For the lack of existing redundancy strategy and the improved ones in cloud storage system, a adaptive data redundancy strategy RIRS (Replication Improved Reed-Solomon) is proposed, which combined full backup with improved RS (Reed\|Solomon) erasure codes. The policy can adjust parameters according to the specific application environment to optimize system performance. When it degenerated into pure full backup method, proposed a dynamic replica management optimization model DRMO （Dynamic Replication  Management Optimized） to improve it. The analysis showed that the strategy can meet the requirements of data validation, save storage space, achieve load balancing and improve the recoverability of the system at the same time.

A new construction of ZCZ(Zero Correlation Zone)sequence pairs set is presented. It can be used in AS CDMA (Approximately Synchronized Code Division Multiply Access) system for the reduction or elimination of the multipath interference and multiple access inter access interference of the system. Based on ternary perfect sequence pairs and orthogonal matrix pairs (binary or ternary), the initial ZCZ sequence pairs set, the performance parameter of which could achieve Theory Circle, is constructed. Then extend the constructed ZCZ sequence pairs set by multiplying with unequal vectors after shifting, we get a class ZCZ sequence pairs set. The number of sequence pairs and the length of zero are related to parameters, different ZCZ sequence pairs sets can be obtained by setting parameters.

In order to overcome difficulty bright by the azimuth sensitivity and translation of SAR (Synthetic Aperture Radar) image,we presents a method of synthetic aperture radar image recognition based on two-dimension discrete wavelet transform and central moments feature extraction. After two dimension wavelet decomposition of the image, feature extraction is implemented by picking up the low-frequency sub-band image.And the central moments with translation-invariant property is extracted as feature vector.Support vector machine is used to classify the feature vector for target recognition. Experimentresults show that the method is an effective method that can reduce calculation and enhance the recognition.

To study synchronization control of delayed triangle chaotic systems, a delayed triangle chaotic system is designed and its chaotic dynamics is analyzed by the bifurcation diagram, Poincare map, power spectrum, the largest Lyapunov exponent and so on. The sliding-mode adaptive synchronization control scheme is proposed for the delayed triangle uncertain chaotic system. On the basis of the proposed chaotic systems, combining with the sliding mode control, the adaptive control technique is applied to achieve a single dimension signal controller. The experimental results show that the proposed sliding mode adaptive synchronization control scheme can realize chaotic synchronization for the delayed triangle uncertain chaotic system.

We optimize the structure and parameters of the punchthrough enhanced phototransistor. We change the structure from traditional form, which is the combination of a NB (Narrow Base) punchthrough transistor and a WB (Wide Base) phototransistor to just one side, to a new shape of the narrow base punchthrough transistor sandwiched in two wide base phototransistors which are both half of the old length. We investigate the electrical properties and optical conversion performance of the device such as dark current, photocurrent and optical sensitivity versus bias voltage with different width of the narrow base region. The optimized width of the narrow base is obtained after we analyze the simulated result. The analysis of the photocurrent and optical sensitivity versus bias voltage with different light power are presented to reveal the character of different light power. When the width of the narrow base is 0.6 μm, the performance of the device is a optimum compromise. The dark current is only 1 μA when the bias voltage is 0.5 V. The optical sensitivity is 4×10⁶ A/W when the light power is 10^-7 W/cm².

In order to provide electrical equipment with high stability, high reliability and constant current, this article has designed a constant current source for high power semiconductor laser system. the design used a low-sensitivity power amplifier tubes with a redundant parallel to improve reliability of the system, used high precision reference voltage source and current series negative feedback based on adder sampling to inerease the system stability. The power tube loss was analyzed Power triode 2N3055 was used and through wind tunnel type integration radiator to carry on the uniform temperature gradient radiation to the power triode, thus realized the constant current system power loss to be balanced. The experimental result indicated that the system can ensure higher reliability and lower than 0.982‰ (output current 50 A ) stability, at the same time, output 0~50 A continuously adjustable high-current, output power up to 350 W, and achieve a successful power loss balanced.

The support vector machine is used in lower limb rehabilitation training for device driving system modeling, one of the important issues is to select training parameters. In order to solve this issue, using the simulink platform collected samples, and through the libsvm toolbox, the related interface is experimented. In contrast experiments, three kinds of regression of advanced objective function optimization algorithm are selected. Algorithm in modeling application scope and value is discussed, and the basis for selecting is given. The grid search and particle swarm optimization have a mean square error close to zero, the correlation coefficient is greater than 99%. It applies to the demand of stability and practicality of the system. The genetic algorithm is not stable, but also has practical application. The results show that the method effectively solves the problem of selecting parameters, and at the same time for the further study of the object oriented objective function external optimization algorithm improvements
, especially for high requirements of real-time dynamic identification condition}, provide some basic support.

A dual buck-type DC/DC (Direst Cunent) parallel power supply system is designed to provide high-efficiency and stable DC bias for WLEDs (White Light Emitting Devices) used in VLC (Visible Light Communication) links. ARM7 processor (LPC2148) is used as the main controller, and based on the feedback voltage, the duty-cycle of a pair of PWM (Pulse Width Modulation) signals controlling the two DC/DC modules is adjusted for regulating the output voltage. By adopting two PMOS (P-channel Metal-OxideSemiconductor) switches, the currents of the two DC/DC modules are shared. To evaluate this power supply system, experiments are performed by using it to drive a WLED. The error between actual output voltage and expected voltage is less than 2%, the current error between the two DC/DC branches is less than 1%, and the power supply efficiency is within the range of 75%~80%. For the transient response, no overshoot or undershoot is observed during power on and power off processes. The rise time is about 20 s for the output voltage increases from 0 to 1.6 V, and the fall time is about 65 s when it drops from 1.6 V to 0. Besides, the life of this system is extended by using two DC/DC modules working alternately. This introduced proposal can be served as a solution for the multiple DC/DC modules power supply system to get high efficiency, stable output voltage and sharing current.

In order to solve the problem of the devices fragile and inaccurate data measurement, detection mode of dualchannel “and” relationship criterion and signal timing logic is presented. The system has four modes, each mode the double inductive sensor signal acquisition can effectively improve the durability of the sensor to protect the accuracy of the detection signal. Testing equipment has accurate test results, touch-screen real-time control parameters, durable and strong cost, with a high market value.

In order to streamline the camera speed in image processing, we introduce the work principle of rolling shutter camera. The perspective projection model of three dimensional space goals and camera imaging target is established. And a detailed theoretical derivation of the velocity measurement algorithms for moving target in space and plane is given. Finally, this velocity measurement method was experimental. The results show that the velocity measuring method using the rolling shutter CMOS digital camera can measure the moving target position and velocity parameters in plane. The test result is reliable, and accuracy error of measuring velocity is within the allowable range(3%). This method will have wide development and application prospect in the future traffic regulation fields.

Aiming at the wide use of the UAV (Unmanned Aerial Vehicles) in designated place to perform reconnaissance, detection or patrol mission, etc, the tasks cost model of the path planning of UAVs is transformed into the traveling salesman problem. The improved ant colony algorithm is employed to complete the path planning of UAV. To overcome the problem of easy convergence to the local optimum in the last stage for the basic ant colony algorithm, the cross tabu search strategy is introduced to improve the basic ant colony algorithm. The simulation study is presented to prove the effectiveness of the developed path planning method based on the improved ant aolony alg
orithm. The simulation results show that the developed algorithm can improve the optimization ability of path planning for UAV.

In order to solve the problem of the decision-making for the air-to-ground attack, the index factors of threaten degree of the target are analyzed and quantified. And a model of online support vector machine is built for threaten degree of the target. On the basis of above analysis, the attack decision is completed by sorting the threaten degree of the target. The developed attack decision-making algorithm using online support vector machine has features such as online training, small amount samples and good generalization ability. The simulation example is given to show the effectiveness of the proposed decision-making for the air-to-ground attack. The simulation results show that the online support vector machine has quick training ability and high-accuracy in the process of threaten degree of the target calculation.

DONG Lian-ying, XING Li-xin, PAN Jun, WANG Jing, LI Li-li, JIAO Jian-nan

In order to achieve the purpose of trajectory for 2DOF (Two Degrees of Freedom) manipulator, we propose an improved Qlearning algorithm which doesn't need the mathematical model of manipulator and can plan trajectory directly. The algorithm can dynamically adjust parameters of greedy strategy according to the study process. The simulation results show that the manipulator reaches the target position more quickly and the trajectory is the most optimal one when the new algorithm is applied to 2DOF manipulator trajectory plan.

To approach a full assessment of network, a network security evaluation index system including 5 first grade indices and 18 second grade indices is established. With the multilevel extension assessment method, a novel network security evaluation approach is proposed. Analysis of example demonstrates that calculation result is really consistent with the realistic situation, the approach is able to reflect accurately the security status of network.

Information accessibility construction is a social welfare service for vulnerable groups to obtain information equally. We will use the Apriori algorithm in Web log data to find the mining association rules to guide the design and to develop the information accessibility website. The Apriori algorithm was used on the information of the web page URL access frequency accessed by a large number of users to get the association rules.According to the 3-click principle and design of website structure, only all 3 frequent item sets should be find out based on the Apriori algorithm, the algorithm improvement reduces the time complexity of Apriori. Through the analysis of the association rule to get the type of webpage which most users interested in, we find the link relations of browsed pages.

To deal with the defect of present grain quantity measure methods, a new recognition method based on laser and three-dimensional reconstruction of bulk warehouse is presented. The implementation process of present methods based on image recognition is summarized, and the principle of the method is analyzed by us. At last compared both of them, and looked into the future of automatic grain quantity recognition.