A complicated regenerative damper model containing regenerative motor and its control model was established to investigate the influence of inertial mass on damping characteristic. Analyzed by structural vibration method, the suspension natural frequency equation and the transfer characteristics of sprung/unsprung mass to road excitation were obtained. Alter the ratio of inertial mass to unsprung mass and analyze its influence on amplitude-frequency characteristic of body acceleration、suspension deflection and wheel dynamic load. The results show that the resonance frequency of regenerative suspension is prior to conventional suspension and the main weakness of inertial mass influence is that it increases the transfer characteristic of higher resonance frequency, although has little improvement for lower resonance frequency. With higher excitation frequency, it affects the ride comfort while has little influence on safety.

To realize precise control of brake pressure, the current-response characteristic of high speed switching electromagnetic vale was studied. For brake pressure fine regulation, the characteristics of braking hydraulic system must be known. By results analysis of the static overflow and current dynamic response tests of inlet valve of Anti-lock Braking System (ABS), the static and dynamic characteristics of the inlet valve was obtained. Taking a pressurization rate control as example, a method of pressure fine regulation was expounded. Considering the method for wheel cylinder pressurization in ABS control, the characteristic test of inlet valves in hydraulic unit was proposed. This can provide guidance for production and assembly test of inlet valves. The characteristics of inlet valves are required to get a good consistency to ensure the portability of ABS software after calibration.

A characteristic-based real-time dynamic steering system model was built. The model is based on the steering angle ratio, torque ratio and softness of Ackermann mechanism. It does not need a lot of parameters of hardpoint positions and is suitable for building model at the automotive predevelopment stage. The dynamic degrees of freedom (DOFs) of the key parts of the steering system were taken into consideration in the model, which reflect their dynamic behaviors. The steering model was embed in the simulation code of a 14 DOFs vehicle model, and an A-class vehicle handling performance simulation was carried out. The simulation results were compared with field test results, which verified the effectiveness of the developed model.

To realize the simulation of vehicle motions, a Hardware-in-the-Loop (HIL) test bench was constructed. The working principle of the simulation of vehicle motion signal was analyzed based on the constructed HIL test bench. A main loop controller with Finite State Machine (FSM) and servo loop control system of motor was designed. In order to prove the efficiency of the HIL test bench and FSM control algorithm, experiments with typical working condition was carried out. The signals collected from the sensors installed on the test bench were compared with target signals,and the results demonstrate effectiveness of designed control system in following the vehicle state.

This paper introduces the variable density based topology optimization into the design of transmission housing structure. The optimization takes the constraints on manufacturing condition into consideration. In order to increase the strength, stiffness and nature frequency of the transmission structure and control its weight, the static and dynamic topology optimizations of the transmission housing are carried out. According to the results of the topology optimization and the manufacturing constraints, a new rational transmission housing structure is designed. The stress analysis and modal analysis of the new design are carried out. The analysis results show that not only the stress level of the designed housing is lower than the material yield stress, but also the modal frequency avoids the powertrain system resonance.

To increase the continuous driving mileage of four in-wheel-motor driving vehicles a control strategy of regenerative braking was proposed based on the analysis of ideal braking force distribution of traditional vehicle. In this strategy, the distribution of braking force and the characteristics of the in-wheel motors are taken into consideration. Under the condition of safe braking performance, the braking force was distributed to the four wheels to recover more energy during braking process. A simulation model was built up using the software CarSim and Matlab/Simulink. Simulation results show that using the proposed control strategy the electric vehicle can efficiently distribute the braking force and recover braking energy.

Numerical simulation was used to calculate the flow field around a heavy-duty truck. This study attempted to reduce the drag by adding four rear-end panels with different sizes and styles. The influence of the rear-end panels on the flow field was studied, and the mechanism of generation of drag and cause of decrease of drag were investigated. The air velocity was set to 25 m/s without considering crosswind to simulate the real working condition. The results confirm the effectiveness of all the four types of rear-end panels in drag reduction, and a maximal drag reduction of 2.8% was obtained. Analysis suggests that the rear-end panels alter the structure of the ring vortex in the wake flow, thus reducing the drag.

When a high-powered bulldozer works under hydrodynamic conditions for a long time, the temperature of the transmission oil will excessively raise and the transmission efficiency will decrease. To solve such problem, the dynamics simulation model and efficiency control model of the hydrodynamic mechanical transmission system were proposed based on the research of it dynamics model. Then the Isight and SimuLink integrated simulation platform was built up to obtain the optimized parameters of PID controller using the design of experiment with Latin Hypercube method. Simulation and experiment results show that with the efficiency control of the transmission system using working delay action and heavy load prejudging techniques, the torque converter can work in its high efficiency region, and the overheating of the transmission oil can be effectively eased, therefore, the working efficiency of the bulldozer is obviously improved.

The working principle the load-sensing full hydraulic system was introduced in this paper. The co-simulation model of the hydraulic steering system of a loader was established based on AMESim and LMS Virtual.Lab Motion. The dynamic characteristics of the steering process was simulated and analyzed. Comparison of simulation and experiment results shows that the co-simulation model is accurate and credible. The multi-body dynamic model can accurately simulate the system load. It is found by simulation analysis that tire cornering stiffness significantly affects the pressure oscillation of the steering system, thus reducing the stiffness can reduce the pressure fluctuation.

To study the robustness of commonly used turbulence models in flow field simulation of hydrodynamic torque converter, fifteen turbulence models were taken to conduct sampling and calculation in a torque converter flow field simulation system of based on ANSYS/CFX. Using Monte Carlo method, the input variables were slightly disturbed using a normal distribution function. Each turbulence model was sampled 200 times, respectively in low and high calculation speed conditions. According to the complexity measurement theory, the complexity of the acquired input and output variables was analyzed. The robustness of the simulation system and its output variables were calculated, by which the quantitative indicators of robustness for each turbulence model were given. Analysis results show that the shear stress model has the best overall robustness in torque convert flow field simulation.

An experiment investigation was carried out to study the effect of fuel volatility on particle matter emission in common-rail diesel engine, and analyze the characteristics of particle size distribution under different load conditions. The results show that refining the fuel volatility can increase the nuclear state fine particle number concentration in low and medium load condition, reduce the emission fine particle number and mass concentration in high load condition, but has little influence on the accumulative state fine particle number concentration. For different volatile fuels, the fine particle number concentration in nuclear state reaches the highest point at the equivalent ratio of 0.4 with the same fuel in different load conditions. In low and medium load condition, adding small amount of volatile fuel into the fuel with poor volatility will result in sharp increase in the nuclear state fine particle number concentration. With further improvement of the fuel volatility when the adding proportion increased, both nuclear state fine particle number concentration and the total fine particle number concentration decrease in low and medium load condition, but the accumulative state fine particle number concentration rarely changes. For high load condition, with fuel of medium volatility, boiling point of 263 °C, both the nuclear state fine particle number concentration and total fine particle number concentration are the highest. For fuel of better volatility with boiling point of 244 °C, both the nuclear state fine particle number concentration and total fine particle number concentration are the lowest.

State Preference (SP) survey was carried out to obtain the travelers' behavior data. To analyze the factors that influence the departure time choice behavior, an overall object model and a typical object model were proposed using the mixed logit model. The proposed models were verified by different ranges of the travel behavior data. The results identify the factors that influence the departure time choice behavior. These factors include the social-economic characteristics of the travelers, the travel delay information, the familiarity of road network, the weather information, the quality of travel information and the travel purpose. There exist certain differences between the overall object model and the typical object model. The results of this study reveal the impact of the traffic information on the departure time choice under different environments, and provide the basis for the dissemination strategy of traffic information.

An integrated model was proposed for the optimization of signal timings for main-signal and pre-signal and their coordination on the approach of intersection with two stop lines and pre-signals. A set of constraints were taken into account in the model, including the number of lanes before the pre-signal and main-signal, the saturated flow of each movement, the cycle length and the length of sorting area. The queue evolution process of different movements (through movement and left turn movement) in the sorting was also addressed in detail. The proposed model is validated by a case study of a signalized intersection using field data. For a given traffic demand, the proposed model can meet the same degree of saturation with less total green time compared with traditional methods.

A method of using Highway 3D Dynamic Sight Distance (H3DSD) to represent the available sight distance was proposed. The geometric data was used as the main data source for H3DSD calculation. The data was used to fit the highway alignment with cubic B-spline curves. Considering the influencing factors of combination of horizontal and vertical alignment, the driver's dynamic visual field, and the illumination angle of the vehicle head lamps, the H3DSD of each point along the centerline roadway could be calculated with spatial geometry. The most important constraint during the calculation was the relationship between the computed angle composed by the line from the position of driver's eyes to the sight distance checking point and the centerline of the driver's visual field, and the driver's maximum dynamic visual angle. Moreover, the criterion of H3DSD was proposed in order to identify the segments with poor sight distance. Also the main factors restricting sight distance were analyzed. A case study was carried out and the results show that the proposed method has the advantages of easy to get the source data and good prospect of application. It can provide significant technique support to highway safety design.

In light of the noisy chaotic characteristics of traffic flow, a new multi-step traffic flow prediction model based on wavelet and echo state network was proposed. Utilizing multi-scale decomposition method of wavelet, the proposed model restricts the interference of noisy components to the dynamic behavior of the traffic flow; meanwhile it extracts the chaotic low-frequency component, which possesses most of the energy of traffic flow. In predicting the multi components concurrently, the strong prediction capacity of echo state network for chaotic low-frequency component was utilized effectively to ensure the accuracy of multi-step traffic flow prediction. The results of prediction of the real traffic flow in Xizhimen Bridge of Beijing show that the prediction accuracy is significantly improved by the proposed multi-step model in comparison with the traditional echo state network model. Under the condition of high prediction accuracy, the maximum predictable step is also increased by the proposed model.

Currently it is hard to get real data for emergency evacuation and rescue work under major disaster. To overcome this situation, a calculation model of the link travel time and route travel time was proposed based on the BPR (Bureau of Public Roads) function. Then a prediction model of emergency evacuation time under major disaster was built. Under the background of road network in De Zhou city of China, the software TransModeler is adopted to build interregional road network and simulate major disaster. The data of emergency evacuation time obtained with TransModeler is used to check the proposed model. Results show that the proposed model has higher precision and fidelity, which can offer data support to emergency evacuation and rescue work when major disaster occurs.

The specific heat and thermal conductivity of rubber-steel composite materials were described using a mixing rule model proposed by the authors. The K-R kinetics model was used to calculate the curing reaction heat. The user subroutine was compiled based on nonlinear finite element software ABAQUS and the integration simulation of tire vulcanization process was achieved. Then the curing process of radial tire 11.00R22.5 was simulated by this analytical method. Simulation result was compared with experiment result and with the results obtained by other researchers. The accuracy and applicability of the proposed method was confirmed by the comparison. Finally, the variations of the temperature, induction time and the curing-state were analyzed during the curing process. The last curing point of the tire was found. These findings can provide theoretical guidance for the design of tire curing process.

Based on the nonlinear fatigue damage model and damage accumulative model of materials, the damage elastic modulus of concrete and the damage residual area of steel reinforcement bars and prestressed tendons were introduced into the section stiffness calculation. The stiffness degradation model of the prestressed concrete beam under multilevel amplitude cycle loading was deduced, and genetic algorithm was applied in parameter determination. Then, the neural network method for forecasting the fatigue stiffness degradation was proposed and Matlab program code of the calculation model was generated. Experiment of prestressed concrete beam under multilevel amplitude cycle loading was carried out to establish the parameters, improve the calculation accuracy and verify the calculation results. It is suggested that the stiffness degradation of the prestressed concrete beam under multilevel amplitude cycle loading could be divided into three stages. The model calculation results agree well with experiment results, with a deviation within 5%. The genetic algorithm based model parameter fitting method and the neural network based forecasting method can give accurate results, thus providing useful reference for the estimation of fatigue resistance of prestressed concrete beam.

Based on the basic principles for producing Reactive Powder Concrete (RPC), a new kind of RPC with compositions of superfine cement, Portland cement, admixtures (ground granulated blast furnace slag and fly ash), sand, water, superplasticizer, steel fibers, without silica fume was produced and tested. Experiment results show that it is feasible to produce RPC without the admixture of silica fume. The strength, fluidity and cost of RPC are satisfied with appropriate proportion of superfine cement and admixtures. The superfine cement in RPC can be partly replaced by Portland cement to reduce the cost. The effect of the amount of steel fibers on the mechanical properties, such as compression resistance, breaking resistance, fluidity and flexural toughness, of the RPC was analyzed.

As reasonable interval between guides during two-roll straightening process was not easy to determine, a 3D elastic-plastic finite element model of the straightening process of bar material was proposed. Based on this model, variations of straightening force, supporting force of guides, degree of plastic deformation, residual stress and the maximum deflection of the bar after straightening were analyzed in different intervals. Reasonable range of interval between guides was determined by comprehensive consideration. The obtained range of interval by this study is larger than previously used range, and the frequency of change of guides is decreased.

To meet the requirements of electromagnetic sounding for emission in the near-surface frequency domain, a System on a Programmable Chip (SOPC) was built using Field Programmable Gate Array (FPGA) as the hardware control platform. The SOPC was based on the unipolar and frequency multiplication Sinusoidal Pulse Width Modulation (SPWM). Verilog was used to realize multi-frequency digital SPWM algorithm, generate the multi-frequency digital drive signal. This signal has a small frequency interval and low Total Harmonic Distortion (THD). It can overcome the defects of other methods, such as the fixed frequency interval and complex system structure of pseudo random method, the over-dependence on load coil and complex algorithm of the bit-stream method. Theoretical simulation and experiment results verify the effectiveness of the method.

A global sliding mode observer for induction motor based on high-order sliding mode was proposed. The model is applied for precisely identifying the rotor velocity and the rotor flux of the induction motor. A global sliding mode surface was designed to improve the robustness of the observer and to implement the observer in the sliding motion through the entire response. A high-order sliding mode control law was utilized to obtain the smooth equivalent control signal directly. The control signal of the sliding mode observer can be applied to state observation, so that the better observing precision is realized. Simulation and experiment results show that the proposed sliding mode observer can eliminate the chattering phenomenon effectively, provide high observing precision. Moreover, the proposed observer is robust to load disturbance and internal parameter perturbation.

In this work, eight types of segmentation measure and four types of interregional distance measure were proposed. Based on these measures 9-intersection model was extended. The measure value was used to replace the corresponding element in the 9-intersection matrix to show the measure property; the measure and the topology were combined together. The extended 9-intersection model can express both topology and measure information simultaneously, while enhances the expressive ability of the interregional relationship, furthermore, improves the expressive ability of topology. It also improves the accuracy and efficiency of spatial data querying, meanwhile, it more fits human recognition.

During the process of wireless video transmission based on IEEE802.16d protocol, how to improve the Quality of Service (QoS) architecture was investigated to meet the needs of real-time data transmission that overcomes the shortage of existing QoS architecture. Improvements were proposed by designing a new QoS scheduling algorithm on the basis of existing QoS architecture. According to the characteristics of four categories of service flow, which are supported by IEEE802.16d, the appropriate scheduling algorithm for each category of service flow was selected to ensure correct data transmission. Simulation results of Unsolicited Grant Service (UGS) and Real-Time Polling Service (RTPS) service flows verify the advantages of the new designed QoS architecture.

When vehicular nodes in Vehicular Ad-hoc Network (VANET) visit the Internet to obtain services, traditional wireless coverage area (e.g. access point) is limited, thus blocking the vehicular nodes from the Internet to enjoy services. Based on the OPNET network simulation platform, this work expands the previous Mobile Ad-hoc Network (MANET) gateway model. The revised gateway is able to establish a connection with the Universal Mobile Telecommunication System (UMTS) core network, and thus the nodes in VANET can access the Internet through UMTS network. Therefore, the shortcoming of traditional VANET can be overcome by the existing large coverage area of the UMTS.

A rotation-invariant algorithm use for image template matching is proposed. This algorithm utilizes the rotation invariance of the image gradient amplitude, the rotation migration of the image gradient angle, and the distribution symmetry of the autocorrelation coefficient of the image gradient angle. The algorithm is divided into two steps to complete precisely and rapidly statistical and matching calculation based on three types of different characteristics of the images. Simulation results show that the algorithm can precisely locate the template image, which is with arbitrary angle. The algorithm has good real-time matching performance and less time-complexity. Even in the case of nonlinear transformation with image grayscale, image scaling and image occlusion, the template matching operation using this algorithm has strong robustness.

Virus analysis is the process of determining the purpose and functionality of a given virus sample. The current problem is that virus dynamic analysis tool can only analyze a single program execution in single computer system, thus the error rate is high. A new analysis model is proposed, which can use abundant resources of cloud computing system. It allows us to explore multiple execution paths and identify malicious actions that are executed when only certain conditions are met. Experiments are carried out in open-source cloud software, Eucalyptus. Results show the proposed model can detect the existence of trigger-based behavior, find the trigger conditions of such hidden behavior and identify the inputs satisfying those conditions. Its performance is superior to single computer system.

With the development of P2P technology, tradition methods based on application port number and matching protocol-specific signatures in payload cannot provide acceptable accuracy any more. In this paper, a content redistribution model of P2P network was proposed. Based on this redistribution model, an identification algorithm was also developed. The proposed model is based on the underlying characteristics of P2P protocol, each peer acting both as a server and a client, so redistributing the received content to other peers. Theoretical analysis and experiments show that the approach is more suitable for identifying traffic that is generated by P2P streaming and highly sought-after resource sharing.

Classical detection algorithms can not describe the defect characteristics in a better way, and it is difficult to classify the defects of similar classes. To solve these problems, an algorithm based on Gaussian Mixture Model (GMM) in nonsubsampled Contourlet domain for fabric defect recognition is proposed. First, the Nonsubsampled Contourlet Transform (NSCT) is used to obtain the representations in multi-directions, multi-scales and translation invariants, and an optimal sub-band is selected by the cost function. The difference of coefficients between defect and non-defect regions of the sub-band is smaller; therefore, the threshold is difficult to choose. The standard deviation method can effectively avoid this problem to get accurate results. After that the feature vectors are calculated by the statistical characteristics of the defect region. Then, the minimum misclassification function is used for joint estimation of the GMM parameters. Finally, the classification is completed by Bayesian classifier. Experiments on defects of nine categories show that the proposed algorithm can achieve better classification accuracy compared with the traditional algorithms.

For 128 Uyghur characters, a handwritten recognition algorithm based on radical dictionary and time division direction feature is proposed. First, the radical dictionary is established by decomposing the Uyghur characters as three type radicals, main, affix and dot. The problem of stroke order and connection can be solved by analysis of the connected and broken strokes. Second, a new online statistical feature, named time division directional feature, is extracted from every radical to reduce the interference from handwritten topological deformation. The statistical feature is designed by dividing freeman codes according to time sequence, with its short term variations being found and adjusted to correct the dither phenomenon. Finally, different classifications are designed for various types of radicals. With the radical coefficient estimated according to the coefficient distribution, the character recognition result is obtained by fusing the outputs of all classifications using the weighted naive Bayesian algorithm. Experiment results show that the algorithm can effectively identify 128 unconstrained handwritten Uyghur characters. An average recognition rate of 93.15% is achieved on the Uyghur character database containing 13056 samples.

Based on the transient analysis theory of 2D Markov chain, an analytical model is proposed to describe the transmission of periodic massage in Controller Area Network (CAN). The statistical data of periodic message transmission in simulation environment demonstrated the effectiveness of the proposed model, and verified the correctness of the assumptions and selected parameters for this model. This model can be used to estimate the blocking caused by time triggered periodic messages in CAN.

According to the optical properties of lenticular lens array, we build a virtual image display platform for one-dimensional integral imaging. Then we analyze systematically the display resolution, the viewing angle and the depth of focus, which can describe the display characteristics. Based on this, a novel method is proposed for reconstructing the viewpoint image during the one-dimensional integral imaging display process by computer simulation. According to the laws of light propagation in lenticular lens, the images, seen from some viewpoint, corresponding to every lenticular lens can be obtained and used to make up a viewpoint image using the proposed method. Experiment results show that the obtained viewpoint image can perfectly match that from the display platform and truly describe the one-dimensional integral imaging display process. The obtained parameters of the display characteristics can be used as the theoretical basis when designing one-dimensional integral imaging system.

An improved Pulse Coupled Neural Network (PCNN) model was proposed, where the linking matrix was optimized with Receptive Field (RF) model. The PCNN model possesses both directionality and scales, and can simulate the segmentation function of visual cells more efficiently. Experimental results demonstrate the effectiveness of the RF-PCNN model in segmenting vehicle images and segmentation results have high boundary detection rate. The problems of under-segmentation and over-segmentation in vehicle plate region are well solved.

Singular Value Decomposition (SVD) is a widely used nonlinear signal denoising approach. Its denoising effect is depending on the selection of singular values for the signal reconstitution. The singular value difference spectrum of singular value can be used to select the expected singular values adaptively. However, sometimes when dealing with signals with strong trend, such as aircraft engine health signals, it may hardly choose the right singular values and fail to restore the desired signals To solve such problem, a signal denoising approach based on Empirical Mode Decomposition (EMD) and SVD is proced. First, the original signal is decomposed by EMD and the trend component is extracted from the original signal. Then, SVD is used to denoise the detrended signal, and the difference spectrum of singular value is used to select the singular values adaptively. After denoising, the detrended part is added to the trend part to get the final signal. Results of Both simulation and real aircraft engine health signal experiment demonstrate that the proposed approach can precisely select the singular values for signal reconstitution and effectively reduce the noise.

To balance the contradiction between the robustness and invisibility of the spatial digital watermarking, a multi-scale contextual discontinuous measure was proposed as new perceptual-mask function to determine the texture and edges of image. Combined with the human visual system the mask function was applied to spatial digital watermarking. In the embedding process, the mask function was employed to control the position and intensity of water mark, resulting in self-adaptive embedding of watermark signal. In the detecting process, the inhomogeneity and image diffusion theory were combined together to evaluate the watermark signal and to make prediction, so that the blind examination is realized. Simulation results show that under the same robustness requirements, a watermarking scheme using a mask via inhomogeneity has a better visual quality. Compared with other methods, the extracted signal of the proposed scheme has higher accuracy under the same signal processes or geometric attacks.

The problem of collision caused by hidden nodes in wireless network is one of the main reasons that lead to the decline in network performances. An existed strategy for hidden node collision avoidance was improved. The improved strategy optimizes the regional grouping method, and adjusts the structure of IEEE802.15.4 super-frame. Then a more flexible time slot allocation algorithm was proposed. The simulation results indicate that the improved strategy can effectively reduce network end-to-end delay and packet loss rate, and improve QoS performance.

A double-queue single-server and limited-cache packet-level vacation queuing model, IBP+MMBP/Geo/1/K, with threshold was established. According to the respective features of different services, which have various arrival distributions to define vacation strategies, the vacation time of H2H service is equated to the busy time of M2M service. In order to seek unified standards to evaluate the vacation strategies, a performance function E to evaluate the Quality of Service (QoS) of different services was defined. At the same time, the influences of different M2M services on the vacation strategies were also studied. Finally, various performance indicators and the change trend of the performance function of the system were given through simulation. The results of this study can be used to guide the radio resource management and the service configuration problems.

The current soft load balancing algorithms in heterogeneous wireless networks do not have a universal wireless resource statistics, have not consider the matching difference between customers' business needs and the resource units. According to grid computing idea, a load balancing algorithm based on grid over heterogeneous wireless networks was put forward. The algorithm firstly maps the radio resources into different three-dimensional resources, which are equivalent to three-dimensional grids. The flow is divided into different subflows (i.e.data packets), and then distributed to different grids. In the condition of guaranteed QoS, networks load balancing can be achieved. Simulation results show that grid-based load balancing algorithm can achieve good load balancing among heterogeneous networks, reduce the load deviation, and improve system throughput.

For efficient network resources dispatching in heterogeneous wireless network scenarios, a Quality of Service (QoS) aware load-balancing algorithm was proposed. Based on the characteristics of different wireless services, the algorithm defines a utility function for each terminal to represent its QoS experience and a utility function for each Radio Access Network (RAN) to represent its load level. The definitions apply to all heterogeneous RANs. The algorithm conducts an iterative load-balancing procedure, where it always tries to handover a terminal, which is with low QoS utility and resource utilization served by the busiest RAN, to another RAN, which has the lightest load among the RANs. This RAN with the lightest load is capable of increasing the terminal's QoS utility, until network load balance is reached. Such a procedure can help to realize better wireless resource deployment and ensure terminal QoS. Simulation results show that the proposed algorithm can effectively increase the throughput, reduce the time-delay and packet loss rate of the network.

A routing strategy based on event-driven clustering and delay grades tree for large-scale emergency monitoring Wireless Sensor Network (WSN) was proposed. In order to save and balance energy consumption of the whole network, when choosing the cluster head this strategy takes several factors into account, including the residual energy of a node, the hops from the node to the sink node, the connectivity of the node with its neighboring nodes, and the number of its father-nodes. In addition, the delay grades tree and the multi-path selection guarantee the real-time and reliable data transmission. Simulation results show that the strategy can improve energy efficiency and prolong the life of WSN up to 200% against LEACH protocol, 140% against AEEC algorithm, and 15% against ARPEES.

In order to improve the network efficiency with limited resource, i.e. energy, capability of calculation and memory. A cross-layer algorithm of resource optimization for Wireless Sensor Networks (WSN) is proposed, which is based on network utility and network lifetime. The energy control in physic layer, the access control in Mac layer and the traffic control in transfer layer are considered together to represent the optimization problem. The vertical decomposing is used by introducing congestion price and energy price to update the node based energy consumption, link access probability and rate. The series of solution is distributed, and simulation results prove the convergence of the resolution.

Traditional Doppler algorithm is not robust to the fast-varying Under Water Acoustic (UWA)Orthogonal Frequency Division Multiplying (OFDM) systems. To overcome the shortcoming, an efficient Doppler estimation algorithm was proposed for the system. First, the sample rate conversion was completed using oversampling and linear interpolation methods. Then, the accurate Doppler shift was acquired by combining the candidate value of Doppler factor. Simulation results show that the proposed algorithm can not only ensure the system performance, but also reduce the computational load of the receiver. Therefore, it is suitable for high-speed real-time UWA OFDM systems.

Decorrelation method based on whiting transformation is proposed to improve the system demodulation performance with nonorthogonal pulse group. Simulation results demonstrate that the performance of the proposed method is better than that of matching filter. Comparison with traditional decorrelation method, the proposed method is better when Signal to Noise Ratio (PNR) is low, while it is worse when SNR is high. Based on the above properties, a new decorrelation method integrating the whiting transformation based method and traditional method is proposed. The performance of this new method is better than either of the two methods.

In order to reduce the encoding complexity of Low-Density Parity Check (LDPC) codes, a method of constructing Quasi-Cyclic LDPC (QC-LDPC) codes with total linear encoding complexity and a short-cycle optimization algorithm for eliminating short cycles were proposed. Based on the short-cycle optimization algorithm, the constructed QC-LDPC codes can achieve good cycle distribution. Simulation was carried out under the Additive White Gaussian Noise (AWGN) channel. The simulation results show that the proposed QC-LDPC codes outperform the IEEE802.16e standard QC-LDPC codes in both error performance and encoding complexity.

The eigenvectors of cross-correlation matrix for Prolate Spheroidal Wave Function (PSWF) pulse set was used as the weighting factors, and the original PSWF pulses were added by the factors to generate new orthogonal PSWF pulse set. Simulation results show that the orthogonality performance of the proposed method breaks through the restriction on the pulse number by traditional Gram-Schmidt method. The high energy concentration and Bit Error Rate (BER) performance of the PSWF pulse set are maintained after orthogonalization process. Meanwhile, the method is easy and fast for hardware implementation.

Magnetoresistance measurement of indium tin oxide/N, N'-Bis(naphthalen-2-yl)-N,N'-bis(phenyl)-benzidine(NPB)/ Tris(8-hydroxy-quinolinato) aluminum(Alq₃)/3-(4-Biphenylyl)-4-phenyl-5-tert-butylphenyl-1,2,4-triazole(TAZ)/LiF/aluminum OLED structures have been made as a function of magnetic field at room temperature. The MR is defined as MR=Δρ/ρ=[ρ(B)-ρ(0)] /ρ(0). It has been found that that the Δρ/ρ increases with increasing magnetic field strength from 0 mT to 110 mT with TAZ thickness of 0 nm at 10 V, which maximal value is 8.22% at about 110 mT. The external magnetic filed can decrease the Δρ/ρ in devices with TAZ thickness from 5 nm to 15 nm. This is because the magnetic field can reduce the singlet polaron pairs converted into triplet polaron pairs due to Zeeman splitting and can lead to an increase in singlet polarons and a decrease in triplet polarons. Furthermore, the Δρ/ρ increases while TAZ thickness is increasing at same magnetic filed strength. When an external magnetic field is 110 mT, the Δρ/ρ is found to increase by up to -16.92% with TAZ thickness of 15 nm at 10 V.

An adaptive threshold method was proposed for phased array 3D sonar systems. First, the beams of were classified into several edge modes and the beam-strength based confidence image was modified by checking the neighborhood symmetry of each beam. Next, the principle of maximum entropy was introduced to search the optimal adaptive threshold for the new confidence image. Experimental results show that the proposed method can distinguish the objects and backgrounds effectively while requiring similar computation as conventional spatial filter methods. This threshold method is suitable for real-time applications in phased array 3D sonar systems.