causing in the paralysis of the entire network.In order to to block and dredge, the BT bandwidth with ACL(Access Control List) rule is used to make up for shortcomings limit BT, Restrict the bandwidth of the effect of BT.

Using physical model method to simulate ink-wash drawing is proposed in this paper, using movement to simulate stationary. The object of this simulation is adsorbing ink-wash drawing which is abstract and easy to simulate. The motion of the ink\|wash is controlled by Navier-Stokes equations which are resolved by semi\|lagrange method and realized on GPU（Graphics Processing Unit）. The result of the simulation is successful; the works of the simulation is full of the character of ink and wash painting, and also have effect on abstract art.

Because the spandex production technology is complex, the production cycle is long and raw material is expensive, it is impossible to do technological parameter optimization experiment and optimize the operation. In order to solve this problem, a simulation system of spandex is built. Then the operation can be carried on this simulation system to optimize the operation, to optimige technological parameter , and to adjust the raw material and formula. After obtaining the success, these optimization methods can be used on the production line. At the same time, this simulation system also provides a good platform for the operators training and the fast diagnosis of equipment failures.

For Infrared images with the background of high and low contrast features, a new adaptive enhancement was put forward based on the balanced spacing density of grayscale equalization. According to the grayscale image theory, we used the recursive window, selected the histogram window adaptively and got the high and low thresholds. then, through the gray linear translation and the balanced spacing density of grayscale, the grayscale image dynamic range is inereased and the infrared images isobtained with rich detail and continuous grayscale. Theoretical analysis and experimental results show that, this proposed method overcomes the discontinuous gray caused by the general histogram equalization and plateau histogram equalization, achieves high contrast and greatly improves the quality of infrared images.

In order to improve the accuracy of automatic monitoring and fault diagnosis in the railway vehicles
and other engineering applications, the five methods of feature point detection are compared and researched in the paper. SIFT (ScaleInvariant Feature Transform) method is selected to detect feature points from the overlapping images, which can mosaic precisely the two images under most relaxed conditions. The experimental results demonstrate that the proposed method can mosaic effectively the overlapping images captured by ordinary camera and obtain high resolution image, eliminating the influences caused by image distortion, overlapping and rotation.

ords similarity has a role which cannot be ignored in the semantic web adaptive learning system. We propose and implement words similarity algorithm based on Tongyici Cilin, in which we fully analyze and use the coding and structural characteristics of Tongyici Cilin. We consider both the words resemblance and the words relevance. After manual test, replacement test and comparison test in which we compared our algorithm with the current popular words similarity algorithm based on Hownet, we found that the algorithm achieved the good results.

To resolve the problem of distinguishing acupuncture point for the unprofessional persons, we design a recognition and detection instrument with SPCE061A. The principles and diagrams of hardware and software architecture of the entire system are described. Based on theories of skin resistances on the acupoint resistance, a new acupoint detect instrument containing electrical stimulating is invented. The recognition acuracy on acupuncture and nonacupuncture can reach to 95% and 98%, respectively. Controlled through a MCU（Micro Control Unit）, the apparatus can output biphasic constant current and special pulses for electrical stimulation. The theory of meridians is significant for the basic theory of traditional Chinese medicine. The effects of environment on acupoint detection is discussed. Compared with the traditional method, this can improve and simplyfy the operation. 

Due to the development of short time, lack supervision of government,  the whole industry was in an out-of-order development stage. The history and actuality of the network payment ware reviewed, several major operating model of network payment  and operating modes were discussed.At last, for promoting the whole network payment industry to develop healthily and rapidly, several supervision proposals were offered.  First use of “social welfare”:a concept of the economics,  the advantages and social benefits of e-commerce and the network payment were evaluated,  the development trends of the network payment in future were pointed out.

In the traditional videoprocessing platform with detached DSP(Digital Signal Processor) and ARM(Advanced RISC Machines) processors, there is some disadvantages such as large print circuit board, the small amount data access capacity and the difficulties for exchanging data. A new method is provided with DaVinci platform for an embedded video surveillance system, which is achieved with a dualcore processor chip DM6446. The realization is introduced in three function blocks, one is embedded Web server, another is encoding for video capture and video transmission, the last one is about protocol stack for network communication. The video surveillance system is built up in B/S(Browse/Server) architecture. The functions are fitted into ARM and DSP core processor correspondingly, which leads the best performance optimization on one chip.

To resolve the problem of insufficient heat supply and energy dissipation caused by cant monitoring the heat effect in the process of heat supply, the design and implementation methods of realtime monitoring far end interior temperature were proposed. Meanwhile, hardware design, software solution and working processing of temperature collection and communicator were discussed in detail. Atmegal16L microcontroller and CMX865 modem were embedded in telephone to collect far end temperature and control FSK(Frequency Shift Keying) to make telephone communication. Test results show that the system has the advantages of simple in structure, low power consumption and stable working. It effectively realizes the real-time monitoring of far end interior temperature.

For early detection of boilers potential failure in power plant the high noise background fault self\|adaptive detection system is studied. A higher background noise fault signal detection algorithm is provided. Simulation experiment results, and points out that the efficiency of the method. The traditional cepstrum analysis is applied to signal based on FFT (Fast Fourier Transformation) transform, and the algorithm is based on the improvement CZT (Chirp Z\| inverse Transform) transform, get the failure characteristics of curve more stable, and more reliable. The experimental results show that this method is fast and effective, fault quoted accuracy over 99%.

In order to resolve the problem in the E-Government systems, such as high platform-dependence, lacking uniform standards and criterions,a system using J2EE(Java 2 platform Enterprise Edition) technology and Struts frame is developed. XML(eXtensible Makeup Language)-based  standards for data exchange method is used to resolve heterogeneous data exchange between different government departments. Practice have proved that system have cross-platform, expandability, distributed, reusing features.