Abstract：As one of the three IT bases, sensor technology has been widely utilization in all fields. But the intrinsic error of non-linearity output and temperature drift for resistance sensor has a strong impact to the entire system accuracy and seriously. So, the system must have temperature compensation and linear calibration. There are many deficiencies in hardware and software compensation method of resistance sensor. With the continuous development of sensor technology, new conditioning methods are constantly emerging. They greatly improve the system accuracy and versatility, to further reduce system power consumption and volume. Sensor conditioning technology will develop to intelligent, miniaturization and multifunction.

Abstract：Resistive sensor in military and industrial control field is widely used. With the development of microelectronics technology, resistive sensor conditioning circuitry also build towards an integrated and intelligent direction from the rapid development of independent components. This paper first to domestic resistive sensors commonly used treatment methods were analyzed , and then to the current foreign resistive sensors measurement circuit are summarized , and finally introduces a new type of domestic HKA2910 resistive sensor control chip , and sums up the domestic resesrch of resistive sensor processing chip of great significance in the field of military and industrial.

Abstract：The intrinsic of non-linearity output and temperature drift for resistance sensor has a strong impact to the entire system accuracy and seriously limited its application. Due to the disadvantages of area, power and weight of conventional solution . Base on the compensation research of the resistance sensor, a sensor signal conditioner chip has been designed and implemented and further more the compensation algorithm has been investigated and finally integrated into the user software.The measurement shows that，a good linearity performance of output versus the measured pleasure was observed at temperature range from -55 ℃～150 ℃,the total error is less than 0.2%. This new efficient way by design and implementation of sensor signal conditioner chip has been realized which make the sensor system more impact and more accurate.

Abstract：In sensor system, for getting the precise data, the signal from sensor must be compensated and processed. Signal conditioning chip of sensors is just about it. The article proposes some application solutions for signal conditioning chip, based on the existing chips. These solutions make compensation, calibration, and loop-power come true. Meanwhile, the article expatiates these solutions used in system by some typical cases. Comparing with traditional ways and means of signal conditioning, these solutions uses less accessories and depresses power. They can be used in operator, instrumentation amplifier and so on. They are able to meet various products′ need.

Abstract：The intelligent deviation compensation calibration software of silicon piezoresistive sensor based on independent designed signal conditioning chip, using modern signal processing technology and applying C# developing language is introduced. That realized the visual operation and control of the core compensation chip, and overcame the weakness of compensating the temperature deviation of pressure transducer through hardware circuit. Calibrating at many temperature points and getting the compensating curve to obtain zero and temperature drift compensation data, solved the consistency of silicon piezoresistive sensor, temperature drift and nonlinear problems. The result shows: the system works reliable, the output accuracy of the transducer that applying high precision temperature compensation algorithm improves a lot, the output signal between -55 ℃ and 125 ℃ has a good linear relationship with pressure, the measuring accuracy of pressure parameters attains within 0.6%.

Abstract：With the expansion of the field of deep space exploration, data transmission task is getting heavier and heavier. Physical Layer Network Coding(PNC) as a new kind of frequency multiplexing transmission technology, can improve the system throughput and spectral efficiency by taking the advantage of the broadcast nature of electromagnetic waves in wireless cooperative environments, which meets the requirements of data transmission capacity in deep space missions. In this paper, the development of deep space exploration at home and abroad is firstly introduced. Then the basic idea of PNC is explained and its benefit over traditional forward and straight forward network coding under two-way relay scenario is illustrated. Finally, the applications of PNC in deep space communication are proposed. And it is pointed out that some technical difficult problems must be solved in practical project application for future deep space exploration.

Abstract：With the fast development of Internet, sheer amount of videos are upload and download in which high proportion are near-duplicate. And these near-duplicate videos arise problems in copy protection, video retrieval accuracy etc, also incurs extra unnecessary cost. It’s becoming more and more important to find out near-duplicate videos from large video sets. This paper systematically surveys current near-duplicate video retrieval technologies, compares trade-off between accuracy and throughput, and illustrates state-of-the-art works and explore the potential trending in this field.

Abstract：This paper presents a 3D-Mesh interconnection network structure with 32 bit data path, which can support dynamic reconfiguration. Based on the 3D-Mesh topology interconnection network structure, a new type of isomerism coarse-grained reconfigurable array with 48 RPE and 16 RSE is designed. Based on the 55 nm COMS process library to design, the total area of ICGRA is 28.52 mm2. At the same time, total power consumption of the system is 2.88 W@300 MHz & 1.08 V Vcc in the room temperature. The network area of 3D-Mesh topology is 3.8% of the total area of the system, and the power consumption is 7% of the total system power. In comparison with the related classes, the structure dynamic reconstruction rate is 2~60 times higher. After using the 3D-Mesh topological network, the utilization rate of the arithmetic unit is also improved greatly.

Abstract：Based on graph representation of multiple constants multiplication, RAG-n algorithm is implemented by MATLAB. RAG-n algorithm can solve optimization problem of adder number in most cases efficiently, and reduce the complexity of FIR filter constant coefficient multiplication. The results of FPGA hardware synthesis show that this method can greatly reduce the number of logic elements applicable to low-cost design of digital systems.

Abstract：A CBGA(Ceramic Ball Grid Array) technology of multi-channels DDS(Direct Digital Synthesizer) packaging structure was introduced in this paper，proposed an improved isolation of the CBGA substrate form and used Ansoft HFSS software simulation to optimize the substrate before and after the isolation. The real test of the proposed design was carried out finally. The test results were in good accordance with simulation results, which show that the proposed optimization design can improve the isolation of multi-channel DDS and provide reference for the design of high-isolation DDS.

Abstract：A CMOS power amplifier(PA) operating at 20 GHz with 23.4 dBm output power is presented in this paper. The PA consists of two fully differential cascode amplifiers. The input, inter-stage and output match networks are all realized by on-chip transformers, resulting in single input and output. Output power combiner is adopted to enhance the signal waveform amplitude. This PA including on-chip transformers is designed based on TSMC 0.18 μm CMOS model and simulated with Agilent ADS. Layout′s post-simulation results indicate the input and output are matched well(S11=-13.85 dB，S22=-10.94 dB)，and the S21 reaches 21.5 dB at 20 GHz，the chip area is only 0.56 mm2.

Abstract：Study the magnetic coupling resonant system based on the equivalent circuit of coil and the coupled-mode theory first. The formulas are derived to calculate the transmission efficiency parameter of wireless power transmission system. Then a simulation using MATLAB programming is established to analysis the theoretical efficiency graphics. Finally, design a circular spiral magnetic coupling coil model using electromagnetic simulation software HFSS. We get the effect of energy transfer efficiency of the system by changing the distance of transmission coil and adding a relay resonant coil based on the simulation of the model. Results show that it can extend the transmission distance and improve the efficiency of transmission by adding a relay resonant coil. Although there are some errors, the simulation results are consistent with the theoretical analysis.

Abstract：A new type of Automatic Gain Control(AGC) circuit with the functions of over temperature processing and battery voltage tracking was designed for class D power amplifier. The AGC circuit blended the chip over temperature processing in AGC gain adjustment, solving the problem of sound disappear and appear suddenly leaded by traditional thermal protection ,and improving circuit user experience.When the battery voltage dropped too low, chip could not work. By mapping the battery voltage to AGC comparison threshold voltage，the circuit could avoid this phenomena and make full use of battery remaining capacity, extending battery work time.

Abstract：In this paper, a new embedded multi-sensor based on embedded technology and inertial, vision and magnetic field information is presented and the calibration method is studied mostly. A simple, practical, low cost, and suitable calibration method for different intelligent sensor is proposed. In addition, acceleration of gravity calibration method is used to calibrate the deterministic terms. Experimental simulation results show the feasibility and practicality of the model and the calibration method and further to extend the research and application of the system calibration method.

Abstract：Current commercial chips are widely used in spaceborne signal processing platform. But the commercial chips are weak to resist radiation and often appear SEU causing system dysfunction or even interruption. This paper aims at studying the subject of SRAM-based FPGA which is extensively used in the spaceborne signal processing platform. By adoptting the way of fault injection, we study different sensitivity to SEU effects presented in different hardware resources in FPGA. According to different sensitivity to SEU effects presented in different hardware resources, the protection for resisting SEU design of SRAM-Based FPGA will be targeted.

Abstract：Due to the situation that traditional commercial vehicle terminal information is displaying is insufficient and vehicle data extraction is difficult, a system is designed based on S5PV210 controller. Vehicle data collected via CAN bus network and resolved using J1939 protocol and it can display in car terminal and achieve remote monitoring system by 3G network. An expansion card is designed based on STM32 controller to achieve the driver identification, accident automatic alarm function using the acceleration sensor and RFID technology. The whole system has simple user interface，hardware and software are easy to expand to meet the diverse needs.

Abstract：To meet the requirements of smart home for intelligent controlling, a kind of smart home control algorithm is put forward, the algorithm is called by "TF algorithm" as being based on true and false value bitmap. Firstly, expression model is established, and then simple algorithm is presented, finally ultimate method is determined by improving simple algorithm. TF algorithm has characteristics such as requiring no training samples, coping with changes quickly, owning a small amount of calculation and being easy to be implemented. Practical experiments show that the algorithm has a strong adaptability and robustness, can be well applied to practical projects.

Abstract：Recently, it has become a hot topic that measuring Blood Pressure(BP) continuously and non-invasively via pulse waves(Photoplethysmography, PPG, Modulated Magnetic Signature of Blood, MMSB and Pressure Pulse Wave, PPW). This paper studies the relationship between the Phase Difference(PD) of three kinds of pulse waves(mentioned above) and BP through exercise recovery experiment and then puts forward the new method of BP measurement through PD between PPG and MMSB. The result reveals that the correlation between Systolic Blood Pressure(SBP) and PD is slightly higher than that between Diastolic Blood Pressure(DBP) and PD for different pulse waves. The PD between PPG and MMSB presents the highest correlation. It is promising to apply PD in the non-invasive and continuous BP measurement.

Abstract：Aiming at the problems of the current application of air quality monitoring system, such as the limitation of the scope and the poor real-time performance, the urban environmental health monitoring platform based on vehicle GPS terminal was designed and developed. The system includes vehicle environment sensing node, data service center and PC monitoring terminal. On the basis of STM32, environmental sensing node collects sulfur dioxide, PM2.5, temperature, humidity and other air quality information. Node gets its real-time location information and communicates with data service center by the SIM5320 module. PC monitoring terminal can check the air quality of the node's current position in real time, and draw the regional air quality distribution monitoring image. The system monitors the situation of environmental quality timely and accurately, makes up for the lack of the government's macro environmental monitoring, and provides a powerful reference for the control of pollution and the solution of environmental problems.

Abstract：With the development of the computer technology and sensor technology, location-based services gradually become a hot research topic. This paper uses the Texas instruments′ CC2530 chip designing a set of ZigBee indoor positioning system, and proposes weighted the neighboring location algorithm based on fuzzy clustering. By using ZigBee indoor positioning system conducting experiment, the experimental results show that using the weighted neighboring location algorithm based on fuzzy clustering, ZigBee indoor positioning system positioning accuracy have improved, the average precision reached 1.47 m. And the positioning effect is compared with NN localization, the KNN algorithm and Bayesian algorithm.

Abstract：To prevent the ice damage on contact wires and provide support to the melting ice, a monitoring system of ice coating on contact wires is designed, integrating the functions of video transmission, ice thickness measurement, wire temperature measurement, and the micro meteorological collection. The structure of the system including control unit and sensors, the selection of micro meteorological sensor, the design of wire temperature sensor, the electromagnetic shielding and anti-freezing design of camera are described briefly. The hardware and software of the control unit are introduced in details. The application of the HPI interface communication between DSP and ARM is studied and applied, which enables the transmission of videos and process of images by DSP, and the transmission of control signal and measurement data by ARM through EPON. The edge detection algorithm based on modulus maxima is optimized by the algorithm with the multiple threshold edges. The equations of ice thickness and the length of ice are put forward in the end.

Abstract：The forearm prosthetic hand designed for the disabled need achieve it′s first web keeping in parallel with the earth in the processing of movement like drinking and other actions. That is, the hand realizes the attitude self-balancing problem. Aiming at the estimation of the forearm prosthesis′s attitude angles, an attitude solution algorithm based on quaternion is used. Using Kalman filter to fuse the data from gyroscope and accelerometer in MPU6050, an optimal estimation for the attitude of forearm prosthesis is achieved after the compensation of the angular velocity drift error of gyroscope. The system consists of android phones, STM32 microcontroller、MPU6050、high performance of steering gear. Through the voice recognition, android mobile phone sends the instructions whether the forearm prosthetic hand need to keep the balance to STM32. Actuating signal can be send to steering gear through the serial port after data are collected and processed with algorithm from acceleration sensor and gyroscope in MPU6050 via STM32. The experimental results show that the attitude estimation algorithm is effective and the forearm prosthetic hand can keep balance during motion processes. This facilitates the disabled with the lack of the front arm.

Abstract：Considering the performance gain of channel assignment algorithm for wireless Mesh networks is closely related to the network traffic load characteristic，through analyzing the traffic load model in multi-radio multi-channel wireless Mesh networks, this paper proposed a static channel assignment heuristic algorithm called LPFCA. The proposed algorithm estimated the expected traffic load of wireless links by its location in the network, quantified the expected load of wireless link and divided all links into different levels. After that the channel assignment was formulated as integer linear programming and the overall network interference weight is minimized by using objective function for optimizing channel assignment. Simulation results show that LPFCA can averagely increase 18.9% throughput compared with the existing algorithm.

Abstract：According to the problem that the pattern matching engine based on software having slowly matching-speed and occupying large system resources，this paper proposed a FIBF structure based on Bloom Filter. Using the FPGA NIOSII microprocessor(MUP)，a matching engine combination of hard and soft core is designed. Though FIBF, the engine filtered out most of the normal data, then sent the suspicious string to the precise matching module. The FIBF module and the precise matching module were connected together by an index generator. The whole system was performed in pipeline method. The FIBF has low consumption of hardware resource. N-FIBF process in parallel can achieve high system throughput. Simulation results show that this engine is superior to other algorithms by using the same resources.

Abstract：In heterogeneous cellular networks, because of the different distribution density, coverage area and serving capability of different tiers, the network deployment causes complex and variable communication environments. There are key problems have to be solved for overall throughput improving. In order to enhancing the system throughput, this article compares existing representative sub-carrier allocation algorithms. The advantages and disadvantages of these schemes are addressed and the improvement strategy is elaborated in this paper. The proposed scheme is a sub-carrier self-organizing distribution based prediction, which can restrain interference, and provide theoretical reference for improving throughput, spectrum and energy efficiency.

Abstract：In this paper，a mathematical model of multi-delay false target smart jamming signal was established, and the digital algorithm was realized. The working mechanism how multi-delay false target signal jams pseudorandom code phase modulation fuse was theoretically analyzed. This jamming method was proved to be effective through digital simulation, and the main factors that affect the jamming effect was analyzed at the same time. The jamming efficiency of this method was offered in the last.

Abstract：Based on linear wave theory and Choppy Wave Model, this paper simulates linear and nonlinear dynamic sea surface, and compares the statistical characteristics of two ocean models under different sea conditions. Using small slope approximation method to solve the issue of microwave electromagnetic scattering, the article analyzes the echo Doppler spectrum with different factors. By using wave inversion algorithm to verify the simulation data, the modeling algorithm is proved to be correct. Results show that nonlinear waves have typical characteristics of Stokes waves, and nonlinear effects have a more significant influence on doppler spectrum at grazing incidence. The electromagnetic scattering of ocean surface added with nonlinear features are more consistent with sea surface conditions.

Abstract：Massive MIMO can achieve higher spectral and energy efficiency by large antenna array. The complexity of receive algorithm becomes the bottleneck of implementation. The MCMC detection algorithm can obtain the near optimal performance with low complexity. This paper presents an improved MCMC algorithm which introduces the over-relaxation iterative method to the MCMC detector. A relaxation factor is used to improve the convergence rate and reduce the complexity. Simulation results show that the proposed algorithm ameliorates the BER performance and the computational complexity simultaneously. The stalling problem in high signal-to-noise ratio of conventional MCMC detector is also resolved.

Abstract：In order to solve the problems that dynamic routing algorithms based on load balance have the redundancy control packet field, routing is not updated timely and the flow regulating mechanism exists defect, a Dynamic Routing Algorithm based on Load Balance(DRLB) is proposed in this thesis. The required information for updating routing is recorded by the satellite nodes, and then the backward agent reads the required information from the satellite nodes. Moreover, DRLB reduces the length of forward agent to reduce the control overhead of network and it also optimizes the selection of the destination node of forward agent to improve the efficiency of routing updates. Finally, DRLB optimizes the flow control mechanism to achieve better load balancing. Simulation results show that DRLB improves control overhead, the average end to end delay and so on.

Abstract：With the increasing demand of personalized products market, the degree of customization of products is higher and higher, the product design level, manufacturing capacity, service content of the manufacturing enterprises put forward higher requirements. But the existing enterprise design, production, management and other issues have seriously restricted the development of personalized market. In order to be able to promote the in-depth application of personalized customization, combined with the cloud manufacturing technology, based on the cloud manufacturing service platform, the mode of personalized service and the idea were proposed. A cloud manufacturing service platform for personalized customization was constructed, the architecture and workflow of the platform were analyzed. At last, the Apache server, PHP programming language and MySQL database were used to build the platform, and some functions were explained. And taking the design process of spiral bevel gear with high tooth number as an example, the feasibility of the platform is verified.

Abstract：In order to improve the performance of fall detection system for the elderly based on sole pressure sensor, and accurately to judge the fall direction of human body, a method was put forward based on self-organizing map neural network(SOM) and the information of sole pressure sensor to cluster and analyze the human motion. To verify the recognition results of the SOM method, 130 samples of 13 common action including fall were participated in the SOM network testing. The results show that the sensitivity, specificity and accuracy of the new system were 92.5%, 93.3% and 93.1% respectively. These results were better than those of the method of threshold value.

Abstract：In remote sensing image, the areas with a large number of color information points are prone to have spectral distortion during the process of image fusion. To solve this problem, firstly the image is divided into green vegetation area, water area and other area before the fusion process. Then a proper fusion algorithm is selected for each area to implement fusion, which can focus the advantages of different fusion algorithms to improve the fused performance. According to the experimental results, the new method has a better fusion effect, and the fused image can obtain a relatively high spatial resolution as well as a reduced spectral distortion, spectral fidelity of the fused image can reach a relatively high level.

Abstract：Modern data center network(Date Center Network，DCN) often uses multiple paths(MultiPath，MP) topology to avoid a link failure between two nodes caused by network congestion, and increases bandwidth and fault tolerance rate of network. Traditional OSPF(Open Shortest Path First) routing algorithm selects the shortest path as the final single path, that may lead to most of the data stream centralized in a single path and course the congestion of network, but other available paths are idle and can not be fully utilized the DCN link resources. Centralized scheduling based on SDN can improve network efficiency. This paper designs the overall framework and the algorithm flowchart of load balancing mechanism based on OpenFlow protocol, and simulation experiments prove the feasibility and effectiveness of the algorithm.

Abstract：In monitoring of high-speed train bogie working state, aiming at many freedoms of motion and strong correlation characteristics of different monitored data points of train, this paper puts forward fault feature extraction method by combination of multivariate empirical mode decomposition(MEMD) and permutation entropy. After the multi-channel synchronous joint analysis of vibration signals of seven kinds of working conditions of high-speed train bogie by using MEMD, the common pattern between different data channels can be accessed. Sensitive intrinsic mode functions(IMFs) that reflect the characteristics of fault signals are used to reconstruct the original fault signal via correlation coefficient, and the permutation entropy of reconstructed signal are calculated and taken as the fault feature. Finally, the support vector machine(SVM) is used to identify the fault state classification. Various experimental results show that the recognition rate can reach more than 85% of the classification results at various speeds, verifying the effectiveness of the proposed method.

Abstract：In order to solve the problem of low accuracy for the PFC in parallel, putting forward a power factor regulating circuit basing on automatic current averaging. This idea is aim to achieve automatic current averaging of the power factor regulator, introducing a negative voltage sampling from the output and automatic current averaging circuit. Saber simulation is performed to test the power factor regulator with automatic current. The result of simulation not only proves the practicability of this circuit, but also shows this design can reliably control the accuracy of average current of this power module within 5% when slight load in parallel.

Abstract：Wireless power recharging has widely application in the electric vehicle and wireless sensor network system. To improve the efficiency of wireless power recharging is the core of wireless energy applications. In order to enhance the transmission efficiency of magnetically coupled resonant wireless power transmission system, we have investigated the effects of enhanced coil(or relay coil), multiple receivers and magnetic shield slice on the efficiency of transmission of system. By optimum design of a system including an enhanced coil, multiple receivers, and adding a magnetic shield slice on the side of the received coil can highly improve the transmission efficiency of magnetically coupled resonant wireless power transmission system. The efficiency of the new and optimized system has a 17% increasing in the efficiency, in comparison with individual original component.

Abstract：A new type of automatic current-sharing technique for parallel DC/DC converter is studied, and the principle of the current-sharing technique is analyzed in detail. On the basis of this, the improvement is made, the exclusion scheme of fault module is put forward, and the realization process of fault elimination is clarified. At last, the principle prototype of a three phase parallel BUCK converter is developed, so that the correctness of theoretical analysis is verified.

Abstract：The control of energy storage system has important influence on the stability of the voltage and frequency of the isolated micro-grid system. In this paper, a small-signal model of micro-grid is established, especially including energy storage model. Based on the existing small signal analysis, the influence of the energy storage state and the control parameters of the inverter on the stability of islanded micro-grid is analyzed, and the rule of those influences is summarized. The simulation results show that the system has strong adaptability to load mutation and energy storage state fluctuation with controller parameters determined according to the rule.