The CSM (Coherent Signals-subspace Method) is a effective method to estimate the DOA(Direction Of Arrival) of the coherent wide-band signals by using focusing matrix. But the CSM needs more operation and is easy to be affected by advanced estimation, and it can not process signals in real time, because it needs focusing matrix and estimates in advance. This paper proposes a modified ISM (Incoherent Signals-subspace Method) which is based on the forward／backward averaging method. According to decomposing the wide-band signals into many narrow-band signals, and decreasing the relativity of the every narrow-band coherent signals by using forward／backward averaging method, we can estimate the DOA of the incoherent and coherent wide-band signals using ISM. This algorithm is realized easyly in decreasing the relativity. Although the capability is not as good as the CSM to process the coherent wide-band signals, the operation is less than the CSM. This is a fast algorithm for the DOA estimation of the wide-band signals. Simulation result and theory analysis show that this algorithm is effective.

In order to mitigate the receiver complexity and reduce the multipath fading effect in UWB(Ultra Wideband) wireless communication system,a multiple transmit antenna and one receiver antenna MISO-UWB(Multiple-Input Single-Output\|UWB) receiver model with preprocessed transmit signal of each antenna and the MRC(Maximum Ratio Combining) ware presented. And theoretical analysis of the proposed model was given. This method can effectively improve system performance in relatively low complexity by using both Pre-Rake diversity and transmit antenna diversity. Simulation results demonstrate that this type receiver possesses better performance when compared to the conventional Rake receiver and Pre-Rake system with one transmit antenna.

In order to improve the efficiency and reliability of the data transmission and satisfy the different QoS(Quality of Service) requirements. A kind of slot-scheduling mechanism  is introduced which is based on prediction and priority in IEEE 80216 Mesh mode: According to different priority of different service, slots are reserved for high priority service to meet different QoS; at the same time it predict the data quantity coming between the time SS sending messages and the time BS granting messages, so that it can improve the scheduling mechanism, and improve the characteristic of delay.

The vector control system of PMSM(Permanent Magnetic Synchronization Motor) has a wide application prospect in the fields of industrial control and medical treatment etc. The simulation research of vector control PMSM system can provide methods for PMSM vector control system design and realization. This thesis involves in simulation research of PI(Proportion Integration) adjustment, speed loop modulation and dq／αβ transformation, gaining SVPWM (Space Vector Pulse Width Module) waves and main loop based on module structure under the environment of MATLAB／SIMULINK. Scope space was used to observe the stator current, rotating angle, revolution speed of rotator and rotating of torque. Through adjusting the model parameters timely, vector control and velocity modulation of PMSM was realized. The simulation results indicat that vector control system has the characteristics of fast speed up, strong overload capacity and ideal speed adjustment.

When sensorless control system is applied in vector control and direct torque control system the responses are different for the different characters of the vector control system and direct torque control system. We analyze mathematic models of  PMSM(Permanent Magnet Synchronous Motor) and the results of rotor angle are generalized by the voltage and current which were meatured by sensors, the simulation results of vector control system and direct torque control system without position sensor are provided. The direct torque control system adopted MRAS (Model Reference Adaptive System) sensorless system are discussed and the results of the system are provided. The results indicate that rotor speed of vector control is steady. Ripple of rotor speed of direct torque control system is big and response of rotor speed is rapid. The ability of anti disturb of direct torque control system is better.

Dynamic test data of shooting range with non-stationary noise, this drift from the gyroscope, which seriously affect the accuracy of test data. The proposed use of wavelet can be an effective method for testing the dynamic range signals to deal with noise as to control the rate of drift in gyroscopes. To analysis, process and simulation the actual dynamic testing data of shooting range with Matlab. Simulation results prove that this method can greatly enhance the level of dynamic range test data accuracy.

In order to remove the nonlinear factor of the DC(Direct Current) rotor speed regulation system in large photoelectric tracking table, a servo system with sliding mode variable structure control was designed . It achieves quick response performance and insensitivity to variation of plant parameters and external disturbance of the variable structure control. According to the simulation and relative to the PID(Proportional_Integral_Differential) controller, results show that sliding mode control can reduce adjust time and tracking error, also has a strong robustness for external disturbance and parameter change.

Considering the complexity of the nodes in Ad hoc network in actual situation, we propose a routing algorithm based on MWMCDS(Maximal Weight Minimum Connected Dominating Set). This approach takes account of the diversity and complexity of the nodes in the network, so it weights each node on its power or online time. In this way, we will make the weights as large as possible while finding the CDS(Connected Dominating Set)as little as possible. So that we can find a highly efficient and reliable communication route to manage other nodes in the network. The network simulation and performance comparison has also been made for the 3 kinds of routing protocols. The result shows that in the same simulation scenarios, the PDF(Packet Delivery Fraction) in both DSR(Dynamic Source Routing)and AODV(Ad Hoc On-Demand Distance Vector)can reach up to 94％, better than DSDV(Destination-Sequenced Distance Vector). Only the route which uses higher PDF can cause the broadcasting routing algorism of Ad hoc network more effective.

Based on the research of handover process of GSM(Global System for Mobile Communications) network, corresponding parameters of handover were described, the setting principle of these parameters were analysed. Emphasizing the role of handover process of the GSM network,we put forward optimum way to set handover parameters and give GSM network optimization method from the perspective of handover, reduce the number of failure handover in GSM network, finally optimize the GSM network.

||To address organizational and retrieval of news video of news video content-based segmentation of the issue,we proposed a new shot segmentation method using adjacent frame difference between the characteristic values to determine the initial shear lens position.The Eigenvalue poor adaptive threshold can be adjusted and the potential location for the camera flash can be detected. Test results show that the algorithm can quickly and accurately detect news video camera shear. Which recall rate and precision to achieve an average of 94% and 89%, higher than the original methods have improved.

To resolve the problem of processing Chinese language in embedded DSP（Digital Signal Processing） real\|time video communication system, the design and implementation methods of message (in Chinese language) exchanging function were proposed in embedded IP multimedia communications system. Hardware design and software solution are given. Software processing technologies are discussed in detail. The method of multi-thread synchronization on the embedded platform is proposed. A compact Chinese Library is implemented in the embedded multimedia communication terminal device. Keyboard input is read through the I2C(Inter-Integrated Circuit) bus and edited into Chinese words using T9 pinyin processing tool. While guaranteeing real\|time audio and video communications, it also provides bidirectional speedy message (in Chinese language) exchanging function.

Most of the encryption engines in smart cards are designed with high operation concurrency. It takes one clock cycle to execute many encrypt operations, such as S box operations. This  deeply decreased the effect of DPA（Differential  Power  Analysis）. To remove this affect,  the plain-text chosen attack with the DPA was combined. The improved DPA  will only change the inputs of one of the DES（Data Encryption Standard） S box. Using this improved DPA,  the correct key on one of the smart cards with DES encryption has been found. The result shows that the improved DPA on DES encryption can increase the correlation value by 150％ than the traditional DPA analysis.

Computing the whole energy consumption of wireless sensor networks, analyzing the key factors of influencing the energy efficiency, and then optimizing the energy efficiency of networks are necessary to improve the whole energy efficiency. An energy model to evaluate the upper bound of energy consumption in wireless sensor networks based on the capacity of networks is proposed,beginning with analyzing the nodes status in data transferring. The optimum routing design for the network of link architecture is implemented using this model.

The characteristics of multivariate strong coupling, time varying, nonlinear and long time-delay in the magnetic separation process of Dagushan Ore Dressing Plant,make it difficult to use the conventional methodologies of optimal control based on mathematical model. With simple network topology constituted by input layer and output layer only, the new neural network with weight functions is proposed. The weight is function instead of traditional constant. On constructing of weight function, according to the production data in actual production process of Ore Dressing Plant and the gap of these data,different interpolation functions are selected as the weight functions. Simulation examples show the good performance of this method that little calculation work, high calculation speed, with no local minimum and slow convergence problems. Model mentioned above has minor error and the better prediction effect is obtained.

Aiming at the disadvantages of single BP net work, an ensemble BP(Back-Propagation) net work is presented by using AdaBoostR2 for improving the prediction accuracy of single BP net work. A new updating method is proposed for the characters of ensemble BP net work based on AdaBoostR2. The new method can update the model effectively and overcome the disadvantage of traditional updating methods. The new methods are used to predict the rock burst in Fushun Laohutai mine. The main influence factors for rock burst are analyzing detailed and used as the inputs of prediction model. The ensemble BP net work based on AdaBoostR2 is used as intelligent algorithm. The prediction model for rock burst predicting has a good prediction accuracy.

This article describes a parallel process method to  autogenerate LL(1) analytical table on multiprocessor computers, and it also discusses the designing and executing tactics in this parallel algorithm in order to improve the efficiency of compiler. With the help of mechanism of thread intercommunication, LL(1) analytical  table is autogenerated on multithread mode.

The energy-saving architecture has huge influence for total energy consumption in our country, in which emphasis is the design of thermal insulation. The structure composing and the temperature changing are analyzed. The model is derived by heat conduction equation. The design for proper thickness of flat roof heat preservation layer in cold area is presented, and numerical solution of indoor temperature is brought out by finite-difference method. The inversion process and numerical simulation are solved by dichotomy and Matlab. Economical optimum of insulation thickness is achieved. And the simulations are performed in order to offer complete theory analysis and experiment study. The proper thickness of flat is 29 cm.Thus, all the above researches have the positive meaning for practical application in engineering.

In order to solve the problems of mountain flood prediction, in this article we establish a mathematical model of mountain flood and develop a calculating program. We can simulate and predict the mountain flood situation under different conditions through changing parameter and coefficient in the model. It provides scientific basis for reducing geological disasters, preventing loss of water and erosion of soil, and protecting ecological environment.

The performance equivalence formulas of stochastic Petri nets can not be applied to analyze the time performance relative to instance arrival rate. To address the issue, the Petri net model was extended. The authors provided a new method of workflow modeling, i.eETdPN (Extended Timed Petri Net). Formal definition of extended timed Petri net was presented. Thereafter, this new method can be used to analyze four basic workflow models, i.e the sequence, parallel, choice and iteration model. We also provided a new calculation method of workflow response time by using queuing theory and stochastic Petri nets theory.  Finally examples of the four basic workflow models were evaluated with the proposed method using Exspect. Simulation results show that the largest relative error is about 2%, which proves the validity of the proposed method.