15 October 2014, Volume 40 Issue 10
-
Select all|
-
Computer Engineering. 2014, 40(10): 1-5. https://doi.org/10.3969/j.issn.1000-3428.2014.10.001
For power and rate constraints problem that exists in mobile sign language video communication,this paper proposes a Power-rate-distortion(P-R-D) model. It chooses the quantization step size as the model parameter by the experiments on the software platform JM8． 6 of H． 264 video coding,then defines three energy states according to the remaining amount of battery energy of mobile devices. With quantization step size as the model parameter,it establishes the power consumption model, coding rate model and distortion model under the three different energy levels. Experimental results show that the established P-R-D model for sign language video coding can accurately reflect the relationship among power consumption. The maximum prediction error of power consumption model and coding rate model respectively are -1． 179 5% and -7． 926 8% .
-
Computer Engineering. 2014, 40(10): 6-10. https://doi.org/10.3969/j.issn.1000-3428.2014.10.002
In order to provide larger capacity of the hidden secret messages and maintain a good visual quality of stegoimage, a novel image steganography based on visual perception and Diamond Encoding(DE) is proposed. According to the Human Visual System(HVS),human eyes have different sensitivity on high gray areas,middle gray areas and low gray areas,texture areas and smooth areas. The different amounts of secret messages are embedded in the different areas of cover images. Especially,the modified DE is used to minimize the change of the image pixel value. Experimental results show that the proposed method can effectively improve the embedding capacity than HVS-OPAP algorithm by 4. 09% , while stego-image has better imperceptibility,PSNR average of 1. 5 dB,and RS steganalysis methods can not detect whether the secret information is embedded,so it has a better security.
-
Computer Engineering. 2014, 40(10): 11-16. https://doi.org/10.3969/j.issn.1000-3428.2014.10.003
The average scheduling strategy based on task numbers in fault-tolerant has shortcoming such as processing span and Quality of Service(QoS). Aiming at this problem,this paper proposes a grid scheduling algorithm based on priority,and then puts forward the design scheme of the hierarchical cluster system. The algorithm introduces the task in the process of task scheduling,task execution time remaining concept value density,cost budget and span,to make up for the task scheduling in the treatment of two span and QoS. Experimental results show that the algorithm has better effect in the completion rate,value realization rate,treatment rate compared with the original mechanism of scheduling strategy and Max-Min algorithm. The original mechanism is improved by using this algorithm. It can effectively improve the efficiency of system task execution,and the availability of system.
-
Computer Engineering. 2014, 40(10): 17-19,24. https://doi.org/10.3969/j.issn.1000-3428.2014.10.004
As a new generation video coding standard,HEVC has more compression efficiency than the existing video coding H． 264 standard,but it has higher complexity. Aiming at this problem,after brief analysis of the intra prediction algorithm,a fast algorithm based on Rate-distortion (RD) cost of intra prediction block dividing and the intra-frame prediction mode selection fast algorithm is proposed to reduce the HEVC intra-frame prediction complexity. This algorithm takes full use of the advantages of rate distortion cost and the probability of the prediction mode is selected in the candidate pattern quick descending law. Experimental results show that complexity reduction from HM10． 0 is over 59% and stable for various sequences,which makes the proposed algorithm suitable for real-time applications. The corresponding bit rate increase is lower than 1． 34% .
-
Computer Engineering. 2014, 40(10): 20-24. https://doi.org/10.3969/j.issn.1000-3428.2014.10.005
Attribute revocation is crucial to the practical use of Attribute-based Encryption(ABE). Most of the existing revocable ABE schemes under the indirect revocation model suffer in terms of delaying in revocation or updating keys and ciphertexts. To address this,this paper proposes a proxy-based immediate attribute revocation Key Policy(KP) attributebased encryption under the indirect model without issuing new keys or re-encrypting existing ciphertexts. It achieves attribute revocation by introducing a proxy in the decryption process and reduces the burden for the key authority. The proxy is semi-trusted which revokes user access privileges and cannot decrypt ciphertexts. Analysis results show that the scheme supports fine-grained access control policies and achieves three kinds of revocation including system attribute revocation,user revocation and user attribute revocation．
-
Computer Engineering. 2014, 40(10): 25-31. https://doi.org/10.3969/j.issn.1000-3428.2014.10.006Most existing eXtensible Markup Language ( XML ) keyword searches are based on Lowest Common Ancestor(LCA) semantics tree to generate search result,but they do not consider the data which is not included in LCA semantics tree while is relevant with user search intention. To solve this problem,an XML keyword query method based on extended query expression is proposed. The query expansion statistical model is based on user query log. Through analyzing query log and combined with optimal retrieval concept,it can judge whether the query expression should be expanded. After that,an XML TF-IDF method is employed to calculate the weight of candidate attribute. According to the context information and using cluster method,it gets the query expression keywords which are most relevant with search intention. Then the expanded query expression is generated. Compared with XSeek and semantics dictionary based query expression method,experimental result shows this method can improve the query quality by average 7% and 17% in Fmeasure respectively.
-
Computer Engineering. 2014, 40(10): 32-36. https://doi.org/10.3969/j.issn.1000-3428.2014.10.007In order to configure the system parameters of database connection pool effectively,a Geom / Geom / c / c discrete-time queuing model with multi-server is built according to the operation mechanism of its manage process and the thought of discrete-time queuing theory． It uses the embedded Markov chain method, and the matrix of transition probability and matching recursive relations of the stationary queue length are given． The mathematical expressions of system performance measures are derived by the theoretical analysis results of the model,such as request blocking probability,average number of connections,system utilization,and system throughput,etc． Experiments are intuitively given to prove the dependencies between database connection pool performance index and system configuration parameters.
-
Computer Engineering. 2014, 40(10): 37-42. https://doi.org/10.3969/j.issn.1000-3428.2014.10.008Through analyzing the time consumption of the existing online New Event Detection(NED) algorithm based on news elements,this paper improves an online NED algorithm for large-scale data environment． The algorithm uses efficient reported similar search mechanism based on inverted index to reduce the similarity comparison of single path clustering algorithms． Through parallelization of report pretreatment, report and event comparison, index search, it improves the efficiency and scalability of the algorithm in multimachine. Experimental result shows that the algorithm can greatly improve new event detection speed without affecting the miss probability and false-alarm probability,and its throughput reaches 150 ~200 reports / s at the scale of 10 ~100 million reports.
-
Computer Engineering. 2014, 40(10): 43-46,51. https://doi.org/10.3969/j.issn.1000-3428.2014.10.009distribution,this paper proposes an improved data stream update algorithm———Extended Hybrid Join(EH-JOIN). The algorithm improves the traditional Hash join method,and it can adapt to common skewed data and greatly reduce the disk I / O cost through using index structure and storing some parts of the master data in memory. Experimental results show that the service rate of proposed algorithm is improved by 96% and 80% compared with MESHJOIN algorithm and R-MESHJOUIN algorithm as the relation set keeps an appropriate size,and the service rate of proposed algorithm is improved by 57% and 48% compared with MESHJOIN algorithm and R-MESHJOUIN algorithm as the memory size differs.
-
Computer Engineering. 2014, 40(10): 47-51. https://doi.org/10.3969/j.issn.1000-3428.2014.10.010Fuzzy C-means(FCM) algorithm requires constant iteration to calculate the characteristics of the membership value of the sample data and cluster center,using MapReduce model to solve the FCM under massive data. Map stage calculates membership degree,and Reduce stage completes computing cluster center. Each iteration needs to start a MapReduce implementation process. Through multiple iterations,it calculates the value of membership and cluster center, and updates cluster center file for the use of next round job. Repeat this process until get the final clustering results. Experimental results show that the algorithm can effectively reduce the number of iterations during the calculation and improve the overall efficiency of the implementation.
-
Computer Engineering. 2014, 40(10): 52-60. https://doi.org/10.3969/j.issn.1000-3428.2014.10.011Integrated Modular Avionics(IMA) partitioned system is put forward to improve the reliability and security of the system,but in the process of system design and implementation,analysis and validation should be used to ensure that the system time requirements are met. In allusion to IMA system which meets ARINC653 specification,according to the characteristics of hierarchy scheduling partitioned system,this paper proposes a schedulability determination method based on simulation method. With the aid of Cheddar tool and the function of custom scheduling strategy,Architecture Analysis and Design Language (AADL) has the ability of modeling partitioned system,then the tool operates on the generated AADL model to determine the system schedulable. Example analysis result shows that the tool has automatic, accurate and fast advantages to describe task scheduling process by Gantt chart and the result is accurate and detailed.
-
Computer Engineering. 2014, 40(10): 61-65,70. https://doi.org/10.3969/j.issn.1000-3428.2014.10.012As the basics of running application on Android system,performance of Dalvik virtual machine restricts the Android’s user experience. By researching Dalvik architecture in the Android system and analyzing some key techniques of the interpreter and Just in Time(JIT) module,it finds that short Trace’s compiler dissipation is large and there are some irrational situations on register allocation in JIT. Combining nowadays JVM technology with modern compiler technology,and based on the Trace selection strategy and register allocation mechanism of Dalvik,this paper proposes algorithms of combining Trace and optimizing strategy of register allocation. These algorithms are implemented in high performance embedded CPU CSKY architecture. The experiments prove that this Dalvik can improve the performance by about 10% .
-
Computer Engineering. 2014, 40(10): 66-70. https://doi.org/10.3969/j.issn.1000-3428.2014.10.013Aiming at the problem of low precise and slow response in the state of JavaScript code recommendation technology,this paper introduces a JavaScript code recommendation method based on dynamic analysis and implements a code recommendation plug-in for Eclipse. This approach builds model offline for storing the simulated runtime environment,and creates indexes for all objects in the environment. It simulates executing user code,while applying abstract syntax tree of user code for blocking and updating undefined variables at runtime,to achieve a JavaScript code recommendation tool based on dynamic a nalysis. Compared with current code proposal tools with static analysis, experiments show the tool improves both the precise and response time.
-
Computer Engineering. 2014, 40(10): 71-75. https://doi.org/10.3969/j.issn.1000-3428.2014.10.014Elastic buffer is a very important function in receiver at high-speed serial protocols such as PCI Express (PCIE) and USB. The frequency delta and phase delta are managed perfectly to ensure data integrity by the elastic buffer when bridging local clock domain and recovered clock domain. This paper proposes elastic buffer logic in primed method to satisfy PCIE2． 0 protocol specification. The width and depth of the elastic buffer are both 10. The proposed logic is integrated into Physical Layer (PHY) and implemented based on 55 nm CMOS process of SMIC. The result to be measured well agreeds with the actual requirement. It can work well under the frequency of 500 MHz and ensure data to receive data correctly.
-
Computer Engineering. 2014, 40(10): 76-80,85. https://doi.org/10.3969/j.issn.1000-3428.2014.10.015Aiming at the problem that existing cache technology has computational complexity and poor applicability, instruction cache optimization techniques based on statistic analysis by using GNU coverage analysis tool and performance analysis tools are proposed. It uses optimization techniques,including optimized cache line coloring algorithm,static use of cache locking and code selectively non-caching,can significantly improve the efficiency of instruction cache reading. Also a Cache Locking Selection Sorting(CLSS) is proposed to evaluate the code segment which can be locked or noncached. Simulations show that these techniques make the program running time reduced by 8% ,and make the cache hit rate increased by 23% .
-
Computer Engineering. 2014, 40(10): 81-85. https://doi.org/10.3969/j.issn.1000-3428.2014.10.016Because current virtualized devices in the system simulation process and I / O instructions execute serially, which can not play performance advantages of accelerated I / O access technical such as Direct Memory Access(DMA), asynchronous I / O in the real architecture,there is a certain gap between virtual platform I / O performance and the real I / O performance. To solve this problem,this paper presents a stand-alone I / O virtualization method,which separates the I / O simulation activities such as disk access from the Virtual Machine Monitor(VMM) only retained the appropriate interfaces by putting the activities in a stand-alone process to achieve I / O simulation and parallel instruction execution, and based on QEMU platform implements the method. Experimental results show that the I / O performance is better than the original QEMU I / O subsystem by this method. Because I / O instructions do not have to wait for I / O simulation ends, this method not only improves the speed of the CPU instruction execution,but also can effectively improve disk read speed by 4． 9% and increase software Viewpref VGA benchmark score by 9． 2% .
-
Computer Engineering. 2014, 40(10): 86-91,97. https://doi.org/10.3969/j.issn.1000-3428.2014.10.017The quality of Graphical User Interface (GUI) directly affects the reliability and usability of the entire software system. When testing the graphical user interface,it generally adopts the method of extracting model,and the frequently models based on the graphical user interface are Event Flow Graph(EFG) and Event Interaction Graph(EIG).However,the algorithm of EFG into EIG is complex,this paper proposes an algorithm. In this paper,the GUI events with a specific division and the use of the special nature of the modal window event optimize the original algorithm. It indicates the new algorithm can be used for GUI and the obtained data is more accurate and more efficient with the real cases than Memon in GUI.
-
Computer Engineering. 2014, 40(10): 92-97. https://doi.org/10.3969/j.issn.1000-3428.2014.10.018Due to the serious dependence on the topology of the network in DV-Hop algorithm,a DV-Hop localization algorithm based on multi-mobile beacon called MMB-DV-Hop is proposed in this paper. The algorithm incorporates Range-based and Range-free algorithms by introducing multi-mobile beacons into the DV-Hop positioning network,and the beacons keep a certain relative position while traversing the entire network and ranging with RSSI to provide information to unknown nodes to estimate locations. Simulation results show that the proposed algorithm has better locating performance in positioning accuracy by reducing the dependence on the topology of the network and the influence of distance estimation error on the positioning accuracy.
-
Computer Engineering. 2014, 40(10): 98-103,108. https://doi.org/10.3969/j.issn.1000-3428.2014.10.019Target tracking is a basic application in Wireless Sensor Network(WSN). It is a core problem to make a high tracking precision with low energy consumption and prolong the network’s life cycle. Aiming at this problem,a target tracking algorithm based on energy-efficient dynamic clustering is proposed. It firstly presents a new method of generating the dynamic cluster from the view of greatly saving energy. Then,the generated cluster structure uses the Unscented Particle Filtering(UPF) algorithm to track the target and predict the location coordinates in next moment. Finally,according to the predicted results,this paper puts forward a cluster head replaced policy. Simulation results show that,compared with PPF algorithm and DPF algorithm,this algorithm not only has higher target tracking precision,but also effectively reduces the network energy consumption and extends the network lifetime.
-
Computer Engineering. 2014, 40(10): 104-108. https://doi.org/10.3969/j.issn.1000-3428.2014.10.020When the users adopt the HARQ ACK / NACK Multiplexing model feedback in TD-LTE, the PUCCH channel carries feedback results,which sometimes are ambiguous. So evolved Node B(eNB) cannot obtain the exact feedback results. Therefore, this paper puts forward a method called ACK / NACK multiplexing feedback results judgment algorithm based on User Equipment(UE) air interface quality, which judges the ambiguous feedback results to be more accurate. In the scheduling system, UE reports CQI / PMI / RI parameters as characterizations of UE air quality to eNB. And eNB amends reported CQI parameters, based on the parameters of each UE and the received UE Block Error Rate(BLER), which are output to the packet scheduler to decide which UE to be scheduled in next scheduling period. And through the corresponding platform simulations it proves that the algorithm can decrease BLER and increase the system throughput.
-
Computer Engineering. 2014, 40(10): 109-113. https://doi.org/10.3969/j.issn.1000-3428.2014.10.021A relay selection jointly with power allocation scheme is proposed to improve the outage performance of an Amplify-and-Forward(AF) cooperative network with partial Channel State Information(CSI). Firstly,it is proved that the system Signal-to-Noise Ratio(SNR) is subjected to the smaller one of the SNR between Source-Relay and Relay-Destination links. According to that,the potential relays choose different feedback schemes based on their statistical CSI to the source and the destination. Then the source node selects one best relay to forward messages based on the information feedback from relays. A corresponding power allocation scheme is also developed by the convex optimization method. Simulation results show that the proposed scheme significantly reduces the outage probability compared with the conventional schemes,and outperforms other schemes with the same feedback load in outage behavior.
-
Computer Engineering. 2014, 40(10): 114-117. https://doi.org/10.3969/j.issn.1000-3428.2014.10.022Toward the complicated structure of low power implementation of the Viterbi decoder in wireless communication,a low power (2,1,9) Viterbi decoder with the structure of series and parallel combination in four-level pipeline is proposed in the paper. To increase working rate,with the consideration of the implementation hardware complexity,a modified Add-compare-select(ACS) unit is used to satisfy its low power decoding requirment． In order to increase the efficiency of decoding and decrease the latency of decoder,a method of path mutual eliminating is employed in the design. Implemented by TSMC 0． 18 μm standard CMOS technology under 1． 62 V and 125 ℃ ,and analysed with placement and route,the chip’s highest speed is about 50 MHz,the area is 0． 212 mm2 ,and the power comsumption is 23． 9 mW.
-
Computer Engineering. 2014, 40(10): 118-121. https://doi.org/10.3969/j.issn.1000-3428.2014.10.023At present,the study of the Digital Predistortion (DPD) mainly focuses on the simulation of different algorithms,and the specific implementation technology research is less. A kind of adaptive digital predistortion device based on System on a Programmable Chip(SOPC) is proposed in this paper,the design adopts polynomial lookup table circuit to achieve the predistortion, and uses EDK project complete coefficient calculation. This method avoids the complexity brought by the whole hardware implementation,can satisfy the requirements of the adaptive function,and has the very strong practical value. Finally it downloads the design to Xilinx FPGA,and tests it in a test platform. Results show that the predistorter can effectively improve the linearity of the power amplifier. For the OFDM signal of 8 MHz bandwidth and the doherty GaN amplifier,ACPR can improve 8 dB and with the adaptive function,it can be a good application to the actual working environment.
-
Computer Engineering. 2014, 40(10): 122-126. https://doi.org/10.3969/j.issn.1000-3428.2014.10.024Concerning the problem of resource waste in fixed assignment algorithm and uncertain transmission delay in contention assignment algorithm, a dynamic slot assignment algorithm of contention-avoid based on Time Division Multiple Address(TDMA) for clustered Ad Hoc network is proposed. The length of frame can be adapted to the net traffic. When the net traffic increases,the frame length increases to improve channel utilization; when the net traffic reduces,the frame length reduces to reduce the delay of accessing. It is proved by simulation results that the algorithm can increase throughput of the system,compared with NEBS and slot-ALOHA．
-
Computer Engineering. 2014, 40(10): 127-131. https://doi.org/10.3969/j.issn.1000-3428.2014.10.025Due to characteristics of time-varying channel in the wireless cooperative communication system,the Channel State Information(CSI) at the time of relay selection may be different from that at the time of data transmission. Aiming at this problem,the selection cooperation with outdated feedback is investigated in this paper. First,the system model and the relay selection process with outdated feedback are introduced,and the closed-form result of outage probability and closed approximation in the high SNR regime are derived. By analyzing Diversity-multiplexing Tradeoff ( DMT),it obtains the maximum diversity order which is 2 when the CSI used in relay selection is outdated. Finally,numerical simulation results verify the correctness of theoretical analysis,and show the outage performance will be greatly reduced when the CSI at the time of relay selection is slightly different form it at the time of data transmission.
-
Computer Engineering. 2014, 40(10): 132-136,142. https://doi.org/10.3969/j.issn.1000-3428.2014.10.026This paper presents a classification approach for Uyghur text sentiment,such as angry and happy,based on discriminative key word extraction. Combined with the characteristics of sentiment expression in Uyghur text,the term frequency and document frequency are derived as primary statistics. Various discriminative statistics which reflect the discrepancy of the positive and negative sentiment datasets are derived from the primary statistics for each vocabulary word, and are used to extract discriminative key words. Features are extracted based on these keywords and are used to train discriminative sentiment models. This paper builds a sentiment text database by excerpting two sentiments:angriness and happiness from Uyghur movie transcriptions and novels,and verifies the proposed approach. Experimental results show that the method based on discriminative keyword extraction is effective in Uyghur text sentence sentiment classification.
-
Computer Engineering. 2014, 40(10): 137-142. https://doi.org/10.3969/j.issn.1000-3428.2014.10.027The grouping form for collaborative learning group based on learners’ characteristics is one of the factors that enhance the learning effectiveness． A new learning grouping algorithm based on enhanced bacterial foraging is proposed． In order to meet the requirements of different learning activity that is associated to the learners’ characteristics, the regulatory factor and feature weights are used to grouping． At the initialization step of algorithm,there are two method which are used to ensure the effective grouping space,one is that the bacterial population is coded by real number coding, and another is that a random perturbation is used to increase the diversity of bacterial populations． And the algorithm is joined the second mutation to avoid the premature convergence at the later step． Simulation experimental results show that the proposed algorithm is advantage to increase the effectiveness and the accuracy for grouping form． And it has a good stability for data sets with different sizes
-
Computer Engineering. 2014, 40(10): 143-149,154. https://doi.org/10.3969/j.issn.1000-3428.2014.10.028For Extend Kalman Filtering (EKF) algorithm disadvantage on the Simultaneous Location and Mapping (SLAM),that is simple linearization of nonlinear systems resulting from the inaccuracy of the system state equation,Jacobimatrix calculation resulting from computational complexity,and noise uncertainty caused by the filtering reduced stability and other issues,this paper proposes a noise adaptive UKF-SLAM algorithm. In order to achieve adaptive UKF-SLAM algorithm,the paper scales the noise to change the noise model. Using the observed innovation sequence to accurately estimate the covariance of the measurement noise model. And using a new message convariance and IAE fenestration to find the system noise scaling factor,and thus accurately setimates the convariance of the system state noise model,it achieves a adaptive UKF-SLAM algorithm. The sampling strategy of UKF Sigma points is scaling symmetric sampling. Experimental results show that the algorithm has a high accuracy on SLAM compared with the EKF-SLAM and UKF-SLAM.
-
Computer Engineering. 2014, 40(10): 150-154. https://doi.org/10.3969/j.issn.1000-3428.2014.10.029Chinese semantic annotation aims to find out the senses of a multi-semantic word,which is widely applied in natural language processing. This paper proposes a novel approach for semantic annotation of multi-semantic Chinese nouns. Given a multi-semantic Chinese noun,the proposed method can get its senses. The procedure is described as follows. The search results including URLs and abstracts of related Web pages are acquired through a search engine. The initial semantic classes are automatically generated by an online URL classifier using information gotten at the first step. Clustering algorithms are adopted to make full use of the Web page abstracts to get the final semantic classes. Experimental results demonstrate the proposed novel approach can obtain a considerable precision and recall rate with little manual intervention.
-
Computer Engineering. 2014, 40(10): 155-160,167. https://doi.org/10.3969/j.issn.1000-3428.2014.10.030Most studies of opinion dynamics adopt Agent-based modeling and simulation for theoretical research and have serious gap with the real social problems. Aiming at this problem,this paper verifies and improves the interpretation and forecasting capabilities of the model with social statistical data of online rating. On the process of online rating,the individual opinion is influenced by its initial opinion and the group’s opinions. The final opinion determines whether the individual to join the group and makes a rate or not. The rating of the individual affects the opinions and the behaviors of subsequent individuals. A simple dynamic model with continuous opinion based on this process is introduced to predict the number of personnel in online rating. It carries out experiments with the online rating data of film on the Internet website of Douban and analyses the effects of change of score proportion. Experimental results show that the model can effectively predict the number of online rating;Individual final opinion is mainly affected by the opinions of bad-normalgood in the group and almost has nothing to do with its initial opinion;The larger deviation of the Poisson parameter to optimum value leads to the lower accuracy of prediction.
-
Computer Engineering. 2014, 40(10): 161-167. https://doi.org/10.3969/j.issn.1000-3428.2014.10.031To classify and recognize the movement imagery electroencephalogram,is an important problem in Brain Computer Interface (BCI) research. This paper presents a novel method of extracting Electroencephalogram (EEG) features based on Maximum Overlap Wavelet Transform(MODWT) and Autoregressive(AR) model. The EEG signal is decomposed to three levels by MODWT and statistics of wavelet coefficients are computed. Meanwhile,in the EEG signal’s approximation part,the eighth-order AR coefficients are estimated by Burg’ s algorithm,and energe feature vector is also extracted. The combination features are used as an input vector for Neural Network(NN) classifier,Support Vector Machine ( SVM) classifier,and Linear Discriminant Analysis ( LDA) classifier. The recognition result using BCI2003 competition data set is compared with the best result of the competition,and the classification results show the higher recognition rate of the algorithm. Moreover,transplanting the trained successfully model into embedded motor steering control system based on EEG,and the average recognition accuracy of 91. 3% is obtained. The method provides a new idea for the study of embedded BCI system for practical application.
-
Computer Engineering. 2014, 40(10): 168-174,180. https://doi.org/10.3969/j.issn.1000-3428.2014.10.032In order to enhance the global search ability of bat algorithm in solving high-dimensional function optimization problems,a Cross-entropy Bat Algorithm(CEBA) is proposed by combining bat algorithm with CE method.The CE global stochastic optimization algorithm which is based on importance sampling and Kullback-Leibler divergence,is embedded into bat algorithm. By using adaptive smoothing technique, CEBA improves the rate of convergence. The improved algorithm fully absorbs the ergodicity,adaptability and robustness of CE,adaptively avoids the stagnancy of population,and enhances the global search ability. Simulated results conducted on classical benchmarks and 10 CEC2005 benchmarks show that the proposed algorithm possesses more powerful global search capacity,higher optimization precision and robustness.
-
Computer Engineering. 2014, 40(10): 175-180. https://doi.org/10.3969/j.issn.1000-3428.2014.10.033Method of Moments uses two-dimensional ideal edge model to describe sub-pixel edge,but the actual image exists a gradual change on the edge. Thus leads to a principle error that the two-dimensional ideal edge model cannot accurately describe the edge. This paper presents an algorithm which uses an error correction table to regulate error. This algorithm can reduce the principle error from the two-dimensional ideal edge model. And it improves the accuracy of subpixel edge detection. A two-dimensional error correction table is constructed after detecting an ideal image’s sub-pixel edge by method of moments,and simulating the known edge of an ideal image by square sampling theorem. Error correction can be completed after computing the error by querying the error correction table combined with a bilinear interpolation. In a comparison test of gray moments and Zernike moments,experimental results show that compared with methods without regulation this method can effectively improve the accuracy of sub-pixel detection. For the main computational correction algorithm is bilinear interpolation,this method can maintain the real-time sub-pixel edge detection.
-
Computer Engineering. 2014, 40(10): 181-185. https://doi.org/10.3969/j.issn.1000-3428.2014.10.034The research on introducing ontology into semantic virtual environment at present,only uses ontology to visualize the specific domain ontology,and does not display ontology’s advantage. Aiming at this problem,this paper proposes a semantic virtual environment querying and reasoning model based on ontology. The web ontology language named OWL is used to uniformly describe the graphic content and semantic information,and query them respectively. In the process of querying graphic content,it introduces reasoning method of ontology to get out the connotative graphic content,then queries the information which is needed. In the process of querying semantic information,it introduces semantic search method,according to the method which is based on semantic distance to compute ontology concept similarity to compute the similarity between classes in the semantic virtual environment ontology,then searches the instance which has the maximum similarity with the queried instance,and finds out the relation between them with the help of reasoning. This model makes ontology play bigger role in semantic virtual environment,and the feasibility of this Mapping Ontology(MO) is proved by querying and reasoning a semantic virtual furniture shop ontology.
-
Computer Engineering. 2014, 40(10): 186-191. https://doi.org/10.3969/j.issn.1000-3428.2014.10.035An Automatic Color Equalization(ACE) algorithm based on double selective pixels and local adaptability is proposed in this paper to solve the problem that traditional ACE algorithm can’t keep original color information,and it is difficult to meet the real-time application for its highly complexity. The proposed algorithm transforms the obtained image into the YCbCr color space to hold the original color information, and reduces the computation complexity of the algorithm through selecting pixels with the gradient information of intensity image and the local mean-variance information. The local contrast is enhanced because of using local adaptive filtering to adjust the brightness control function of ACE algorithm. Experimental results show that the improved algorithm gets a better result which satisfies the human visual properties compares with Multi-scale Retinex with Color Restoration(MSRCR) algorithm and traditional ACE algorithm. Computational complexity is greatly improved as the algorithm runs faster than conventional ACE algorithm by about 150 times.
-
Computer Engineering. 2014, 40(10): 192-197. https://doi.org/10.3969/j.issn.1000-3428.2014.10.036It is very difficult to visualize the high dimensional solution set of the multi-objective optimization problem for its large number of objective and solution． To solve the above problems,this paper proposes a new method to visualize the high dimensional solution sets with dimensionality reduction and non-dimensionality reduction techniques of data analysis． This method pretreats the solution set of the multi-objective optimization algorithm,uses Principal Component Analysis(PCA) to analyze the characteristics of the data and get the converted data and its corresponding contribution rate． According to the contribution rate order,it adjusts the the order of columns of the converted data,and calculates the distance between the rows of the converted data with the contribution rate use and runs the hierarchical clustering algorithms based on the row distance to reorder the rows and reorganize the data． It displays the result on heat map． Experimental results show that the method can let the user know the contribution rate of the each converted target, offer satisfactory visual effects,facilitate the understanding of the distribution of the data and make decisions．
-
Computer Engineering. 2014, 40(10): 198-203. https://doi.org/10.3969/j.issn.1000-3428.2014.10.0370/ 1 knapsack problem is studied,and adaptive cellular particle swarm optimization algorithm is presented． In the design of the algorithm,the rules about updating the particle’s velocity and position are redefined,an adaptive factor is introduced to provide a basis for the active evolution of the valid particle and the active degradation of the invalid particle,a new coding mode is given to make new particles be valid with great probability and fast speed,cellular and its neighbor are introduced into the algorithm to maintain the swarm’s diversity and the algorithm uses evolutionary rule of cellular in local optimization to avoid local optima． Simulation experimental results of different scale 0/ 1 knapsack problem and comparisons with other algorithms show that the algorithm not only can solve the 0/ 1 knapsack problem effectively,but also can get the good second-best solution even for the global optimal solution with a faster rate,and has a certain degree of stability
-
Computer Engineering. 2014, 40(10): 204-209. https://doi.org/10.3969/j.issn.1000-3428.2014.10.038Based on the analysis of the advantages and disadvantages of the current appliance of swarm intelligence algorithm into Gene Regulatory Network ( GRN), this paper studies the gene regulatory network constructed under Boolean network model using Artificial Fish Swarm Algorithm ( AFSA). Especially, the comprehensive network of predefined attractors and limit cycle is formulated by applying Boolean network model into flower growth morphogenesis.After comparing AFSA with Simulated Annealing(SA) and analyzing the ability of the networks to preserve the attractors when the updating schemes is changed from parallel to sequential,the paper finds the network has robustness within the limit cycle length equal to two and specific network topologies. Experimental results show that the intelligence algorithm outperforms simulated annealing in network discovery and robustness. Therefore,it is feasible to learn Boolean network using AFSA.
-
Computer Engineering. 2014, 40(10): 210-213. https://doi.org/10.3969/j.issn.1000-3428.2014.10.039With the continuous expansion and complication of the decision system,evaluation indexes described as the real value are slightly defective． This paper studies multi-attribute group decision that the attribute value information is in the form of the interval numbers and the weight is given by real numbers． It establishes nonlinear programming model which is used to gather the information of interval numbers． Plant Growth Simulation Algorithm(PGSA) is proposed to gather each program on each attribute group decision preference ordering intervals which are given by different experts, while achieving the preference matrix． Projection theory with the weights of the attribute value is introduced to sort and choose the best decision-making program． A numerical example results indicate the PGSA both calculates simply and flexibly and respects the views of each expert in order to keep the integrity of the information． It shows the effectiveness of artificial intelligence algorithms in information gathering.
-
Computer Engineering. 2014, 40(10): 214-218. https://doi.org/10.3969/j.issn.1000-3428.2014.10.040Aiming at the problem of low prediction accuracy of actual service flow,a novel State Prediction algorithm based on Wavelet transform and LFSN (SPWL) is proposed by Linear Fractional Stable Noise (LFSN) and wavelet transform． In this algorithm,the characteristic of LFSN distribution is defined,and the fitting method of actual data which is satisfied LFSN progress is given with discrete Fourier transform． The prediction accuracy is improved by fusing the results of LFSN model． Simulation is conducted to study the key influence factor of algorithm with NS2 and Matlab,such as time delay,dropping rate,as well as utilization rate． Results show that,compared to FARIMA algorithm,SPWL algorithm has high accuracy prediction,the prediction error of SPWL is 12． 83% ．
-
Computer Engineering. 2014, 40(10): 219-223. https://doi.org/10.3969/j.issn.1000-3428.2014.10.041Ontology plays an important role in knowledge expression,sharing,reuse and semantic query,but in the process of ontology integration,it is difficult to fuse the concept layer. Aiming at this paper,this paper proposes a method to find concept semantic for ontology. Based on the instance similarity metric extract from instance data,it alternates between instance level and schema level to get concept semantic,mines the property semantic,and then turns instance level to amend similarity metric and complement concept semantic on schema level. This paper tests this method with test ontologies provided by OAEI2012,and the result shows that the precise and recall ratio are increased,which proves the feasibility and effectiveness of the method
-
Computer Engineering. 2014, 40(10): 224-227. https://doi.org/10.3969/j.issn.1000-3428.2014.10.042Aiming at nonlinear,multivariable,strong coupling system control problem,a nonlinear PID control method for self-learning based on hyperbolic function is proposed. Due to the process in PID controlling,it forms a consensus in the control field that the curves of proportional,integral,differential three gain parameters following error changes. Thus, this method uses hyperbolic function to construct the rule curves of PID three gain parameters following error changes, which puts forward a control theory about making the traditional linear PID control law into nonlinear PID control law, and uses self-learning algorithm to adjust the three nonlinear gain function coefficients in real time. The self-learning control on nonlinear PID is implemented based on the hyperbolic function. Compared with other controlling methods,the simulation results show that the approach achieves a better robust stability and anti-disturbance capacity,which is an effective control method and has important application value in the nonlinear control filed.
-
Computer Engineering. 2014, 40(10): 228-231,238. https://doi.org/10.3969/j.issn.1000-3428.2014.10.043In some engineering applications,there are some cases like the effect of environment,the instability of measurement devices. Large error comes into being from the measurement equation in the nonlinear filtering system. The standard nonlinear filtering algorithm cannot remove this kind of system error. This paper addresses this problem,assumes that process noises and measurement noises are subject to Gaussian distribution,establishes the incremental measurement equation using the difference between the adjacent measurements,and obtains the cubature points and its weights using the third-degree spherical-radial rule. The cubature points are employed to approximate the integrals in the Bayesian filtering process numerically,and the incremental cubature Kalman filtering is obtained. Simulation results show that the filtering accuracy of the incremental cubature Kalman filtering is not only better than that of the standard cubature Kalman filtering but also better than that of the incremental extended Kalman filtering. The new algorithm can eliminate the system error of the measurement equation successfully.
-
Computer Engineering. 2014, 40(10): 232-238. https://doi.org/10.3969/j.issn.1000-3428.2014.10.044Camera Response Function(CRF) appears in the formation of a photo,mapping the irradiance to the image intensity. CRF is nonlinear,thus it has effects on deblurring blurred image. But many current deblurring algorithms do not consider the influence of camera response function. Therefore,this paper discusses the algorithm of deblurring based on a single image with considering the influence of CRF,and presents a model combining the motion blur model with the CRF. This model can describe the forming process of a real blurred image accurately. Further more,it presents a flexible edge selection CRF estimation method based on the motion model. Experimental results prove that the method has advantage in solving CRF method and removes the influence of CRF on deblurring.
-
Computer Engineering. 2014, 40(10): 239-244. https://doi.org/10.3969/j.issn.1000-3428.2014.10.045To recover the original shape of the imperfect triangular mesh model with missing sharp feature effectively, this paper proposes a hole-repairing method based on region growing. The basic procedures of the method are as follows: Hole boundaries are detected according to the topological relationship of the model. Taking hole boundaries as the starting point,mesh around holes are segmented into regions according to breath first research. The normal new facet is obtained by analyzing the geometric property of relevant feature region. Afterward, according to the obtained normal, an asynchronous growth of new mesh is realized through an iteratively way inward and meanwhile,a controlling signal is set for every hole boundary facet to control the asynchronism of region growing. New facets in different region are intersected to rest sharp feature line and feature corner. Experimental results show that the approach works well for recovering the original shape of imperfect model with missing sharp feature.
-
Computer Engineering. 2014, 40(10): 245-251. https://doi.org/10.3969/j.issn.1000-3428.2014.10.046Considering the adjacent frames scene small changes of surveillance video images,this paper proposes a neighborhood calculation histogram equalization algorithm for video image to enhance the contrast. This algorithm employs the adjacent video frame correlation,calculates the neighborhood information of the frame,chooses a suitable enhancement parameter according to the former frame background and detail information obtained by Laplace operator, and generates a modified histogram of the image. The algorithm separates the modified histogram into two parts using its mean value to get two transformation functions. Histogram equalization is operated for the latter frame. Experimental results show that this algorithm efficiently enhances the contrast of video images,preserves the mean brightness,alleviates the over-enhancement problem and keeps the interest region details. Compared with the traditional algorithms,it acquires the lowest average absolute error of brightness value as 1． 474 1,highest entropy as 7． 099 3,and highest peak signal to noise ratio as 20． 671 0.
-
Computer Engineering. 2014, 40(10): 252-257. https://doi.org/10.3969/j.issn.1000-3428.2014.10.047Tensor product wavelet only emphasizes on the edge of the horizontal and vertical direction. Nonseparable wavelet is isotropic,can extract the edge of the image in all directions,and can obtain relatively complete outline of the image. When this kind of characteristics is applied to the image registration,it can acquire the accurate positions of the invariant points of the affine transform. Based on this characteristic,this paper proposes an image registration method through calculating the registration parameters of the high-frequency sub-images of the nonseparable wavelet decomposition of the original image. This method can transform the registration of the original image into the registration of its high-frequency sub-images. The correctness of the proposed registration method is proved according to the fast algorithm theory of wavelet decomposition. A four channel nonseparable wavelet filter bank is constructed and the registration steps of the method are given. Experimental results show that the method has better registration results. The amount of computation for computing the registration parameter is less than a quarter of the registration parameters computation of the original image. When compared with registration method based on the tensor product wavelet decomposition,the method has higher registration precision.
-
Computer Engineering. 2014, 40(10): 258-263,269. https://doi.org/10.3969/j.issn.1000-3428.2014.10.048For a single method of extracting image feature retrieval defective,this paper proposes a combination of Nonsubsampled Shearlet Transform (NSST) statistical features and Rotation Invariant Local Phase Quantization (RILPQ)texture image retrieval method. NSST exhibits highly directional sensitivity and shift invariance,even it can be sparse representation of the image. In contrast,NSST acquires the natural texture and edge information with the traditional wavelet and has higher computational efficiently with nonsubsampled contourlet transform. This paper acquires the statistical features of the image NSST coefficients by Generalized Gaussian Distribution(GGD) function. Image features are directly extracted by RI-LPQ description operator. Texture images on the Brodatz image database are retrieved by the formula of similarity measure with weight coefficients. Experimental results show that average retrieval rate of NSST statistical features method is 4． 77% and 1． 44% higher than traditional wavelet method and Contourlet method respectively. The average retrieval rate of this paper method based on fused features is 7． 36% and 1． 98% higher than NSST method and RI-LPQ method respectively.
-
Computer Engineering. 2014, 40(10): 264-269. https://doi.org/10.3969/j.issn.1000-3428.2014.10.049To solve the problem of low efficiency and weak stability in searching k-nearest neighbor of large-scale scattered point cloud by using sphere space,a fast algorithm for finding k-nearest neighbor is presented. Point cloud data is divided into different sub-space by using partition strategy without considering k. The radius of the initial dynamic sphere is determined adaptively based on the approximate density of sampling points in a sub-space. The k-nearest candidate points are searched by the circumscribed cube of a dynamic sphere. When the number of k-nearest candidate points does not meet the requirement,or the search fails,the expanding extent is ensured by a circumsphere of the circumscribed cube. Experimental results show that the proposed method obtains not only a better performance and automation than the existing algorithms,but also a quite stability for the anticipated point number of the sub-space and sampling density.
-
Computer Engineering. 2014, 40(10): 270-274,280. https://doi.org/10.3969/j.issn.1000-3428.2014.10.050In order to avoid the big gyration that may occur when the PID controller is used to the Unmanned Surface Vessel(USV),a straight line path following control method based on fuzzy PID is presented. Using the angle deviation between course angle and the straight line as well as the distance deviation between the USV and the path as input,the fuzzy algorithm is used to adjust the PID parameters dynamically,and the PID controller outputs the control voltage of the motors driving the propellers,so that the USV can track the straight line. Simulation results show that the proposed method can avoid the big gyration that appears when PID controller is used at large initial deviation of course angle. In comparison with PID controller,the fuzzy PID has better performance in overshoot and regulation time at small initial angle deviations.
-
Computer Engineering. 2014, 40(10): 275-280. https://doi.org/10.3969/j.issn.1000-3428.2014.10.051With the development and the wide application of mobile communication technology,mobile phone crimes become a new trend,and mobile phone forensics is an effective means to crack down on mobile phone crimes. In view of the problems that the traditional method calling API directly cannot restore the data of Android mobile phone and existing methods cannot restore the deleted data that section covered,through deep analysis of Android SQLite database file structure and the addressing mode of data record,the method that detects and estimates each Type field in the table is proposed. This method uses extraction recovery to process deleted data,combines “restore method with best effort” to discuss the recovery possibility of the deleted data that part covered,and obtains the deleted data. This method is carried out on the Android emulator. Experimental result shows that this method can restore deleted data successfully,and compared with traditional algorithm,it improves the recovery rate of deleted data with no affection of the information extraction accuracy.
-
Computer Engineering. 2014, 40(10): 281-286,291. https://doi.org/10.3969/j.issn.1000-3428.2014.10.052Considering the contradiction between the efficiency of wireless power transmission and rate of data transmission on the choice of carrier for visual prosthesis,this paper presents 16DAPSK high-order modulation and demodulation technology for visual prosthesis application. Application advantage for 16DAPSK is analysed combined with the algorithm,then Very Large Scale Integration(VLSI) for 16DAPSK is designed and optimized by reducing circuit scales of filters with symmetrical characteristics of coefficients and RAG method. The hardware is verified on the DE4 platform as well as the efficiency of wireless power transmission and rate of data transmission. Experimental results show that the system works correctly,16DAPSK technology can solve the problem of channel delay and improve performance for visual prostheses with the simplified demodulation hardware structure. By optimizing,hardware resources are reduced by 58% ,which increases the feasibility of clinical application for visual prosthesis. Compared with traditional methods, energy efficiency experiment shows that the lower communication frequency for 16DAPSK settles the aforementioned contradiction.
-
Computer Engineering. 2014, 40(10): 287-291. https://doi.org/10.3969/j.issn.1000-3428.2014.10.053According to the linear deviation problem of blood cells parameters detection for the hematology analyzer,a piecewise curve fitting method of blood cells is put forward in the paper. Based on the one-to-one relationship between the original value for all kinds of blood cells which are detected by the hematology analyzer and the theoretical value,for piecewise linear fitting according to the data of different period. In order to avoid the jump phenomena of different curve intersection,using three data points of the two linear curve segment intersection to fitting polynomial by Least Square Method(LSM),to realize the smooth transition between different linear segment. Experimental results show that the algorithm is simple,practical and the fitting curve linear deviation is better than the requirements of hematology analyzer industry standard. In addition,it can fully meet the needs of the clinical application.
-
Computer Engineering. 2014, 40(10): 292-295,303. https://doi.org/10.3969/j.issn.1000-3428.2014.10.054Conventional firmware update is totally refreshed,which is inflexible for updating the specified modules. In the paper,based on Unified Extensible Firmware Interface (UEFI),an approach of Capsule type firmware update is proposed,which encapsulates the update data in Capsule,transfers the Capsule to firmware system,unpacks the Capsule for recollecting the update data,and writes the refresh data to the required memory. The proposed method is able to update the specified firmware function,module,protocol,or application. Experimental result shows that the given method can perform the firmware update in module,and holds the advantages of security,high-efficiency and strong flexibility.
-
Computer Engineering. 2014, 40(10): 296-303. https://doi.org/10.3969/j.issn.1000-3428.2014.10.055Traffic guidance system is a method publishing the processed traffic data in a public platform and as a reference for the user to select the path. Amount of information user receiving in a short time,not only a effects the identify rate of information,but also distracts the driver’ s attention,threats to their safety. Aiming at this problem,a system is proposed with a method of road network dynamic programming. With the environmental data of users and mapmatching algorithm to improve the accuracy of locate position. the method of road network dynamic programming is involved into this system to reduce the unnecessary road status information,and the data exchange technology and coordinate transformation are used to improve system performance at the application of the mobile terminal. Experimental results show that the system can prove the positioning accuracy rate is above 95% ,and compared with the traditional method of publication,which can effectively reduce the amount of information published by 70% .
-
Computer Engineering. 2014, 40(10): 304-307,313. https://doi.org/10.3969/j.issn.1000-3428.2014.10.056To obtain accurate quantitative data as a reference for evaluating and selecting real-time Linux system,this paper researches several more critical characteristics of the real-time preemption patch of real-time Linux operating system in-depth,and proposes a method of program instrumentation for real-time performance evaluation. In the evaluating process,it loads and runs the test program on the Linux operating system with Preempt-RT patch,gets the important system information data by testing the key position of program with code probe to evaluate its real-time performance,and does the comparison test of the interrupt response time and context switch time as the most important criteria between Linux system with Preempt-RT patch and standard Linux system based on the numerical control system platform. Experimental result shows the interrupt response time on the patch system decreases by about 10% compared with standard Linux system,while context switch time decreases by about 99%,which satisfies the hard real-time system and meets the real-time demand of numerical control applications.
-
Computer Engineering. 2014, 40(10): 308-313. https://doi.org/10．3969/j．issn．1000-3428．2014．10．057The optimization path finding algorithm computes a moving path of the object with a width in game scene, which searches a continuous node composed of a node-set from map mask. Starting from the initial node,along the path from start node to find the farthest node and no visible obstacles,as a starting point,move in circles until the termination node set. Then sequentially connect these visible nodes,which can get optimal path. The redundant node with node set, makes the path smooth,reduces the times of changing direction in the process of moving objects,eliminates the object with a width not pass through a narrow channel,needs to re-compute the path problem,which achieves a better user experience.
-
Computer Engineering. 2014, 40(10): 314-317. https://doi.org/10.3969/j.issn.1000-3428.2014.10.058Aiming at making the comprehensive evaluation of virtual maintenance training more reasonable, the improvements on the basis of traditional Analytic Hierarchy Process(AHP) and fuzzy synthetic evaluation method are put forward. The optimal transfer matrix is used, and the AHP is improved with no need to verify the consistency in determining the weight of indexes and eliminate the blindness in adjusting judgment matrix artificially. The traditional fuzzy synthetic evaluation is improved,and weighted average model is used,so all factors can be considered. It is more credible to use the method of interval evaluation according to comment rank to calculate the result of comprehensive evaluation. The UAV’ s virtual maintenance training is used to evaluation. Results show that the improved method is simple,the method is used in Virtools for project implementation and verification,the operational information of trainee can be recorded efficiently, the system can give scores according to the operational record, and the effect of comprehensive evaluation is satisfactory.
-
Computer Engineering. 2014, 40(10): 318-320. https://doi.org/10.3969/j.issn.1000-3428.2014.10.059Binary translation system is a bridge between codes of different platforms,while the performance of the system restricts its application. In binary translation,after data flow analysis on those flags,the translation of conditional flags still needs plenty of native instructions which greatly affects the performance of systems. To solve this problem,a flag pattern optimization method is presented,this method which is based on data flow analysis makes ARM instruction of definition flag and usage flag become a fixed pattern,and depends on different patterns to achieve the same semantic by choosing appropriate MIPS instructions. Experimental results show that MIPS code generated by translating has a 14% reduction,and the system performance is improved by 13． 7% .
2024
Volume 50 No 3
Volume 50 No 3
(Started in 1975, Monthly)
Sponsored by：
East-China Institute of Computer Technology Shanghai Computer Society
Edited & Published by：
Editorial Board of Computer Engineering
Chief Editor：Jiang Bo
Executive Chief Editor:
Hu Bing
Tel：021-67092217
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428
Sponsored by：
East-China Institute of Computer Technology Shanghai Computer Society
Edited & Published by：
Editorial Board of Computer Engineering
Chief Editor：Jiang Bo
Executive Chief Editor:
Hu Bing
Tel：021-67092217
E-mail: ecice06@ecict.com.cn
CN: 31-1289/TP
ISSN: 1000-3428