最新刊期
卷 36 ， 期 8 ， 2008
- 摘要：A new algorithm for distortion correction based on pixel-movement model(PMM)in angiography system is proposed.The parameters of PMM are derived by analyzing the pixel movement induced by the distortion(calculated by a single-image distortion correction method)in a series of phantom images acquired by a certain interval of orientation angle.The model allows distortion correction of angiographic images acquired in any arbitrary C-arm orientation.Experimental results demonstrate that the proposed algorithm is fast with high accuracy for distortion correction of angiography images.关键词：distortion correction;image intensifier;X-ray angiography2|962|1更新时间：2025-07-16
- 摘要：Single-layer homojunctional organic light-emitting diodes based on tris-(8-hydroxylquinoline)-aluminum(Alq3)as emitting layer were fabricated using vacuum deposition method.The device structure is:indium-tin-oxide(ITO)/Alq3(x nm)/Mg:Ag.According to trapped charge limited current(TCLC)theory,the current conduction regime in OLEDs devices with different thin film thicknesses was investigated systematically by varying the thickness of organic functional layer.A fitting analysis of device current via TCLC theory was adopted to study the relationship between experimental data and theoretical prediction.The results demonstrated that the current of single-layer device with Alq3 thin film has a change from ohmic conduction region,TCLC region to TCLC-SCLC transition region;while the current of device with relatively thick film has no SCLC region due to the increase of traps in the Alq3 layer.The numerical fitting is very beneficial for Alq3 film thickness optimization during device fabrication process.关键词：organic electroluminescence;current conduction;trapped charge limited current;film thickness;numerical fitting2|1667|1更新时间：2025-07-16
- 摘要：A sequence of decimal numbers generated the chaotic piecewise linear map is used to define a bijection map;two sequences of decimal numbers individually generated by two chaotic piecewise linear maps are used to determine the noise vectors by comparing the element of the two sequences;three operations(bit-by-bit exclusive OR,module multiplication,module addition)in the group and the permutation determined by the bijection map are alternately applied on plaintext with block length 64n bits to produce ciphertext blocks of the same length.Lastly,the performances were analyzed from the theory and experiments.关键词：chaotic map;block cipher;permutation;group theory2|1020|2更新时间：2025-07-16
- 摘要：In this paper we propose a novel integrate-and-fire(IF)neuron model with exponential synaptic conductances that can be simulated exactly.The postsynaptic potentials and spontaneous discharge statistics of the new model are compared with those of commonly used models,such as the leaky IF model with instantaneous synaptic interactions or the passive membrane equation(PME)model with exponential synaptic conductances in which conductances are explicitly integrated.The proposed model is much closer to the PME model with respect to the spiking response dynamics,while still being nearly as computationally efficient as simple leaky IF model.Then we present an event-driven simulation strategy for the new model.Using event-driven and clock-driven simulation strategies we simulate random network with dynamic synapses,the results indicate that(1)the simulation time scales linearly with the total number of spiking events in the event-driven simulation strategies and(2)the temporal precision of spiking events impacts on neuronal dynamics of network in the different simulation strategies.关键词：integrate-and-fire neuron;synaptic conductances;event-driven;clock-driven;simulation2|1213|4更新时间：2025-07-16
- 摘要：This paper presents a GA-based decision-tree algorithm to deal with SVM multi-class classification problem.First,GA is used to create optimal or near-optimal decision-tree automatically,which makes the margin between two classes maximal at every decision node.Then at every decision node,standard SVM is used to make binary classification.Finally,the SVM decision tree achieves multi-classification.Theoretical analysis and experiments show that the proposed method is more precise than the traditional "DT-SVM" and "DAG-SVM" methods,and has higher training and testing efficient than classical "1-a-1" and "1-a-r" methods.关键词：support vector machine;genetic algorithm;decision tree2|2283|20更新时间：2025-07-16
- 摘要：CABAC is the entropy coding method adopted in H.264/AVC main profile,and the combination of RDO and CABAC can represent an improvement in bit rate of around 20%.During the RDO mode decision,entropy coding is performed for every candidate encoding mode,which increases the computation complexity of video encoding drastically.To avoid the expensive computation of Lagrange costs,this paper proposes a transform-domain bit-rate estimation method for CABAC for the first time,in which the bit-rate is estimated based on the quantized transform coefficients and the motion vector differences.After that,a fast mode decision algorithm based on the bit-rate estimation technique is presented,in which the bit-rate estimation is performed instead of actual entropy coding in mode decision,thus reducing the computing time of H.264 encoder.Simulation results show that the proposed transform-domain bit-rate estimation method for CABAC achieves about 51% computation time reduction of Lagrange cost for mode decision and saves about 33% total encoding time with FS(full search)motion estimation algorithm,while the losses of PSNR and bit-rate are negligible.关键词：CABAC;entropy coding;bit-rate estimation;RDO mode decision;Lagrange cost2|1085|1更新时间：2025-07-16
- 摘要：We propose a new approach to track memory reference trace and present a hardware-based Memory Trace Tool(MTT)for real systems.This paper details the design of MTT,including sampling configure,translating address,outputting trace and analysis of memory trace,which implement a complete solution to acquire and analyze memory reference trace.Compared to existing methods,MTT has several features:(1)be transparent to applications and system software;(2)be nearly zero-overhead and no memory pollutions;(3)be available to obtain full-system trace on real systems;(4)provide various online analysis approaches;and(5)be independent of system platform,support different CPU and OS platform.关键词：memory reference trace;profiling;memory access behavior analysis;field-programmable gate array(FPGA)2|1423|1更新时间：2025-07-16
- 摘要：Based on conventional design concept of the fan,the procedure of the parametric design and numerical simulation with application to the axial-flow fan is developed.The procedure and aerodynamic design process of the CPU fan are embodied into parametric design and numerical simulation software utilizing VB language as well as axial-flow fan aerodynamic design software,which can generate high quality HEXA mesh automatically.The numerical simulation can also be performed continuously by batch program and the results about performance curve of the fan can be characterized without resorting to anyone.At last geometry parametric investigations achieved quickly by the procedure are carried out to illustrate the effect of various design parameters on the performance of the fan by choosing one parameter at a time as the control variable,which can provide a more in depth understnding of the flow characteristic of the fan and deep understanding for the optimum of the fan.The procedure can be adopted widely to help engineers to improve work efficiency in practice.关键词：axial-flow fan;aerodynamic design;parametric design;computational fluid dynamics2|3125|0更新时间：2025-07-16
- 摘要：As feature size shrinks and the frequency increases,on-chip L2 cache with large capability is the main source of leakage power in micro-processors.Supporting not only the State-Preserving but also the State-Destroying low leakage power mode,two leakage power controlling policies for L2 cache are proposed.One is the Conserved State-Preserving and State-Destroying (C-SP&SD) policy;the other is the Speculative State-Preserving and State-Destroying (S-SP&SD) policy.Only one copy of a data block in cache hierarchies is in the active mode,the others are turned into the low power mode.The State-Destroying mode with least leakage power should be turned into at most as possible if no obvious performance cost occurs.Compared with the traditional leakage power controlling policies for L2 cache,C-SP&SD policy can save the leakage power of L2 cache more efficiently without obvious performance cost;S-SP&SD policy can save the most leakage power of L2 cache and reach the best energy efficiency of whole processor.关键词：level 2;cache;leakage;power2|793|1更新时间：2025-07-16
- 摘要：The existing Environmental-Stress-Screen(ESS)plans precipitate defects by stressing all products for specified durations.For high reliability and long life products,the plans usually require long screen durations to allow nearly all defective items to fall,and thus generate excessive aging effects of good items.For some products,failures are defined in terms of performance characteristics exceeding their critical values.This paper describes the principles of using degradation measurements of performance characteristics to screen the products.In an ESS,the performance characteristics of defective items degrade considerably faster than those of good ones,yielding a bimodal distribution of the characteristics.As screen duration increases,the 2 modes of the distribution shift apart.It is possible to find a tightened critical value to weed out the defective items before they fail.This paper,based on these principles,by choosing optimal screen durations and tightened critical values,designs the optimal screen plans which meet the screening efficiency and which simultaneously meet the reliability requirement.Because the ESS regime of this paper allows defective parts to be screened out before they fail,the developed ESS plans can shorten screening duration and alleviate aging effects of good products.A numerical example is given to illustrate the effectiveness of the method.关键词：Environmental-Stress-Degradation-Screening(ESDS);mixed degradation path;long life product;random weighted method;good products percent after screening2|891|2更新时间：2025-07-16
- 摘要：A new Hopfield neural networks with dynamical synapses and multiple time scales is proposed in this paper,based on some recent progresses in biological science.Fixed point theorem,inequality technique and Lyapunov functional methods have been used in this paper to study properties such as the global exponential stability,existence and uniqueness of the equilibrium solution for both constant and variable time delays.关键词：dynamic synapses;time scales;neural networks;exponentional stability3|757|0更新时间：2025-07-16
- 摘要：By using the leveled characteristics of fine granularity scalable (FGS) compressed video,the MPEG4 video object coding rules and the improved C&S(chain and sum),the base-level encryption algorithm,which is based on Video object plane (VOP) in Fine-grain extensible coding ,retrieve and encrypt the key data,including shape,texture,motion and global background,of the base-level VOP to encrypt the FGS steaming.The algorithm supports the transform coding of the network node to adapt to the variation of the bandwidth without any encryption and decryption operations.The use of encryption in VOP unit and C&S algorithm effectively reduce the media stream package loss and transport errors.There is no size increment in the encrypted data steam and the variation of the encrypted key withstands the Known-plaintext attack.By the experiments on three series forman,akyio and carphone provided by Moving Pictures Experts Group (MPEG) in a sampling rate of 3 of 300 processing frames,the processing speed of C&S,RC4 and RC5 are 23.5, 64.5 and 42.7M b/s.The results also show a satisfying encrypted security and descrambled visual effect.关键词：fine granularity scalable;scalable coding;video encryption;DRM2|1216|0更新时间：2025-07-16
- 摘要：This paper presents a parallel model termed as towerlike master-slave model (TMSM) for artificial immune systems.Based on TMSM,the parallel immune memory clonal selection algorithm (PIMCSA) is also designed for dealing with large-scale TSP problems.TMSM is a two level coarse-grained parallel artificial immune model with distributed immune response and distributed immune memory.In PIMCSA,vaccines are extracted and migrated between populations rather than antibodies as has been done in parallel genetic algorithms,it is a good balance between the diversity maintenance of populations and the convergent speed of the algorithm.PIMCSA shows superiority over other compared approaches both in solution quality and computation time,and the lager the problem size the more outstanding the predominance will be.TMSM is a good simulation of biological immune system,and PIMCSA is a parallel artificial immune algorithm with good extensibility,which is capable of solving large scale and complex optimization problems.关键词：TSP;parallel artificial immune system;clonal selection;immune memory2|1338|8更新时间：2025-07-16
- 摘要：In overload conditions,the QoS of applications with weakly hard real-time constraint is degraded diversely.To deal with the unfairness case caused by only considering an infinite interval or a fixed finite window loss-rate,this paper brings up a concept with a (p,k)constraint,which is based on variable interval.Based on the (p,k)constraint,an algorithm is proposed,named as AWCS(Any Window Constraint Scheduling).A simple version of AWCS is put forward,which is called KWCS(K-Window Constraint Schedule).Extensive experiments show that KWCS can supersede AWCS,and not only achieve comparative performance but also get lower complexity.The properties of two algorithms are addressed,and a general representation of delay bound of the scheduling algorithms is brought forward.Results show that both AWCS and KWCS can provide better performance than other weakly hard real-time schedule algorithms in heavily overload circumstances.关键词：any window constraint schedule;weakly hard real-time;K-window constraint schedule;minimum success ratio of an interval;delay2|919|4更新时间：2025-07-16
- 摘要：Based on the four-event periodic waveform models,we proposed the timing constraints for dynamic gate,LO-CMOS,NTP dynamic gates and N-C2MOS.Hybrid timing analysis method was employed to calculate the delay of dynamic circuits.We proposed the delay test waveform generation algorithms for dynamic gates,which taked into account multiple inputs switching.The algorithms have been implemented in SpiceTime and applied to the design of a dynamic 32-bit adder.Without false paths,the error of evaluation delay and precharge delay are within 3.62% and 8.26% respectively.关键词：dynamic circuits;timing veri?cation;hybrid timing analysis;test waveform generation2|1014|1更新时间：2025-07-16
- 摘要：A noise variance optimization method is proposed for the time and frequency dimension separate (2×1-D) Wiener-filtered channel estimation of OFDM based systems.According to Wiener-filter theory,the noise variance is necessary to achieve optimal solution.For 2×1-D Wiener-filtered channel estimation,the Wiener-filtering will be applied twice respectively in time and frequency dimension.Hence,the effect of variety of noise variance induced by the first filter should be considered on the second filter in this method,but it has not been considered in the existing 2×1-D Wiener-filtered channel estimation method.This paper presents a novel algorithm which takes into account the effect of variety of noise variance.In the proposed method,the noise variance used by the second filter is optimized according to the mean square error (MSE) of channel estimation by the first filter.The exact MSE of channel estimation is derived in this paper.Moreover,the channel estimation performance is evaluated with different noise variance optimizing criteria.The simulation results show that the performance of the proposed method is better than the 2×1-D filters method without noise variance optimization,and is very close to that of the Wiener 2-dimension filter.关键词：channel estimation;Wiener filter;OFDM;noise optimization2|1800|1更新时间：2025-07-16
- 摘要：It’s pointed out that singular value decomposition (SVD) has the very similar signal processing effect as wavelet transform when Hankel matrix is used.It’s proved that original signal can be decomposed into the linear superposition of a series of component signals by SVD using Hankel matrix,and what the component signals reflect essentially are the similarity extent of original signal to standard orthogonal bases of m and n dimension space constituted by left and right orthogonal matrix in SVD.The similarity mechanism of signal processing of SVD and wavelet transform is analyzed from the angle of basis of vector space and characteristic of Hankel matrix.The orthogonality difference of component signals got by SVD and wavelet transform is studied theoretically.It’s pointed out that SVD also has the singularity detection ability when Hankel matrix is used,but compared with wavelet transform,SVD has two characteristics in singularity detection,one is that the vanishing moment of SVD component signals is increased progressively and the n’th SVD component signal has the n-1’th order vanishing moment so that singular point with different Lip index can also be detected,the other is that the width of the impulse that indicates the position of singular point will keep the same in all SVD component signals and this width is determined by the column number of Hankel matrix.关键词：SVD;wavelet transform;Hankel matrix;signal processing;similarity2|2317|35更新时间：2025-07-16
- 摘要：LDDoS (Low-rate Distributed Denial of Service) attack is a new class of DDoS,which exploits TCP’s RTO (Retransmission Time Out) mechanism.An LDDoS attack can elude the monitor of traditional detection approach by sending low-rate packets in the way of periodic pulse to a victim.This paper proposes an approach of detecting LDDoS attack based on kalman filter.The error between one step prediction and the optimal estimation is used as the detection criterion.Experiments in simulation environment and practical network are conducted to test the detect performance and about 89.6% detect probability is achieved.Results show that this approach has an expected effect in detecting LDDoS attack.关键词：low-rate distributed denial of service (LDDoS);retransmission time out (RTO);TCP;traffic;Kalman filtering3|1017|12更新时间：2025-07-16
- 摘要：Internet has become an essential infrastructure of communication.Limited survivability in the face of failure impairs the performance of Internet.Recent years,fast failure recovery of IP network has drawn much attention.As the state-of-art,there three ways to realize fast failure recovery of IP Networks:(1) to accelerate the convergence of IP routing;(2) to employ the proactive recovery strategy;(3) to increase the speed and accuracy of failure detection.This paper analyzes the defects of the existing solutions.In order to realize fast failure recovery of IP networks,we should:(1) balance the traffic load after recovery;(2) do interoperability test and redesign the routers' architecture.关键词：IP network;failure recovery;routing convergence;backup path;load balance2|1032|6更新时间：2025-07-16
- 摘要：Graphs are widely used to model data in various applications.In recent years,a lot of algorithms have been developed to mine frequent pattern in graph data.We review some representative algorithms in this area.We classify the algorithms based on their properties and mathematical foundation,focusing on the algorithms underlying the approaches and how the algorithms relate to each other.Some typical algorithms are investigated.The comparison is then given based on the properties of the algorithms.New progress and future research directions are pointed.关键词：frequent subgraph;frequent pattern mining;graph match;graph isomorphism2|2841|5更新时间：2025-07-16
- 摘要：Along with the increasing miniaturization and multi-functionality of electronic products,the electromigration of micro-interconnections has become a severe and urgent issue,and turned to be an important reliability and durability concern.This paper presents a brief overview on the electromigrtion problem in aluminum and copper alloy interconnections,and then delivers a review on the electromigration issues in flip chip solder connection which is currently employed widely in mirco-electronic packaging technology.These issues include current crowding effect,joule heating effect,polarity effect,intermetallic compound,electromigration under multi-loads and lifetime of electromigration.关键词：electronic packaging;micro-interconnection;solder joint;reliability;electromigration6|3577|13更新时间：2025-07-16
- 摘要：In order to improve the efficiency of instruction loading into DSP,two techniques of instruction cache optimization on H.264 encoding are proposed in this paper.As the first technique,adjust encoding procedure to prevent the quantity of the instruction codes inside marcoblock encoding loop over the capacity of instruction cache.Thus a new encoding procedure is proposed which fragments the conventional encoding procedure from one macroblock encoding loop at the frame level into three macroblock encoding loops at the slice level,while the results of motion estimation and encoding mode decision are unchanged.As the second technique,compact source code suitably.That is,emphasize adaptively either source code compacting or unrolling according to the length of source codes as well as the quantity of instruction cache miss.关键词：video coding;H.264;media processors;cache optimization2|894|3更新时间：2025-07-16
- 摘要：In this paper,we present two novel algorithms,one for discovery logical topology,the other one for discovering physical topology in heterogeneous IP networks.The algorithm for logical network,which adopt heurism＼greedy algorithms,can be applied to large scale IP network.For the algorithm can self deduce IP addresses,adjust the search domain,it has strong adaptabilities.Another novel algorithm is presented for physical topology which includes switches,hosts and hubs.At the same time,a new data structure named Treelike- Graphics for topology discovery is presented with precise mathematic definition,relationship set and proprietary operation on this data structure.The algorithm,which we present in this paper,does no base on the STP protocol to get the information of topology.So,there are few preconditions to the algorithm.We have implemented the algorithm presented in this paper in Perl in Tsinghua University own research network.The experimental results clearly validate our algorithm,demonstrating that our algorithm can consistently discover the accurate physical network topology with reasonably small running-time requirements.关键词：topology discovery;treelike-graphics;network;data link2|2429|5更新时间：2025-07-16
- 摘要：In this paper,an approach to text categorization named LSASGT is proposed,which combines Latent Semantic Analysis(LSA) with Spectral Graph Transducer(SGT) for the task of text categorization.For both LSA and SGT are originated from spectral analysis theory which can mine some latent structure information within all training and testing data,we integrate them tightly in one model.Firstly,according to the characteristic of natural language,LSA is used to represent documents in a latent semantic space in which documents and their semantic relationships can be reflected more pertinently.Then we construct a graph based on the latent concept-based subspace,and apply the graph into SGT for text categorization.The experiments demonstrate that LSASGT can improve classification performance on both English and Chinese datasets of Reuters21578 and TanCorp-12.关键词：spectral graph transducer;latent semantic analysis;text categorization;graph construction2|1392|3更新时间：2025-07-16
- 摘要：One major problem in semantics-based component retrieval process is the discrepancy between nature language which users are familiar with and formal semantic descriptions of components,which makes users difficult to describe retrieve queries precisely.An interactive query generation method is proposed to solve this problem.The method uses concept identification process to map users’ nature language to ontology descriptions of components,in which concept identification uses conversational process to interact with users.This method extracts implicit semantics in user queries so that it leads to improve recall and precision rates,reduce query effort in component retrieval.关键词：component retrieval;semantic;ontology;conversation;interaction;query generation2|896|2更新时间：2025-07-16
- 摘要：A multilevel security(MLS)model with credibility characteristics was proposed to solve the problem of trusted subjects’ hidden security flaw and poor system usability in present MLS systems.By introducing credibility labels of subjects and objects and credibility evaluation functions in original BLP model,it can evaluate credibility of access requests as well as corresponding credibility variation of subjects and objects.Since this model establishes restriction mechanism against trusted subjects and assigns limited privileges to all subjects,it is more flexible and practicable than present security-label based MLS models.关键词：multilevel security;credibility;BLP security model;access control2|1091|12更新时间：2025-07-16
- 摘要：Automatic gain control (AGC) is very essential for Orthogonal Frequency Division Multiplexing (OFDM) receivers,because the significant features of OFDM signals are large dynamic range and high Peak-to-Average power Ratio (PAR).An adaptive AGC strategy is proposed to improve the convergence speed and maintain the stability of AGC circuits,in which Mean Absolute Deviation (MAD) architecture with hybrid gain compensation parameters and programmable power deviation reference is employed.At the same time,to reduce the power dissipation of receivers,an adjustment successful signal can be given for following digital processing modules.Besides,it can be employed in multi-antenna OFDM systems.关键词：adaptive automatic gain control (AGC);orthogonal frequency division multiplexing(OFDM);hybrid gain compensation parameters;absolute power deviation reference2|1787|2更新时间：2025-07-16
- 摘要：For the problem of fault diagnosis in dynamic system,a principal component analysis model based on credit assigned cerebellar model articulation controller is proposed to carry out on-line fault detection and isolation for multiple sensor system.Firstly,the forecasting values of sensors are available from historical data measured in fault-free condition based on principal component analysis model.Secondly,the Squared Prediction Error of the system is calculated,the fault occurred when the SPE is suddenly increased.Sensor values are reconstructed respectively to newly calculate the SPE to locate the faulty sensor.Finally,Compared to BP and CMAC,the method proposed is proved feasible and effective by a simulation of multiple sensor fault diagnosis.关键词：principal component analysis;fault detection;fault isolation;signal reconstruction2|973|5更新时间：2025-07-16
- 摘要：This paper applies the remainder of analogue quantity and the residue of analogue quantity to the implementation of pipelined folding and subranging ADCs.Two improved circuit structures are proposed,one structure makes use of the conversion of residue of analogue quantity,while the other doesn’t.Experiments on dynamic performance of two structures are conducted.The test result shows that ACD+ built without conversion of residue of analogue quantity can achieve a performance of 2bits@40MSPS ADC+ and 2+8bits@40MSPS ADC.Finally,the structure and dynamic performance test graph are given in detail for practical implementation of ADC circuit.关键词：analog to digital converter (ADC);folding;subranging;residue of analogue quantity;remainder of analogue quantity2|1121|4更新时间：2025-07-16
- 摘要：To resolve the multisensor multiplied maneuvering target tracking problem,a new algorithm is proposed.Firstly,multiple tracking models are set and the synthesis similar degree between measurement and prediction is calculated with applying method of the fuzzy mathematics.Then,association probability can be calculated with using the threshold distinction principle and experiential probability principle,and the state estimation and estimation covariance is updated for each target with applying the association probability.Finally,each model probability is calculated and the ultimately state estimation of each target is calculated in fusion center.Some typical simulations are used to compare the performance of this centralized interacted multiple model multisensor fuzzy joint probabilistic data association (CIMM-MSFJPDA) algorithm proposed here with the centralized interacted multiple model multisensor joint probabilistic data association (CIMM-MSJPDA) algorithm,and results show that the tracking precise of CIMM- MSFJPDA algorithm is better than that of CIMM-MSJPDA algorithm,and the computational burden of CIMM-MSFJPDA algorithm is much lower than that of CIMM-MSJPDA algorithm.Therefore,the synthesis performance of CIMM-MSFJPDA algorithm is better that of CIMM-MSPDA algorithm.关键词：centralized;interacted;multisensor;fuzzy theory;data association2|1195|1更新时间：2025-07-16
- 摘要：A novel robust approach to video text detection and localization is presented.Through utilizing the adaptive threshold and the statistics coarseness feature of each horizontal (vertical) pixel line of difference image,the text detection is carried out fast and effectively.To suppress the complex background interference,and perform the video text localization accurately,the magnitude of color gradient is computed only for key feature points (selected pixels),based on the direction information of gradient vector.The proposed method is robust to various video image quality,background complexities and text appearances.Experimental results on a large number of video images demonstrate the effectiveness of our approach.关键词：video text detection;localization;color image gradient;coarseness2|1219|3更新时间：2025-07-16
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0