Abstract:A fast fault location mechanism on memory data error, which is called FFLM for a self made CMP-16 multi-core processor’s functional validation, is proposed and realized. FFLM builds a multi-port golden memory model based on the hardware emulation accelerator. It monitors the packages of memory access between memory system and processor cores during the emulation, real-time compares the data from real memory system being tested and the data from golden memory model, judges whether they are consistent, and finds the errors once any wrong data is sent to processor core from memory system. Compared with traditional ways, FFLM has the advantages of fast emulation speed, low hardware cost and low fault Location time cost. Statistical results from the emulation for a self-made CMP-16 multi-processor show that FFLM improves the speed of date fault location in memory system by 6.5 times averagely.

Abstract:Buffer resources are key components of the on-chip router, and their structures exert significant influence on the performance and power consumption of the interconnection network. In general, asynchronous FIFO based on shift registers is adopted to implement on-chip buffer resources. Packets transmitted traverse the FIFO queue step by step, leading to higher propagation delay of packets and larger transition counts in the circuit. In this research, an asynchronous FIFO based on hierarchical bit-line buffer is proposed, and then a new asynchronous on-chip router is presented in detail. Compared with the traditional asynchronous router, the newly presented one has lower hardware complexity and power consumption. Experiments show that the new router can achieve 39.3% area saving and 41.1% power reduction when the depth of asynchronous FIFO is configured with 8.

Abstract:Coherence messages in token protocol are always broadcast, which creates heavy network traffic and limits its scalability. An efficient predictor technique was used to eliminate the useless message in token protocol, which employs a predictor to predict the data’s sharers and only sends coherence messages to the sharers to avoid broadcasting. Three predictor policies depending on the write miss and read miss ratio of application were proposed. The results show that: the owner predictor with 512 entries can reduce interconnect traffic by 3.8% on average, the sharer predictor and the hybrid reduce interconnect traffic by the average of 11% and 7% respectively. The method proposed efficiently reduces the traffic and improves the scalability of token protocol. 

Abstract:A graph coloring based management optimizing algorithm for cache, namely Cache Coloring, has been proposed. This algorithm first partitions the data into several data objects according to their memory accessing behaviors. Then it partitions the cache into a pseudo register file with alias according to the size of the data objects. Each pseudo register in this register file can hold one of the data objects. Finally, it uses an extended graph coloring register allocation algorithm to determine the position of each data object in the cache and their replacement relationship. The data object partitioning divides the management of cache into two levels, one for the coarse-granularity management of the data objects in the compile-time and the other for the fine-granularity management of the cache lines in the run-time. So the advantages of both compiler and hardware are exploited. Cache Coloring is implemented in GCC. A hardware simulation platform which supports Cache Coloring is built based on the Simplescalar processor simulator. The primary experimental results show that Cache Coloring can exploit the locality well and reduce the cache miss rate.

Abstract:To reduce the power, the energy-efficient embedded microprocessor always adopts the distributed and hierarchical register file structure (DHRF). Many data need to be stored in the second level general register file (GRF) because of the small capacity of the tiny operand register file (TORF), and this challenges the design of compiler. A new VLIW scheduling algorithm is proposed to solve the problem through analyzing the program characteristics. The variables are detected while the compiling and virtual copy operations are inserted at the appropriate time. Through instruction scheduling and communication scheduling for the copy operations and constructing new data transfer route including GRF for the global variables and software pipelining variables, which have great demand for the register, the pressure on TORF is transferred to GRF. The experimental results show that the VLIW scheduling algorithm is consistent with the starting point of the energy-efficient microprocessor. On the condition of 8% program performance decline, the energy consumption on register accessing is reduced by about 51%, and the energy consumption of the processor is reduced by about 43%. At the same time, the burden of programmer is avoided.

Abstract:Bubble flow control is an efficient technique to avoid deadlock in torus networks without using virtual channels. If a flow control mechanism has knowledge of buffer utilization within a dimension, it can make resource allocation decisions based on global network conditions to improve network performance. The previous critical bubble scheme has a risk of blocking. To resolve this problem, a false packet protocol was presented, and a non-blocking moveable bubble scheme was designed, which is an improvement of critical bubble scheme with a requirement of one packet buffer at least, which halves the buffer requirement of two. Network simulation results show that this scheme is apparently better than the existing methods, avoids permanent blocking, displays a throughput improvement of more than 20% under distribute, hotregion and uniform traffic patterns, and maintains a steady throughput after network saturation without sharp drop.

Abstract:To meet the stringent requirements on both high-throughput and high-flexibility for FFT computation in modern wireless communication applications, an application specific reconfigurable architecture called ASRA is presented. ASRA is a VLIW-like static multi-issue processor with custom multi-grained reconfigurable fabric as extensible function units. The run-time context manager in ASRA offers multi-grained Custom Instructions selection and is bound to the appropriate reconfigurable fabric while considering run-time changing scenarios. Mapping of the FFT algorithm on ASRA is studied and optimized. ASRA employs on-chip scratchpad as fast local memory. The read/write operation and butterfly operation is pipelined to support hardware pipeline execution of a loop kernel. Experiment results show that ASRA achieves a high performance improvement and a good trade-off between area and performance. 

Abstract:Targeting at the mapping of key loops onto CGRA (Coarse-Grained Reconfigurable Architectures), this research proposes a novel approach called memory-aware kernel loop pipelining mapping (MALP). The RCP_CGRA model and the critical loop mapping formulation were shown first. Based on polyhedral model, then the array clustering and data domain partition were presented. An analysis of the critical loop storage requirement was described. Based on this analysis result, the MALP provided an efficient way for loop mapping under the resource constraints of CGRA. Experiment results show that MALP can improve the data throughput rate while costing less resource. MALP makes the loop mapping on CGRA more efficient.

Abstract:A novel motion object detection method in H.264 compressed domain is proposed, based on multi-features accumulation and analysis, which can reflect the motion of object and change in the area of borders. Firstly, the MVs of macroblocks were accumulated and filtered. The Jacobian matrix was used to describe the relationship between parameters of global motion and MV in each macroblock, which makes the computation of global motion parameters easier than before. Then the motion areas were detected by the similarity of local motion and global motion. Secondly, the remarkable blocks of DCT energy were accumulated in temporal by the reliable motion around them. Then the border and texture area was found by the accumulations of DCT energy with the local self-adaptive threshold which selected by entropy. Finally, the results which had been detected by MV and DCT respectively were combined. The experiments shows that the method proposed can detect the motion object in compressed video accurately, and the result of can achieve good recall and precision simultaneously.

Abstract:As the parallel computing technologies are becoming mature, the research on parallelization of geospatial raster data processing algorithms has been a hot spot issue. Focusing on the complicated algorithms whose processing procedures consist of multiple computing steps, this research proposes a task scheduling method based on the theory of spatial computational domain with which the task scheduling solution is not static, but adjusts itself as the algorithmic procedure proceeding. Experiments have verified the effectiveness of our method. Because the method keeps adjusting the task scheduling solution at every computing step, the load-balancing effect is better, and thus the parallel running time is shorter compared with the conventional task scheduling method.

Abstract:In a wireless sensor network, some sensor nodes will consume their energy at a much higher rate and die much faster if the energy dissipation of nodes is unbalanced. Accordingly, the network will lose its ability prematurely. A novel clustering data aggregation algorithm is proposed to resolve the problem of unbalanced energy dissipation among nodes during data gathering in clustering wireless sensor networks. The algorithm divided the network into rectangular grids with unequal size and made cluster heads rotate among the nodes of each grid respectively in terms of residual energy. A grid whose cluster heads consumed more energy is bigger and has more nodes to take part in the cluster head rotation for sharing energy load. By this way, the algorithm is able to balance the energy dissipation of nodes. In addition, considering that wireless sensor networks are energy-restricted because the nodes are equipped with limited and non-rechargeable energy resources, the algorithm adopted a series of measures to save energy. The results of simulation show that the algorithm is able to achieve better performance in aspects of efficiency of energy usage, lifetime of network and balanced extent of energy dissipation.

Abstract:Coded exposure photography makes the ill-pose motion deblurring problem well-posed, and searching for the optimal code has been the key problem in coded exposure photography. For practical purpose, an optimal code searching method for coded exposure photography in the presence of CCD sensor noise is proposed. Firstly, the sensor noise in the process of imaging was systemically analyzed based on the affine noise model. Then, the basis of Raskar’s code was theoretically analyzed for the first time and based on the signal to noise ratio (SNR) gain of deblurred image, an analytic formulas between the photon noise and the structure of the optimal code was presented. Finally, based on the noise calibration of real coded exposure camera, the criteria for the optimal code was presented, a fitness function was designed and the optimal code using the genetic algorithm was obtained. Experimental results with synthetic and real data demonstrates the validity of the proposed method.

Abstract:Sosemanuk is one of the stream ciphers that have been selected into the final portfolio for the eSTREAM project. Sosemanuk from the view point of bit-level was described. By studying the S-box of Serpent1, the modular addition, and the linear feedback shift register, a kind of equation between the bits of the internal states was obtained, based on which an improved word-oriented guess-and-determine attack on Sosemanuk is proposed. This improved attack utilizes the Groebner basis method to solve the equation system and can determine the other 5 words of the internal state by guessing just 7 words, and thus the complexity is pvoved to be O(2192).

Abstract:For the multiplicative quaternion-error attitude determination approach, the attitude sensors for Euler angle measurements are studied. Two different methods were used to demonstrate that the measurement sensitivity matrix of Euler angle error with respect to error quaternion solved in many references is false. The correct measurement sensitivity matrix is presented and the reason for this mistake is analyzed in detail, and the numerical examples are given for the validation and verification of the conclusions.

Abstract:Aimed for the question of computational efficiency during conceptual design stage, a rapid engineering aero-heating calculation method for complex shaped hypersonic vehicles is established. Firstly the pressure distribution along vehicle’s surface was calculated by modified Newtonian theory. Secondly, the streamline that passes through the target point was calculated with Newtonian steepest decent concept. Then by using the fitted function of high temperature gas parameters, reference enthalpy method and engineering calculation function of aero-heating, the heat current of target point was presented. Finally, the heat flux on the surfaces of blunted cone, lifting body and wave-rider vehicle was calculated The analysis result shows the method used in this paper is fit for complex figure, and can satisfy the aero-heating calculation during conceptual design stage in both efficiency and precision.

Abstract:Understanding drop formation of gel fuel is the first stage to conduct experimental investigation on combustion characteristics of single gel fuel droplet, and the formation of liquid drops at the end of a capillary tube is one of conventional methods. In order to investigate the formation of gel propellant drops at the end of a capillary tube, the Navier-Stokes equations were solved, and the VOF (volume of fluid) method is employed to track the gas/liquid interface during process of drop formation. The variation of dimensionless neck diameter DN/D0 and dimensionless drop height L/D0 with time was analyzed, which agreed with the existing experimental results and validated the numerical model. The results show that drop is not always at the steady state; there is high shear rate at the drop neck, leading to dynamic viscosity decreasing and drop pinch-off; then drop pinch-off, the part of conjoint drop, coalesces into drop rapidly, and high positive velocity appears; however, the part of conjoint capillary tube coalesces into gel propellant at the end of capillary tube rapidly, and the high negative velocity appears.

Abstract:The active evasion strategy of spacecraft is discussed when the distance between spacecraft and target is relatively short. Firstly, the relative dynamic model is analyzed. The relative motion can be decomposed into two submotions, viz., the relative motion in the instantaneous rotation of the line of sight (IRPL) and the rotation of this plane. Secondly, the optimum evasion direction of the spacecraft is proposed, through the analysis of the relative motion in IRPL. Thirdly, the determination of this direction in the inertial space is discussed. Finally, the numerical simulation is conducted with different maneuver accelerations and different detective distance of the spacecraft. The effectiveness of the active evasion strategy is demonstrated with the simulation results.

Abstract:Based on time-varying characters of spacecraft propulsion system and stochastic properties, the semi-qualitative fault isolation method which combines the Z-test and Analytical Redundancy Relations (ARRs) has been proposed. Firstly, based on the insight of state spatial alternation, it is achievable to mining time-invariant structural characters space using ARRs for Spacecraft Propulsion System which is built on the Diagnostic Bond Graph, and by the tendency analysis of the residuals of ARRs, the time-invariant fault signature matrix with change direction can be established. Secondly, the Z-test for robust qualitative residual generation to established observed signature against infection from multi-uncertainty is used, and then faults are isolated by comparison of observed signature and fault signature. Through the analysis of isolation results of a spacecraft propulsion system, it shows that the method is valid for overcoming the challenges from establishing time-varying diagnostic criteria and the jamming from stochasticity not only in the system model, but also in the measurements, which improves the robustness in fault isolation.

Abstract:The solution efficiency for the traditional structure dynamic optimization process is relatively low. Based on a discussion of the parameters optimization for the ring truss deployable antenna structures, this research chose the sample points by design of experiment method in the variable parameter space. The response values of the first order natural frequency, total mass, and the maximum deformation with specific loading for the ring truss deployable antenna structures were obtained by the finite element method. The Kriging models, which were established by the relationship between the sample points and response values, were used to find the optimum solution with optimization algorithm. It is required that the optimum solution should satisfy all the constrained conditions. The results showed that the response surfaces of the first order natural frequency, total mass, and the maximum deformation with specific loading for the ring truss deployable antenna structures were nonlinear relationship with the design variable. The structure optimization method with the Kriging approximate model can be used to find the optimum structural parameters of the ring truss deployable antennas accurately.

Abstract:Forces and dynamic equations of supercavitating vehicles were analyzed in detail in the paper; the mathematic model of the optimal diving trajectory was established in the vertical plane. Considering the time-delay effect during the expansion and contraction of supercavity, the Time Delay Gauss Pseudospectral Method(TDGPM) was introduced. Based on the information of discretized nodes, the TDGPM interpolated the real time cavity shape for each node and calculated the dynamic equations, which were used as constraints during the optimization process. Numerical solution of optimal diving trajectories and corresponding maneuvers were obtained. Results show that supercavitating vehicles without fins can achieve stable and maneuverable flight by applying proper cavitator control. Results also indicate the feasibility of TDGPM and its fast convergence.

Abstract:Carbon fibers reinforced silicon carbide composites (C/SiC) are the most promising materials for next generation rocket engine components due to their unique properties, such as low densities and extremely high temperature resistance. The intensity and wall thickness design of C/SiC composites combustion chamber is one of the key tasks for rocket engine design. Based on the thin shell theory and the fourth intensity theory, this research deduced an ideal thickness calculation formula for composites chamber, which was used to calculate the wall thickness of a certain type of rocket engine combustion chamber. The validity of the formula was checked through subscale C/SiC tube by burst-test and thrust chamber by hot fire test. The formula and results proposed in this paper can be widely applied to the design of the wall thickness of C/SiC composites combustion chamber.

Abstract:Mining-Task-Machine is the key technology of the Ocean Mining System. To finalize the design，It is the key step to test the traction and control unit of the Mining-Task-Machine by measuring kinematic parameters with high precision. Mining-Task-Machine moves in a large area in water, a binocular measurement method with wide-angle lens is used to make real-time track and measurement of the trajectory of Mining-Task-Machine. Firstly, the lens distortion is calibrated by using gridding board in lab. Secondly, according to the natural feature, the parameters of the lens distortion is modified based on the result in lab and all the camera parameters are calibrated. Finally, the binocular system tracks a cooperative feature on Mining-Task-Machine and measures the motion trajectory including position and velocity, etc. of it on real time. The residual between the result of the system and the total station is about 5 mm. The system has passed the examination and evaluation, and has been put into use.

Abstract:We present a new method to detect the dim changing target for the image sequence of outside stereo scene obtained from the airborne camera. In order to compensate for the movement of the platform, a cascading image registration mode is proposed, which firstly uses the biological vision method to register the images to be detected coarsely and fast, and which then uses least square match based on invariant coefficient to register the images accurately. In order to compensate for the gray-scale changes between the images, a new histogram similarity transformation is proposed, which is ?able to handle not only linear gray-scale changes but also nonlinear gray-scale changes. In order to reduce the stereo effect, a sub-region division method is proposed, which divides an image into several sub-regions, and then calculates the affine transformation parameters for each corresponding sub-regions independently. Using this scheme, we can make sure that different sub-regions correspond to different geometric transformation parameters. Experimental results using many real images obtained from electro-optic pods clearly demonstrate the correctness and validity of the method.

Abstract:The main problem of military antagonism decision is its difficulty to confirm the strategy set of adversary. This research analyzes the military decision under the uncertainty of strategy set based on hypergame. The decision approach of high-level hypergame was proposed after analysis the belief relationship of each level. The result of high level was put back to low-level to refine the decision based on the analysis of the possible decision of adversary. Consequently, this makes decision of friendly forces more pertinent and reasonable against to adversary. Finally an example is used to illustrate the approach.

Abstract:The mission of Earth observing satellite (EOS) is to acquire photographs of specified areas on Earth surface at the requests of users. The goal is to select a feasible task sequence to maximize the sum of weights. This research presents the mission planning problem of the next-generation agile Earth-observing satellite(AEOS). The complex user requests(including multi-strip area, real time download request, and stereoscopic request) and complex satellite constraints were considered, covering eight satellite actions(including observe action, data download, SSR Erase action, attitude movement, heliocentric pointing, geocentric pointing, and instrument action). A chronological look ahead algorithm was designed to solve the problem, heuristic rules based on expert knowledge were used to make choices and arrange satellite actions, which satisfy all satellite physical constraints and operational constraints. For the current experiment instances and applications, the algorithm can give results in very short time. Experiment results suggest that our algorithm works well for the agile earth-observing satellite planning problem.

Abstract:Large-scale space development project (LSDP) is a kind of project which possesses some special characteristics and is significant for the social economy development and national security. Risk analysis is an important guarantee to accomplish these projects designed with high quality and low cost. According to the characters, management process and risk factors of the LSDP, using the theory and technology of Multi-Agent and Disturb Project Management, a new risk analysis method of LSDP based on Multi-Agent technology is proposed. It is applied to a large-scale space development project. The risk level of the project is evaluated and the key risk factors are identified. These results can provide decision information for the project risk control and management.

Abstract:Maximum average correlation height (MACH) filter is formulated by linearly combining the training data, which is statistically optimum and fairly robust for finding targets in clutter when the Gaussian assumption holds. This research proposes a nonlinear extension to the MACH filter by correntropy function which can induce a new feature space. Thus it is possible to construct linear filter equations in the new space, and the proposed filter has an improved performance due to the nonlinear relation between the feature space and input space. The algorithm is applied to synthetic aperture radar image recognition and exhibits better performance under peak-sidelobe-ratio and receiver-operating-characteristic criteria.

Abstract:Blind recognition of convolutional codes is widely used in the fields of information interception, intelligent mobile communication and multicast communication. In order to solve the problem of fast recognition of convolutional codes, the classical Euclidean algorithm is improved, and a method for blind recognition of convolutional codes based on improved Euclidean algorithm is proposed. Code rate is searched first, the check polynomial of convolutional codes is solved through iterative process, and the convolutional codes can be identified quickly. Validity of the algorithm is verified by the simulation results, and computational load is less than the algorithms in the literature reviewed.

Abstract:A new radar signal, the chaotic based random stepped frequency signal, is proposed because of characteristics of chaos such as pseudorandom behavior, sensitivity for initial condition, accessibility for generation and control, and easy realization and processing for the stepped frequency signal. A signal model was introduced, then the ambiguity function and the cross correlation function were analyzed theoretically in detail, and two different echo signal processing methods were proposed finally. Numerical simulations based on four typical chaotic maps and two random series show that the designed signal is effective, and promising for radar application.

Abstract:Code tracking loop is a key component of Global Navigation Satellite System (GNSS) receivers which determines the accuracy of positioning performance. So analysis for its theoretical performance has been a research hotspot for decades. Many productive conclusions have been derived. However, most researches consider the impact of band limited channels in frequency domain, which is too computational to be applied in practical situations. To lighten the computational burden, an analytical expression of code tracking accuracy with correlation functions was derived, and the performance of code tracking loop could be easily assessed with the derived expression, which has guiding significance in the design of GNSS receivers.

Abstract:Oil leakage from cracks is a kind of common fault in mechatronics systems. Responses of artificial olfactory system to the volatile organic compounds (VOCs) emitted by leakage oil can be used for leak detection, which is a novel non-destructive method for fault diagnosis. An artificial olfactory system was applied to detect three different kinds of leakage oil such as diesel oil, machine oil and gear oil. The collected dataset was arranged in a-three dimensionality matrix (sample×time×sensor). Three methods as two-way principal component analysis (PCA), three-way unfolding PCA and parallel factor analysis (PARAFAC) were adopted to distinguish the VOCs. The results showed that PARAFAC was superior to the other two methods because PARAFAC took into account the true three-dimensionality structure of the dataset. The multivariate calibration method, principal component regression (PCR) was applied in the prediction of different concentrations of diesel oil. The results indicate that it is possible to use an artificial olfactory system to give an alarm by setting a threshold. 

Abstract:Module identification plays an important role in modular design and mass customization. In order to further extend the quantitative methodology for product modularization, an approach to product granulation based on the theory of fuzzy quotient space was proposed. The traditional product modularization based on the relativity analysis was translated into the product granulation by introducing the granular computing method, and the product granulation was analyzed by the theory of fuzzy quotient space. The product granular space was established based on the normalized distance and the integrated fuzzy similarity relation, and the optimal granular layer was obtained using a two-stage optimization algorithm. Then the optimal modularization scheme was achieved. A case of the telescopic shuttle mechanism of automatic storage/retrieval machine was studied to illustrate the validity of the proposed method. This approach offers a new way of numerical analysis and evaluation for module identifying and optimizing, and the application result shows that the proposed method is feasible and rational, which can guide the product modularization process effectively.