This paper presents a new powertrain configuration based on power distribution design of an extended-range electric vehicle. First, the parameters match and equipment selection were studied and a control strategy was proposed. Then, combined with the matching results and test data, the mathematical model of the vehicle was built with AMESim and Matlab-Simulink software platform. The model was analyzed aimed at the dynamic performance, EV-mode mileage, regenerative braking efficiency and extended-range model fuel consumption. Results show that the extended-range electric vehicle with the power distribution design has a smaller curb weight and higher regenerative braking efficiency without compromising the dynamic performance. This configuration could improve the EV-mode mileage for more than 30%, reduce the extended-range model fuel consumption for 3% ~ 6%, and has ideal energy-saving potential.

Based on the game rules of World Solar Challenge (ESC), a conceptual solar race car driven by one in-wheel motor is designed. First, the powertrain parameters are matched. Then, according to the input power of the solar panel, the velocity control strategy during the racing is proposed. Finally, the speed control strategy is verified by simulation in Matlab. results show that, based on the designed powertrain and cruising strategy, the solar car can not only complete the race under various conditions, but also can travel at an optimal speed, which can reduce the energy consumption and speed the least time.

Based the quadratic subdivision mathematical model of the braking severity, a design method of the predictive model for parallel cooperative braking system is investigated systematically. An off-line optimization stream is designed. First, the continuous design space, which is constituted of the vehicle speed v, battery State-of-Charge (SoC) and braking severity z, is sampled by Design of Experiments(DOE);two subsystems of regenerative braking energy and braking stability are defined and optimized by collaborative optimization method(CO) based on the sampling points; finally, the predictive model is established for the cooperative braking system. Simulation results show that the proposed model can realize real-time optimum control and maximize the regenerative braking energy recovery efficiency under the condition of ensuring the braking stability. Accordingly, it has high engineering application value.

In order to improve the vehicle dynamic stability, the driving and braking force distribution between the front and rear axles for 4WD electric vehicle is investigated. According to the friction circle of tyre force, the ideal force distribution ratio is obtained, at which the wheels of the front and rear axles reach the adhesion limit simultaneously when the vehicles operate at various conditions. The vehicle handling stability under different distribution methods is analyzed by simulation using 4WD vehicle model. Simulation and theoretical analysis results show that the idea distribution method proposed in this work can improve utmost the lateral stability of the vehicle under the premise of ensuring the current longitudinal acceleration. In addition, for 4WD electric vehicles, the change of turning radius with longitudinal acceleration is minimized and the vehicles have the best maneuverability. Furthermore, the ideal distribution method is simple in calculation and relies little on the vehicle state information, so is easy for application in real vehicles.

The deadbeat control scheme is adopted to improve the current tracking performance of permanent magnet synchronous motor of electric power steering system. However, the stability and accuracy of traditional deadbeat control suffer from time delay, motor parameter perturbation and other disturbances. To overcome this problem, a prediction current control algorithm based on disturbance observer is proposed. The algorithm combines deadbeat control with a current observer and disturbance observer. The current observer estimates the future value of motor current and compensates the time delay; and the disturbance observer estimates the system disturbance to improve the current observation and control precision. Simulation and experiment results demonstrate that the proposed algorithm significantly improves the steady and dynamic performance of current tracking of the electric power steering system.

The rear area of the floor of light bus experiences severe vibration under high speed working condition. In order to solve this problem, order tracking analysis and modal analysis of the floor were performed to identify the root cause. It was found that the excitation frequencies of the engine and transmission system are overlapped with the modal frequencies of the floor, which has some local bump modals, resulting in resonance. A finite element model, including the body and frame of the floor, was established using HyperWorks software. Model sensitivity analysis and size optimization of the thickness of floor the beams were carried out using the OptiStruct module in HyperWorks. Then an optimal scheme was designed according to the finite element analysis results and the verification test was conducted. The test results demonstrate that the investigation method and the proposed optimal scheme can effectively control the severe vibration of the floor with low cost.

A shear-lag analytical model for aluminum-magnesium alloy single-lap joints was established to predict the joint shear stress. The predicted distribution of shear stress was verified using finite element model. Based on the three dimensional elastic-plastic finite element model, the distributions of shear stress and Von-Mises equivalent stress of the alloy joints in the adhesive layer thickness- and width-direction are analyzed. The effects of various parameters, such as adhesive type, layer thickness and adhesive length, on the bonded structure strength are discussed with experiment and simulation. The results of this study may provide guidance for matching different kinds of metals and adhesives and are helpful in strength prediction and performance optimization of adhesive joints.

Based on the description of a dilemma-zone avoidance-guiding system, a method for determining the activation time of the system is proposed. First, the impact of the activation time on vehicle guidance is analyzed. Second, the activation time needed for the vehicle, which would have been caught in the dilemma zone, to cross the intersection with an acceleration strategy is computed. For high-speed vehicles, a gain model is proposed for determining the upper bound of the activation time; then the activation time for vehicle, would have been caught in the dilemma zone, to pull up before the stop line with a deceleration strategy is computed. Finally, the greater value between the two activation times is taken as the activation time of the system. Theoretical analysis and numerical examples show that an appropriate activation time can be determined by the proposed method, which can increase the possibility for vehicle to cross the intersection using accelerating strategy without being caught in the dilemma zone.

In order to describe the time headway distribution characteristics of different control types at roundabouts, an investigation was carried out in different sections and different lanes at roundabouts both with and without signal heads in Changchun. Several distribution models were used to analyze the headway distribution in both weaving and circulating areas, such as Weibull distribution, shifted lognormal distribution, double shifted negative exponential distribution and JohnsonSB distribution. The maximum likelihood estimation method and Levenberg-Marquardt method were used to estimate the parameters in the distribution models, and a single-sample Kolmogorov-Smirnov test was used to examine the goodness of fit to the distribution models. Finally, the relationships of lanes between different controlling types and different lanes in the same roundabout were measured by two-sample Kolmogorov-Smirnov test. The results show that there are significant differences in headway distribution among different types of roundabouts, and certain differences in different regions under the same type of control; and JohnsonSB distribution and shifted lognormal distributions have better goodness of fit among the distribution models.

With the objective of improving the capacity and avoiding overflow for oversaturated intersections meanwhile considering bus priority, a method is proposed to determine the queue length threshold, detect the threshold and estimate the queue length. Then according to the traffic flow conflict and section space constraints, a method of dynamic phase combination is designed. From the traffic flow characteristic of oversaturated intersections the key parameters are defined, in which the weight of bus priority is considered. On the basis of the above work, the control method for oversaturated intersections is studied with gather-disperse theory and fuzzy control. Finally, an interconnected control method of traffic signals for oversaturated intersection is put forward. Simulation results show that the new method can not only restrain the overflow in oversaturated intersection, but also improve the average traffic flow and effectively reduce the average delay time at intersections.

Through the description of the degradation parameters of each road section capacity, based on the long time habitual traffic equilibrium model and combining with features of the two-way branch road sections to expand the travel time functions of roads, a traffic flow equilibrium model of branch network based on travel time reliability is constructed. In order to overcome solving difficulties of the model caused by non-convexity, the general method of successive average for solving traffic distribution is broken up one after another, and a method of successive average is designed, which is implemented step by step from the origin points. Example analysis shows that under different reliability requirements, there are differences in the travelers' choices of the arterial road sections and the branch road sections, and for more conservative travelers the probability to choose the branch road sections is smaller.

Based on the model parameters the characterization factor is defined. Then according to the definition, a method is developed to analyze the vehicle characteristics using observed data. The method consists of five steps: named data collection, data processing and parameter calibration, parameter correlation analysis, characterization factor calculation, and characterization distribution study. Once the characterization factor distribution is known, traffic simulation based on vehicle characteristics can be conducted with conversion of characterization factor and model parameters. With Optimal Velocity Model (OVM) and Intelligent Driver Model (IDM), a study on vehicle characteristics is conducted using the data collected in Nanjing. A heterogeneous IDM model is proposed and used for simulation based on driver-vehicle characteristics.

The performances of several mathematical models are evaluated for the investigation of the excess zeros issue in crash data analysis. These models include: zero-inflated Poisson and negative binomial models, multiple zero-inflated Poisson and negative binomial models, Poisson- and binomial-Lindley models. In the evaluation several goodness-of-fit measures are employed, including Deviance Information Criterion (DIC), Mean Absolute Deviance (MAD), Mean Square Predictive Error (MSPE), Maximum Cumulative Residual Plot Deviance (MCPD) and Cross-Validation assessment (CV). The evaluation is conducted based on the crash data sets of a state of USA in year 2010, and Markov chain and Monte Carlo methods are used. Statistic results show that Lindley models are preferred to multilevel zero-inflated models; among all the six models, the negative binomial-Lindley performs the best.

A mathematical model of EMU was established to analyze effects of suspension parameters on the suspension stiffness of the bogie. Then lab test was conducted to measure the actual suspension stiffness, which is used to verify the theoretical calculation results. The characteristics and distributions of the suspension stiffness under various conditions were summarized. Theoretical analysis shows that the rotation resistance factor of the bogie fluctuates greatly with the longitudinal stiffness of air springs and the lateral distance between two air springs, especially for small radius curves. Therefore, the rotation stiffness of the bogie and the anti-rolling stiffness of the vehicle should be considered simultaneously in the vehicle design stage. Test results of the primary suspension and second suspension stiffness are both a bit higher than the calculated stiffness with maximum error of 11%, which suggests that parameter verification after the assembling is essential and necessary. The test results also show that the greater are the rational angle and speed, the greater is rotation resistance factor. For a 300 m radius curve at inflated state, the rotation resistance factor is 0.023 with rotational speed of 0.05 °/s, and the resistance factor is 0.065 with rotational speed of 0.2 °/s, indicating that the higher is the speed, the greater is the rotation stiffness. However, at deflated state the corresponding resistance factor is 0.068 and 0.095, much greater than that at inflated state. We can be concluded that it is very dangerous when a vehicle passes small radius curve at high speed. The text results of rotation stiffness of the bogie are greater than calculated ones,and the dynamic characteristics of air spring stiffness and other suspension components should all be considered in theoretial calculations.

In order to reasonably evaluate alkali content generated in alkali-aggregation reaction in concrete, the influence of carbonation on sodium ion concentration distribution inside hardened cement paste was investigated; the alkali-aggregate reaction characteristics in concrete under carbonation were clarified. The results show that sodium chloride distribution is relatively uniform before carbonation; during carbonation process sodium ions in the carbonation area migrate to the non-carbonation area, sodium ion concentration decreases in the carbonation area and increases in the non-carbonation area, and reaches maximum peak value in the carbonation interface. Thus the carbonation increases the sodium ion concentration in non-carbonation area in the concrete and reduces the initial critical sodium ion concentration for alkali aggregate reaction in the concrete that accelerates the alkali aggregate reaction.

The axial bearing capacity of thin-walled steel members with web holes and different section shapes are investigated. Ten simply supported compression members with two types of complicated sections, including channels with complex edge stiffeners and web holes, channels with complex edge stiffeners and Σ type web stiffeners and web holes, were tested. Results show that the axial compression loading efficiency of the Σ-section specimens with web holes can increase about 30% to 50% compared with channels with complex edge stiffeners and web holes under the same conditions. Compared with the specimens without holes, there is no change in failure modes, but the failure positions are near to the holes. Finite Element Analysis (FEA) is carried out to simulate the compression tests. FEA results agree well with the experimental data. Based on this, the load carrying efficiency and Von Mises stress distribution of channel columns with web holes are compared with those without holes by FEA. It is shown that the carrying efficiency of Σ-section models with holes decreases by 25.7% compared with the same section models without holes, while the model of channel section with complex edge stiffeners and holes decreases by 6.15% in average. Because of the perforations on the web, the positions of the stress concentration and the maximum stress change from the webs near the mid-height of the models to the locations adjacent to the holes.

This paper is focused on the effects of different injection and spark modes on the in-cylinder equivalence ratio, temperature field as well as the NO formation. The simulation process is based on the experiment on a direct injection Compressed Natural Gas (CNG) engine, which was observed with optical device. Results show that the in-cylinder equivalence ratio distribution, which can be controlled by the injection mode, is the key factor for lean combustion control. Moreover, the equivalence ratio distribution directly impacts the flame propagation and temperature field. The spark mode mainly influences the in-cylinder temperature filed, so that influences the NO formation. The higher the gradient of equivalence ratio distribution, the faster the flame propagation and the more stable the lean combustion. The double spark mode can effectively control the flame propagation speed.

Temperature rise greatly impacts the performance of the GDI injector and causes the ECU of an engine difficult to control fuel quantity accurately. According to the coupling relationship of electric, magnetic and thermal subsystems of the GDI injector, the physical model of the coupling process was established. The energy loss and transformation of the GDI injector in working process was explored. It is revealed that the loss of electromagnetic conversion is a key factor to cause the temperature rise of the GDI injector. The energy dissipation as heat source and the temperature field of the GDI injector was analyzed based on the finite element equation of the heat conduction and thermal radiation. The temperature field of the injector was simulated using ANSYS software; then the simulation results were verified by experiment. The effects of the temperature field distribution and the temperature rise variation on the performance of the GDI injector were analyzed. Results show that higher temperature rise is concentrated on the coil and core part, the response time of GDI injector gets longer as the temperature rises, and the injection quantity becomes lower. The value and the pulse width of the holding current have great impact on the GDI injector performance and temperature rise.

In order to achieve non-contact transportation of heavy load objects, a new non-contact automatic transportation device using aerostatic suspension is proposed. The working principle of the aerostatic bearing and non-contact transportation is studied, and the relative force analysis is carried out. Then, the physical model of the suspension structure and transport structure is established and the experimental prototype is manufactured. The electronic control system of stepping motor is set up and PLC program is designed using WPLSoft software. In addition, the influence of transport pressure and nozzle rotation angle on the transport velocity of heavy load objects is investigated by experiments. The experiment results show that the capability of the aerostatic bearing is enhanced sharply, and the heavy load objects can be transported steadily. When both suspension pressure and transport pressure are 0.2 MPa and the nozzle rotation angle is 32.4°, an object with mass of 1 kg can be transported steadily at velocity of 62.5 mm/s.

In order to overcome the shortages of lower efficiency and higher fuel consumption of truck-mounted concrete pump, a global power matching method is proposed based on both fuel consumption rate of the engine and the efficiency of hydraulic pump. The load distributions are summarized by mathematical statistics tool. The fuel consumption rate of the engine is obtained on an engine test platform. Experiments are conducted on the hydraulic pump to find the high efficiency region with different pressure, rotation velocity and displacement. The combination optimization of engine fuel consumption rate and pump efficiency are carried out using genetic algorithm. The engine speed and the pump displacement are jointly adjusted to make output control indexes adapting to the load change in real time, ensuring all the components work in the high efficiency region. Comparative experiments are conducted with new and old methods. Results show that the new method outperforms the old one that the fuel consumption is reduced by 16%.

Based on the Support Vector Regression (SVR) model, a method is proposed to assess the parameters of the reliability model of machining centers by analyzing censored data. An improved Local best Particle Swarm Optimization algorithm (improved lbest PSO) is developed to tune parameters of the SVR model to keep working efficiently. The improved Lbest PSO, developed from the local best PSO (lbest PSO), introduces a mutation operation and an adaptive inertia fact to improve its ability to search the global optimal solution. Additionally, the Least Square Model (LSM), the Maximum Likelihood Model (MLM), the SVR model selected by the Lbest PSO and the Genetic Algorithm (GA) are employed to compare their estimation performance with the proposed method. Results show that the proposed method is superior to all the other models. Finally, this method is used to estimate the parameters of the reliability model of a type of machining center and obtain its mean time between failures.

A Trajectory Control Method (TCM) and an Impedance Control Method (ICM) are presented. The methods are used to control the robot system so that patients can safely and effectively conduct lower extremity rehabilitation training for standing-up process. During standing-up process, the body segment rotational angles, movement trajectories, Ground Reaction Forces (GRF), Centre of Pressure (CoP) and rope tensile forces are measured by the robot sensor system; the joint moments of ankle, knee and hip are calculated in real-time control program for feed-back-control. Experiment results show that the methods are suitable for standing-up rehabilitation training robot control system, and ensure safe, comfortable and effective rehabilitation training of the lower limbs.

To analyze the change rule of body center-of-mass momentum during sit-to-stand motion assisted by rehabilitation robot, the trajectory and center-of-mass momentum of human body are measured. A mathematical model of the body center-of-mass momentum is derived based on Lagrangian equation. The position of mass center during standing up is obtained by auxiliary parallelogram method. The trajectory and momentum are measured during standing up with low, normal and fast assisting velocities. A simulation model is built up in Matlab/Simulink based on the proposed mathematical model to simulate the trajectory and momentum. The results show that sit-to-stand motion is carried out steadily with different assistant velocities of the robot. Through the analysis of horizontal and vertical peak center-of-mass momentum, it is found that the horizontal momentum is the determinant of the stable standing up motion, and the vertical momentum is the determinant of the complete sit-to-stand process. The simulation and measuring curves fit well, which demonstrates the effectiveness of the proposed model in predicting body position and center-of-mass momentum during sit-to-stand motion. This work provides references for stable standing up analysis and robot assisted rehabilitation.

The 3D microtopography of Micro Wire Electrical Discharge Machining (MWEM) surface is characterized by isotropy. The methods of roundness and second order of principal section are introduced to describe the complexity degree of MWEDM surface texture more precisely and simplify the calculation `process. The Gaussian curvature is adopted instead of zenith curvature to characterize the concave convex character of the surface, which can make the calculation not being limited by the measuring direction and improve the calculation precision. The surface geometry elements are cut apart by watershed method to simplify the characterization parameters and realize accurate counting of the surface peaks. It shows that, for estimating the isotropy MWEDM surface, the improved structural parameters of 3D roughness are more simplified and the characterizations are more precise.

The rigid body dynamics is widely used in current research of the digital prototype of agricultural equipment without considering the mutual coupling problem between the large displacement of component movement and small displacement of the structural deformation. For the complex dynamic characteristics of steering system in high clearance boom sprayer, a digital prototype of steering system is established with flexible part of steering hydro-cylinder fixation. The influence of the flexible-body fixation in the steering system on its performance is researched. The kinematics and dynamics characteristics of the rigid-flexible coupling influence component movement are revealed. Compared with whole rigid components, the results show that the maximum angular displacement, angular velocity and contact force at hinged place are increased dramatically considering the rigid-flexible coupling. Analysis results of equivalent stress and translational displacement show that the strength of the flexible fixation meets the application requirements.

A finite element (FE) model and a theoretical calculation method for the tensile-shear strength of clinched joint were proposed. First, a 2D model for clinching process was established in ANSYS to simulate the performing process and analyze the metal flow law; the simulation results are in accord with the test results. Then, a 3D model for tensile-shear process was constructed from the results of the 2D model; the load-displacement curve and the failure mode were obtained. The simulation results are in good agreement with test results with an error of 3.4%. Finally, theoretical formulas for the calculation of the joint strength was proposed based on the two failure modes in tensile-shear tests, which are neck-fracture failure mode and button separation mode. The formulas were verified by the tests of twelve types of clinched joints. Results show that the calculation error of the joint strength is 8.9%.

In order to study the fabrication method and fluidic flow characteristics of bio-micro-channel, a novel measuring method of quasi-static axial tension (QSAT) is proposed for the compound of biological soft tissue with the characteristics of silkworm's anterior silk gland (SASG). The anterior silk gland of fifth instar silkworm is taken as the test material. The experiment of QSAT is conducted using electronic single fiber machine. The fracture morphology of the sample is observed by SEM. The QSAT performances of SASG and fibrosis process of silk protein solution, called as silk fibrosis, are measured. Using hyper-elastic Ogden model, the QSAT mechanical constitutive equations of SASG and silk fiber are established. Based on the model, the fitting curves of experiment data are obtained by ABAQUS with the correlation coefficient of above 98%. Results show that hyper-elastic mechanical property of both SASG and silk fiber is excellent, meanwhile silk fiber outperforms SASG.

Three-roller edge pre-bending of thick plate is affected by non-uniform material strength distribution along the thickness introduced by multi-pass roll forming and heat treatment. Hot-rolled Q235 plates with thickness of 100 mm and 200 mm were investigated by tensile testing of round bars from equal spaced multi-positions along the thickness. The material stress-strain curves of tensile bars were achieved by tensile testing on WAW-E100 machine at a strain rate of 3.0 × 10^-5 s^-1. The stress-strain relationship of the material considering the non-uniform distribution was obtained. Three-roller edge pre-bending process of thick plate on CCDW11XNC-300\420×2500 mm heavy machine with bottom rollers horizontally adjustable was analyzed and an edge pre-bending model was developed. Numerical results by this model show that a strength variation of 0~20% along thickness could result in top-roller force variation of 0~13%, little increase in springback ratio, and no effect on pre-bending angle before unloading.

Considering the latest developed Friction Stir Welding (FSW) robot, a set of structure design analysis process is established for large-scale equipments of heavy-loading and high precision. The FSW robot is characterized by large volume, complicated structure and extremely bad working conditions. In order to improve the accuracy and evaluate the mechanical property of the welding joints, the mechanical model of the FSW head is built and the welding process under five working conditions is simulated. The obtained forces and torques can be taken as the boundary conditions for robot strength and stiffness analyses; and the analysis results can be used to verify the rationality of structure design of the robot. Simulation results show that the stiffness performance of the robot is good under the most severe welding condition of melon-segments configuration; all the functional components are safe and reliable, and the structure design of FSW robot meets the welding quality requirement. The results may provide references for further lightweight design and mechanical performance optimization of the FSW robot.

This paper proposes a trajectory tracking control method for nonlinear and uncertainty nonholonomic mobile manipulator control system based on adaptive fuzzy control and non-singular terminal sliding mode control. First, a dynamic model for the nonholonomic mobile manipulator is established. The Fuzzy Gaussian Potential Function Neural Networks (FGPFNN) is employed, because of its nonlinear approximation performance, to compensate the uncertainties of unknown parameters of the system optimally. Second, the non-singular terminal sliding mode control is proposed in order to eliminate the influences of unknown outside interference and approximation errors on the system. Finally, the stability theorem of the control system is obtained by applying the Lyapunov stability theory. Simulation results demonstrate the effectiveness and superiority the proposed control method.

With the gradual development of network application fields, the attack patterns have reached their delicacy and multi-steps from the coarse and simplistic pattern in their early days. In order to redeem the flaws of intrusion detection technology, an intrusion alert correlation model based on data mining and ontology (IACMDO) is proposed. IACMDO deals with underlayer alert through cluster and classification, and builds attack knowledge model by ontology, realizing the detection, tracing and predicting against multi-steps attack. The performance of traditional IDS is upgraded through simulations of KDD Cup 1999 and DAPRA 2000 datasets, which verifies the efficiency of the proposed alert correlation model.

Finite Impulse Response (FIR) digital filter design is actually the design of unit impulse response coefficients combination, which is essentially a combination of multi-parameter optimization problem. On this basis, an improved particle swarm optimization algorithm based on opposite learning strategy (OPPSO) is applied to design linear phase low-pass FIR digital filter. OPPSO increases the probability to find the optimal solution by solving the inverse solution of the particle position, thus, to find a set of optimal combination of parameters of the filter. Meanwhile, a more realistic sense of zero phase filters close to the ideal filter is introduced into the experiment for comparison. Experimental results show that the proposed algorithm has good optimal results for linear phase low-pass FIR digital filter design.

The moving trajectory by random sampling distributes unevenly in time dimension. After projecting the three-dimensional spatiotemporal trajectory data into one-dimensional time domain, a spatiotemporal hot spot region discovery and moving pattern mining methods are proposed based on automatic detection of intensive time intervals. Through detecting intensive time intervals dynamically with a bottom-up clustering strategy, the spatiotemporal hot spot regions are discovered in corresponding time intervals. A depth-first algorithm is designed to mine the set of frequency moving patterns. Finally, based on synthetic moving trajectory dataset, the effectiveness and scalability of the proposed algorithms are verified.

To improve the efficiency of deep neural network distributed training, a new method is proposed, which makes neural network model running on distributed memory computing system. A framework of distributed memory is built, which contains functions of data partition and multi-task schedule. It can avoid the impact of I/O on the training rate and makes the training process run at memory-speed across cluster. Within the framework, multiple model replicas of deep believe network are trained in an asynchronous way. In addition, the dropout algorithm is employed to prevent over-fitting. The proposed method is evaluated using CIFAR-10 dataset. Experiment results show that the new method improves the efficiency of training deep neural network and enables scalability.

The dynamic multi-factor combinations and k-Nearest Neighbor (KNN) classification method are adopted to designedly and in-depth analyze each node terminal. The feature nodes are classified according to their similarities. According to the classification strategy the usages of all nodes are organically sorted. The priority is given to the nodes of high usage efficiency and the easily idle device nodes, thus, to realize the network load balancing. Experiment results show that the proposed method can greatly improve the acceleration ratio, expansion ratio and data expansion rate.

Traditional cloud media resource allocation algorithm lacks service satisfaction and distribution utility etc, which results in the low efficiency of cloud media allocation and resource availability. In this work, first, the utility function of cloud media service provider and requester is introduced, which is uniformly described in three aspects: price, response time and bandwidth of the service. Then the cloud media resource is negotiatory allocated based on the utility function values of both sides of cloud media service and concession-making strategy, which maximizes the utility of the cloud media service resource. Finally, a Resource Allocation Algorithm based on Negotiation Mechanism (RAANM) is proposed on the basis of the negotiation utility model. Comparing with traditional resource allocation algorithm, RAANM aims to maximize the utility instead of to minimize the longest finishing time, thus to enhance service satisfaction. Simulation results demonstrate the efficiency of the proposed algorithm.

A Request Migration based Task Scheduling algorithm (RMTS) is proposed. When a new request is received, the scheduler running RMTS calculates a probability based on the current system load and request migration path length to determine whether migration is needed. If migration is needed, the optimal migration path is selected. With one video server failure, the scheduler transfers all the requests in service to another server to ensure service uninterrupted. The request success rate and service delay are used to measure the system performance. Experiment results demonstrate that the proposed RMTS performs better than traditional request migration algorithm that, with RMTS, the request success rate and service delay are improved by 14% and 15% respectively.

An Equivalent Cable Bundle (ECB) method is developed in this paper. First, a cable observed by simulation is taken as an individual group, and all other conductors of the complete cable bundle are sorted into four groups with respect to the modulus of the terminal loads. Then the unit length inductance matrix and unit length capacitance matrix, the equivalent resistance of the terminal impedance of the simplified ECB model are obtained; the position coordinates of the cable harness in the equivalent model are calculated. Finally, the finite difference time domain algorithm is used to calculate the results with these parameters. The efficiency of this method is verified by numerical experiments of automotive cable harness consisting of 14 conductors. Results show that the computing time for the terminal induction voltage of the integrity models of cable 1, 4, 8, 11 and 14 is all about 11 s; while the computing time of the above terminal induction voltage of the simplified ECB model is all about 5.5 s, that the computing speed is increased by 50%. The ECB method and the finite difference time domain method satisfy the calculation precision and improve the simulation efficiency of automotive cable harness time domain radiation sensitivity.

In tilting sampling the relationships between the selection of tilting angle and the spectral aliasing as well as the resolution are ambiguous. To overcome this problem, the source of the spatial aliasing is analyzed based on sampling theory, and an Aliasing Index (AI) is designed, which measures the extent of aliasing. In addition, the effective spatial resolution is introduced to analyze the relationships among the titling angle, spectral aliasing and effective resolution. Results show that, under the same condition, within the tilting angle range from 57° to 89°, the aliasing and spatial effective resolution in tilting sampling are higher than that in normal sampling; the tilting angle corresponding to the minimum aliasing is 78°, and that corresponding to the maximum spatial effective resolution is 75°. Curve fitting shows that the spectral aliasing and the effective spatial resolution are power-law correlated. It is helpful to understand the relationship between aliasing and resolution with various titling angles for practical application to meet specific need.

In order to improve the demultiplexing performance of the mode division multiplexing system, which has Differential Group Delay (DGD), the MIMO equalizer based on Constant Modulus Algorithm (CMA) is used to demultiplex the received signal, in which the SMA is used to iterate and update the tap coefficients of the MIMO equalizer. Simulation results show that the algorithm can successfully demultiplex the received signals with crosstalk and delay and has good convergence. When the Optical Signal Noise Ratio (OSNR) in the system is greater than 18.25 dB, the Bit Error Rate (BER) of the system can reach 10^-5 order magnitude, which meets the requirements of the communication system.

A packet-level queuing system with time-varying Markov modulated service rate was established to model the process that network serves Machine to Machine (M2M) Small Data Services (SDS) when M2M Surveillance Video Services (SVS) use network bandwidth preferentially. The main goal was to investigate the service capability of the network for SDS and the restrictions of SVS on SDS. Stochastic process theory, queuing theory and probability theory were used to analyze the environmental state probability space of the multi-state Markov-modulated Poisson Process (MMPP) attached to the random arrival process of SDS. Then first step analysis and Laplace-Stieltjes(LS) were applied to analyze the complex random service process of SDS under the multi-state MMPP random environment process, and the average and variance of the service time for SDS packets were calculated. Numerical simulation results demonstrate the impacts of burstiness, arrival rate and service rate of SVS on the service features of SDS.

To solve the difficulty and complexity of carrier frequency estimation of multi-h Continuous Phase Modulation (CPM), a non-data aided carrier frequency estimation algorithm for multi-h CPM was proposed. This algorithm is based on existing carrier frequency estimation algorithm only for single-h CPM. The carrier frequency estimation was obtained using the maximum likelihood principles. Then, a modified Cramer-Rao bound for multi-h CPM was deduced for evaluating the performance. Simulation results show that the proposed algorithm is suitable for full response and partial response multi-h CPM; it has good performance in carrier frequency estimation and is not sensitive to the timing offset and modulation index offset.

The outage performance of multiuser transmission network with an Energy Harvesting (EH) cooperative relay is investigated; where a source transmits independent information to multiple destinations with the help of energy constrained Amplify-and-Forward (AF) relay. The relay can harvest energy from the radio frequency signals transmitted from the source, and it helps the multiuser transmission only with the consumption of the harvested energy. First, an EH Cooperative (EHC) protocol is proposed on the basis of the power-splitting relay receiver architecture, which enables the simultaneous information processing and EH for the multiuser transmission system. Then, the outage performance is theoretically analyzed, and an explicit expression of the system outage probability is derived. Finally, the effects of the source power and relay location on the system performance are discussed. Numerical simulation results demonstrate the accuracy of the analytical results, and reveal that the proposed EHC protocol outperforms traditional non-cooperative scheme in term of outage probability. Besides, deploying the relay close to the source can obtain lower system outage probability.

In order to resolve the problem of signal reconstruction error caused by observation noise in Compressed Sensing (CS), a method is proposed to reconstruct error vectors using the theory of non-correlation constraint as the condition. In this method, a linear programming for error vectors is established based on the annihilator property of sparse check matrix; and compressed sensing reconstruction for Low-Density Parity-Check (LDPC) codes is implemented. Under an additive white Gauss noise channel and primal-dual interior point algorithm, simulations show that three selected (LDPC) codes all satisfy non-correlation constraint condition with high reconstruction ability. Moreover, the correlation coefficient of MacKay random codes is smaller, therefore 100% success rate is obtained in reconstructing error vectors with SNR of -1 dB. The simulation results also show that under the condition of satisfying the requirement of Bit Error Rate (BER) performance, high reliability communication can be achieved with low SNR by the proposed CS-LDPC codes.

The complicated computation has always been a serious problem in the parameter estimation of near-field sources. Based on the symmetrical array structure, a new computational efficient method for Range-of-Arrival (ROA) estimation of near-field sources is proposed. First, the polynomial is constructed through the symmetrical array structure, and the angle information of the near-field sources is obtained by solving the roots of the polynomial. Estimation of the range information is combined with the idea of compressed multiple signal classification (MUSIC). Dividing the Fresnel range into sub-intervals, the new spectral function is obtained by constructing the intersection of these noise sub-spaces. Thus, the range search is carried out in the small area instead of the whole area, which reduces the processing time. Simulation results show that the performance of the proposed algorithm is better than that of existing algorithms.

Most existing edge detection methods can not perform well in low contrast image and noise images. Combing the advantages of differential operator and fuzzy edge detection, a novel multi-level fuzzy edge detection method based on weighted neighbor-region is proposed. First, this method computes the image gradient features, and utilizes the adaptive method to divide image into multiple tiers based on gradient. Then, a fuzzy function is constructed to strengthen different image gradient features in different levels. Experiment results show that the proposed approach can highlight the gradient features in low contrast region of an image with the help of strengthening. Also this approach outperforms the state-of-art methods in terms of both visual quality and noise (Salt and Pepper noise and Gaussian noise) suppression.

In traditional Spatially Selective Noise Filtration (SSNF), the correlation is weakened by wavelet coefficient shift among different decomposition scales, and the wavelet coefficients are seriously affected by the noise at fine scales. To resolve these problems an improved Spatially Selective Noise Filtration (ISSNF) algorithm is proposed. The relative shift of wavelet coefficients between two adjacent scales is deduced, and then a shift correlation algorithm is proposed. Considering that the higher the scale is, the stronger the correlation of wavelet coefficient is, and that divergence happens during decomposition, a reverse processing method is proposed, in which constraints are provided from higher scale to lower scale. According to statistic principle, the wavelet coefficient of the highest is processed by threshold filtering. The performances of the proposed algorithm, the traditional algorithm and two existing improved algorithms are compared by processing Blocks signal and Heavy Sine signal. Moreover, the above filtration algorithms are applied in group delay estimation of certain GEO satellite. Results of Monte Carlo simulation and processing of measured data demonstrate that the performance of the proposed algorithm is better in noise reduction.

In magnetic sensor system there are errors (caused by nonorthogonality error, different sensitivity error, and zero-shift errors of single triaxial magnetometer) and misalignment errors of the coordinate systems of triaxial magnetometers. To overcome this problem, a calibration method of magnetic sensor system based on linear error model is proposed. For single triaxial magnetometer, a linear error model about nonorthogonality, different sensitivities and zero-shifts is set up; and for misalignment of coordinate systems, a linear error model about rotation angle of coordinate systems is set up. The error parameters of the magnetic tensor system can be got by generalized inverse. Calibration of the magnetic tensor system can be realized by putting these error parameters into the measuring data. Measuring experiments prove that the calibration method can effectively reduce the errors of single triaxial magnetometer and misalignment errors of the coordinate systems of triaxial magnetometers.

To analyze the technology innovation ability of China's auto industry, an analytical model is developed based on the theory of technology innovation. Using this model, the performances of technology innovation ability in auto industry, household audio-visual equipment industry and telecommunication equipment industry are evaluated. The model adopts a new and more sophisticated two-line approach based on Data Envelopment Analysis (DEA) and Principal Component Analysis (PCA). Furthermore, six factors are introduced into this model. The evaluation results show that the technology innovation ability of China's auto industry is in the low level, and it confronts insufficient innovation input and unbalanced innovation output. At the same time, the innovative transformation process faces inefficiency. Finally, several policy suggestions are proposed on improving the technology innovation ability of China's auto industry.

The particle size distribution and stability of three fat replacers (WPC 7.00, WPC 8.50 and WPI) are studied, and their effects on the viscosity, rheological properties and sensory qualities of low fat milk are analyzed. Results show that the sizes of 72.4% particles of WPD 7.00 fall in the scope of 647.9±291.7 nm, which could imitate fat to produce similar consistency and greasy feel. The average Zeta potential of WPD 7.00 is -45.5±9.71 mV, indicating a stable solution. In terms of the viscosity, rheological properties and sensory qualities there is no significant difference (P>0.05) between the whole milk and low fat milk with 75% fat replaced by WPC 7.00.