15 December 2017, Volume 43 Issue 12
-
Select all|
-
Computer Engineering. 2017, 43(12): 1-5. https://doi.org/10.3969/j.issn.1000-3428.2017.12.001Transmission signed file transfer of digital signature technology based on Public Key Infrastructure(PKI) is usually subject to factors such as the environment,the terminal,and the file format between the users,and the signatures between different signature systems are not compatible either.In order to solve these problems,this paper analyzes and studies the cloud digital signature technology,including its advantages,specific protocol and signature procedure,and designs the cloud signature system.The system applications are put on the cloud server,and users only need a simple text interaction with the cloud server to achieve cross-platform operation of signature files.The application results show that the signature operation of the cloud digital signature system possesses the features of unity,convenience and universality.The system can ensure that the signed files are compatible with each other and have the characteristics of high security and low cost of equipment maintenance.
-
Computer Engineering. 2017, 43(12): 6-10. https://doi.org/10.3969/j.issn.1000-3428.2017.12.002Aiming at the problem of now internal low aircraft maintenance resources utilization,a Prognostic and Health Management(PHM) architecture based on cloud computing is put forward.Allocation model based on particles swarm algorithm is set for PHM resources in cloud computing virtual layer,it ensures utility function between users and resources is higher than utility function between asks and resources under optimal service quality,and the simulation experiments are carried out by using CloudSim platform,results show that the architecture can improve the generality and utilization of knowledge to PHM system,and can solve effectively the problem of resource optimization in aircraft maintenance support.
-
Computer Engineering. 2017, 43(12): 11-16,24. https://doi.org/10.3969/j.issn.1000-3428.2017.12.003
In order to maximize the profit of service user virtual resource leasing in cloud computing,a new virtual resource leasing algorithm of cloud computing is proposed.By the cloud computing environment composed of three functional modules which are Virtual Resource Provider(VRP),Cloud Service Provider(CSP) and end user,it gives the virtual resource rental profit target.Considering the distribution of price and the urgency of tasks,for the weakly stationary price sequence,it uses the outlier detection method to filter the extreme price,designs the weak balance operator which uses the exponential function to control the integral shape of objective function’s curve,and uses the non-uniform mutation operator to adjust local operators and effectively predict the future price.Then it gets the optimal leasing rental price of Virtual Machine(VM) processing tasks.Experimental results show that the proposed algorithm can improve the efficiency and rental cost of virual resource and reduce its usage cost.
-
Computer Engineering. 2017, 43(12): 17-24. https://doi.org/10.3969/j.issn.1000-3428.2017.12.004In order to improve the equilibrium of resources scheduling in cloud computing system and maximize the benefits of resource providers,this paper proposes an allocation model for cloud computing resource based on queuing theory.It considers waiting queue length as the premise of resource allocation,analyzes the resource allocation strategy in terms of Nash Equilibrium(NE) theory and proposes Modified Resource Allocation(MRA) algorithm.Meanwhile,this alogithm is compared with some resource allocation algorithms in distributed computing architecture,Hadoop.Experimental results show that,compared with Fair algorithm,First-In First-Out(FIFO) algorithm and classical Random algorithm,the proposed MRA algorithm not only can realize the fair sharing of resources for users,but also can meet the needs of resource providers in improving the system response time.
-
Computer Engineering. 2017, 43(12): 25-29,37. https://doi.org/10.3969/j.issn.1000-3428.2017.12.005
Cloudlets are small self-maintained clouds.They improve the overall performance of mobile services through balancing user access requests.In practice,the dispersed user demand changes with time,and the service resources of cloudlets need to be deployed in advance according to the prediction of the required resources.Therefore,in order to optimize the overall performance of the system resource utilization,queuing theory is used to construct resource limited service model of cloudlets which fits in with the birth and death processes.Then,the comparison between single and multiple queues show that the single queue model with multiple virtual cores is better than the multi-queue model with single virtual core under the same service intensity.Experimental result showes that the service model of cloudlets can further optimize the overall system performance with the increase of the virtual cores.
-
Computer Engineering. 2017, 43(12): 30-37. https://doi.org/10.3969/j.issn.1000-3428.2017.12.006In order to improve the reliability of on-board cloud computing resource scheduling and reduce the computation time of data processing,a parallel MapReduce heuristics on-board cloud resource scheduling algorithm with Quality of Service(QoS) perception is proposed.Based on the MapReduce parallel computing model,the On-Board Unit(OBU)-based vehicle parallel detection service framework in cloud computing environment is designed,and the relative priority factor is used to construct the on-board cloud computing scheduling model.Then the cloud resource scheduling model is optimized by using heuristic parallel optimization algorithm to reduce the computational complexity of the proposed algorithm.The simulation results in NS-3 show that the proposed algorithm can shorten the job execution time effectively and has higher reliability.
-
Computer Engineering. 2017, 43(12): 38-44. https://doi.org/10.3969/j.issn.1000-3428.2017.12.007Due to the characteristics of concurrent,resource sharing and task migration in multi-core environment,uncertaionty has been introduced into the safety-critical task scheduling for partitioned operating system.In view of this,this paper presents a safety-critical task scheduling method for partitioned operating system in multi-core platform.This method divids the system partition into critical partition and non-critical partition by judging whether it contains safety-critical tasks.On the system layer,resource pre-allocated method is used to ensure the independent operation of the critical partitions.Moreover,it determines the minimum resource upper limit according to the partition utilization,so as to improve the resource utilization rate of the system while ensuring the reliability of safety-critical task scheduling.On the partition layer,a dual-mode fault-tolerant task scheduling algorithm based on dynamic priority and a minimum task priority search algorithm based on binary-search are proposed,which can enhance the fault-tolerance abality of the system.Experimental results show that the proposed method can guarantee the partitioned operating system to have higher resource utilization and reliability of task scheduling.
-
Computer Engineering. 2017, 43(12): 45-50. https://doi.org/10.3969/j.issn.1000-3428.2017.12.008Based on the cluster architecture,hardware and software of Shuguang 5000A,the computational fluid dynamics method is used to investigate pressure distribution around the tennis surface under various speeds and rotational velocities.For the simulation study,the Realizable k-ε turbulence model is utilized specially for the Magnus force of rotational tennis, a rotaling tennis aerodynamics simulation method is proposed.The air-pressure difference between the upper and lower surface of tennis increases with the increase of rotational velocity,and the enhanced Magnus effect makes the tennis fall down immediately.The increase of ball speed decreases the controllability of tennis and leads to the disorder of air flow around tennis.The drag coefficient and lift coefficient increase with the increase of ball speed and a bigger rotational velocity.Experimental results show that the computational accuracy and efficiency can be improved by the multinuclear parallel computation.
-
Computer Engineering. 2017, 43(12): 51-54,59. https://doi.org/10.3969/j.issn.1000-3428.2017.12.009In order to take full advantage of the communication bandwidth,this paper presents an improved task mapping algorithm based on heterogeneous communication mechanisms for a multicore processor with both Network on Chip(NoC) and shared-memory inter-core communication mechanisms.The proposed algorithm uses the branch and bound algorithm to reduce the search space,and then improves the quality of mapping by iteratively exchanging the positions of the tasks.Experimental results show that,compared with the NMAP algorithm,the communication cost of the proposed algorithm is reduced by an average of 18.3% for random task graphs,and the communication cost can be reduced by up to 30.6% for parallel sorting applications.
-
Computer Engineering. 2017, 43(12): 55-59. https://doi.org/10.3969/j.issn.1000-3428.2017.12.010Coverage-Based Fault Localization(CBFL) method can locate the fault by analyzing the information and results of success and failure test cases.However,CBFL ignores the impact of accidental successful test cases,and the existence of accidental successful test cases will reduce the accuracy of the fault location.Aiming at this problem,this paper presents a new fault localization method.It reduces the influence of accidental success test cases based on program mutation anatysis,improves the doubt degree calculation formula and adds calculation of the influence of variation.Experimental results show that this method can significantly improve the accuracy of fault localization compared with the traditional CBFL method.
-
Computer Engineering. 2017, 43(12): 60-64. https://doi.org/10.3969/j.issn.1000-3428.2017.12.011To reduce the time overhead of assigning multiple Fast Interrupt Request(FIQ),a vectored interrupt controller for FIQ interrupt source identification and priority arbitration is designed.It is described by Verilog-HDL.Modelsim simulation and FPGA verification are also passed.Synthesis is completed with SMIC 0.13 μm CMOS technology,as well as place and route.Simulation and verification results show that the area is 0.107 mm2,the average power is 3.56 mW and the frequency is 80 MHz.This design can meet the real time demand of the chip in mobile communication RF SoC.
-
Computer Engineering. 2017, 43(12): 65-68. https://doi.org/10.3969/j.issn.1000-3428.2017.12.012
The schedulability can be determined based on deadline analysis and response time analysis for the global Fixed Priority(FP) real-time scheduling algorithm.The traditional method takes into account the real-time tasks with a carry-in job.The processor cannot meet the computing requirement of real-time tasks.So this paper proposes a schedulability method.The interference caused by the real time task is analyzed,which takes into account the number of jobs and the number of processors in real-time system.Experimental results show that the method can reduce the amount of interference,obtain a more compact schedulability criterion,and increases the number of tasks that can be passed by schedulability in multiprocessor real-time system.
-
Computer Engineering. 2017, 43(12): 69-72. https://doi.org/10.3969/j.issn.1000-3428.2017.12.013The evaluation effect of current various types of software reliability models is heavily dependent on the assumption of failure behavior,resulting in poor model applicability.The appropriate software reliability model is selected by Analytic Hierarchy Process(AHP),and the appropriate reliability model is modeled according to AdaBoosting algorithm.In the failure data set,the participating model is learnt training,and AMCM model is established.The AMCM model is applied to the four classical software reliability models.Results show that the proposed combination modeling method is effective,it can significantly improve the evaluation performance of the model.
-
Computer Engineering. 2017, 43(12): 73-77. https://doi.org/10.3969/j.issn.1000-3428.2017.12.014To locate the fault points in the core network of mobile operators,previous methods require too many data samples which need long time to find the fault points.Aiming at this problem,this paper proposes an abnormal traffic detection method using the historical data as reference.It establishes a database containing historical traffic data.On this basis,it uses the surface feature and gradient features in short term to describe traffic behaviors of network,so as to detect the abnormal traffic in time.Meanwhile,it proposes a hierarchical and multi-granularity method to find the fault points that lead to the anomalous traffic.Experimental results demonstrate that,compared with adaptive threshold method,K-means clustering method and multidimensional entropy method,the proposed method can not only perform better in online abnormal traffic detection,but also locate the equipments and points that have faluts and abnormal traffic more accurately.
-
Computer Engineering. 2017, 43(12): 78-82. https://doi.org/10.3969/j.issn.1000-3428.2017.12.015Device-to-Device(D2D) communication is an effective technique to solve the shortage of spectrum resources in cellular networks.With the increasing diversification of mobile multi-media services,users need more bandwidth.Aiming at this problem,this paper proposes a heuristic D2D resource allocation scheme based on Proportional Fairness(PF) scheduling algorithm.According to the D2D pairs’ priority in descending order,it allocates resources for D2D pairs one by one.Simultaneously,to ensure that D2D pairs with poor channel state can be scheduled,D2D pairs whose data rate requirements are satisfied are prohibited to reuse Resource Block(RB).When all D2D pairs’ data rate requirements are satisfied,users’ priority will be readjusted to improve the system throughput as much as possible.Simulation results show that the proposed scheme can guarantee the data rate requirements of D2D pairs and improve the throughput of the system while taking into account fairness between different D2D pairs.
-
Computer Engineering. 2017, 43(12): 83-87. https://doi.org/10.3969/j.issn.1000-3428.2017.12.016For the features of high speed moving,sparse distribution,diversity of data service,etc.,and Quality of Service(QoS) demand of high reliability and low latency when transmitting time-sensitive information in airborne tactical networks,a random access multi-channel Medium Access Control(MAC) protocol is proposed based on Reed Solomon Polar(RS-Polar) coding technology.The proposed protocol adopting error correcting coding technology of RS-Polar cascaded and the multi-packet reception technology,ensures the real-time and reliability of packet transmission without the use of channel reservation and time-slot allocation,and improves the network throughput effectively.By establishing the channel collision model and the multi-packet reception model in the burst packet propagation stage,the mathematical expressions of packet success probability,packet end-to-end delay and network throughput are derived.Simulation results show that packet end-to-end delay of the protocol is less than 2 ms,and the packet success probability is higher than 99% when the load is less than 1 000 packet/s,can meet the performance requirements of the airborne tactical network effectively.
-
Computer Engineering. 2017, 43(12): 88-92,97. https://doi.org/10.3969/j.issn.1000-3428.2017.12.017In the actual Wireless Sensor Network(WSN),the perimeter forwarding method can easily cause the hole edge nodes’ energy consumption to be too fast because of the routing void,resulting in the expansion of the void,and the formation of energy holes.In order to prolong the network lifetime,and avoid energy hole,this paper defines the blind area and the blind node,designs blind area search rules,and proposes a routing algorithm based on blind node search.This paper uses the blind node and blind area information in the network to route,thus avoiding the blind node of the hole edge,and reducing the frequency of using perimeter forwarding.Simulation results show that compared with GPSR and GEAR,this algorithm can effectively prolong the network lifetime,and also improve the performance of packet delivery rate and transmission efficiency.
-
Computer Engineering. 2017, 43(12): 93-97. https://doi.org/10.3969/j.issn.1000-3428.2017.12.018In order to improve the low interception performance of aircraft formation,a new method of Low Probability of Intercept(LPI) formation communication is proposed.The communication of aircraft formation is considered as an adaptive optimization problem,and the radiation power and topology link of the aircraft node are taken as the optimization object.The Hybrid Chaotic Particle Swarm Optimization(HCPSO) algorithm is used to carry out iterative calculation to realize the optimal low intercept communication of the aircraft formation.Simulation results show that the chaotic particle swarm optimization can effectively reduce the formation radiation power,improve the communication topology uncertainty and improve the low intercept ability of aircraft formation under the premise of meeting reliable communication.
-
Computer Engineering. 2017, 43(12): 98-102. https://doi.org/10.3969/j.issn.1000-3428.2017.12.019In the design of routing algorithms in mobile opportunistic networks based on node social information,there is a problem that it is hard to balance transmission success ratio with average transmission delay effectively.In order to solve this problem,the routing algorithm based on social relationships is proposed,which makes use of center of the improved bridging centrality degree to evaluate the node heterogenicity and speeds up the messages forwarding through the introduction of forwarding discrimination factor in the community,and chooses forwarding nodes according to simple community identification algorithm with aging mechanism.Simulation results show that the proposed algorithm can effectively improve the message delivery success rate and reduce the average transmission delay compared with the classic algorithm Bubble Rap and its improved algorithms BiBubble and BridgingCom.
-
Computer Engineering. 2017, 43(12): 103-107,114. https://doi.org/10.3969/j.issn.1000-3428.2017.12.020For down-link multi-user Multiple Input Multiple Output(MIMO)relay system,a linear precoding algorithm is proposed with the channel estimation errors and the antenna correlation taken into account.The cost function is designed with Minimum Mean Square Error(MMSE) rule under the power constraint at the transmitter and relay terminals.The precoding matrix of relay and transmitter is obtained by theoretical deduction.Assuming that each user of receiver is independent of each other,the processing matrix is a diagonal matrix,so the closed form of receiving precoding matrix can be obtained by direct derivation.The iterative method is designed to jointly optimize the precoding matrix and the processing matrix at the receiver.Numerical simulation results show that compared with precoding algorithms that failed to take each port into account,the proposed algorithm can reduce bit error rate of the system.
-
Computer Engineering. 2017, 43(12): 108-114. https://doi.org/10.3969/j.issn.1000-3428.2017.12.021For the detection and parameter estimation problem of Linear Frequency Modulated(LFM) interference signals over periodic periods,this paper proposes a fractional spectrum analysis and parameter estimation method.Taking the sampled data across two period as an example,the spectrum distribution characteristics of the sampled periodic linear frequency-modulated signal in the fractional Fourier domain,the relationship of the spectrum overlap width,the fractional rotation angle α and data length in each cycle of two periodic spectrum,and the conditions of the peak shifting are analyzed.The relationship among each parameter is inferred in the digital fractional Fourier transform,as well as the reason of the frequency modulation prediction by fast linear frequency modulation technique.Experimental results show that the proposed method has higher accuracy in parameter estimation in two simulation cases.
-
Computer Engineering. 2017, 43(12): 115-119. https://doi.org/10.3969/j.issn.1000-3428.2017.12.022Existing congestion control algorithm still exists the problem of packet loss rate is too large in the complex network environment.In response to this phenomenon,through studying the problem of network congestion control,this paper proposes a congestion control algorithm,which has the ability of prediction and self-regulation.It uses fuzzy neural network controllers to predict the congestion in the network.According to the queue length in buffer for real-time predictioin.Before the congestion is about to occur,it uses transmission rate of the control input to suppress.It dynamicly adjusts the transmission rate combined with variables like increment parameters and decrement parameters.Experimental results show that,this algorithm can be maintained better convergence effect at different signal-to-noise ratio.And its network packet loss rate is not affected by the exchange rate of the network.It has good stability and fidelity.
-
Computer Engineering. 2017, 43(12): 120-123,129. https://doi.org/10.3969/j.issn.1000-3428.2017.12.023In order to improve the decoding efficiency of Spinal code,the beam search and Cyclic Redundancy Check(CRC)decoding are considered as a whole for joint decoding,a CRC-aided multiple traceback decoding algorithm is proposed.The correct decoding route is selected from multiple surviving routes by CRC.The correct decoding time is remarkably shortened,and larger system throughput is achieved as a result.Simulation results show that compared with the fixed backtracking method from a single node,for given coding and channel parameters,system throughput of the CRC-aided multiple traceback decoding algorithm can be increased by 8% to 20% at low Signal to Noise Ratio(SNR) and 58% at high SNR.The decoding amount of improvement increases with the traceback factor.
-
Computer Engineering. 2017, 43(12): 124-129. https://doi.org/10.3969/j.issn.1000-3428.2017.12.024In order to prolong the life cycle of wireless sensor network,this paper proposes an energy aware routing protocol MAPA-MIPC based on clustering network,which combines the supplement of nodes’ energy and the energy saving routing.Using solar energy as an energy supplement,according to the energy remaining of each node on the link,consumption and prediction of the next moment of solar energy collection,it calculates the best upload path of the node data.At the same time,the energy threshold detection is introduced to the cluster head and the relay node election.Nodes that do not reach the threshold value cannot become a cluster head or a forwarding node,thus ensures that the cluster heads and the low energy nodes have a longer life cycle.Simulation results show that compared with MRRCE and EHA-LEACH protocol,MAPA-MIPC has obvious advantage in extending the network life cycle.
-
Computer Engineering. 2017, 43(12): 130-135,146. https://doi.org/10.3969/j.issn.1000-3428.2017.12.025Focusing on the issue of multiple user set-point adjusting the status of dynamic physical process with real time controlling in the Wireless Sensor/Actuator Network(WSAN),an algorithm to solve multiple user set-point scheduling is proposed.The algorithm includes the scheduling model and the state model of dynamic physical process.Based on the optimal estimation of Kalman filter,the optimal control is calculated step by step using the simulated annealing algorithm so as to obtain the global optimal control.Simulation results indicate that compared with Scheduling Algorithm Based on Quadratic Programming(SAQP) and Cross Layer Optimization Scheduling Algorithm(CLOSA),Multiple user Set-Point Scheduling algorithm Based on Simulated Annealing (MSPSBSA) has faster convergence speed,better system stabilization as well as stronger real-time processing capacity.
-
Computer Engineering. 2017, 43(12): 136-140,154. https://doi.org/10.3969/j.issn.1000-3428.2017.12.026In the high speed mobile environment,due to less consideration of single router and adjacent router connection rate,it is easy to lead to poor accuracy of congestion detection and the high error rate of data flow transmission.In order to solve the problem of multi-node network congestion,a kind of sensing network control algorithm based on router group is proposed in combination with the characteristics of dynamic,fast and timeliness of large network data.Through the acquisition of parameters and data-aware to obtain more accurate information as a basis for congestion judgment,and divided the connected routers into the hierarchical routers group,and adjacent two layers of the router was set up as a virtual node.By analyzing the three different congestion cases in the virtual node,the controller will be control and adjust the parameters of the edge router in the virtual node.Experimental simulation results verified the reliability and stability of the algorithm.
-
Computer Engineering. 2017, 43(12): 141-146. https://doi.org/10.3969/j.issn.1000-3428.2017.12.027The traditional feature extraction methods are mainly focus to the role of the category on the characteristic word for text classification,which do not express the impact of the sample on the classification.In this paper,aiming at the problem that the contribution of the sample to the classis is not detected out from the Sprinkling,and the K-Sprinkling is proposed based on these detected sample tightness and sample membership.Then,by considering the Sprinkling advantages,the sample weights are mapped into the vector feature space to achieve the text classification through the potential semantic indexing method.The experimental results show that the K-Sprinkling method proposed in this paper can obtain better classification performance.It outperforms the traditional method by 1.89% on the balance sample,as well as 3.30% on the imbalance sample in terms of F1-score.
-
Computer Engineering. 2017, 43(12): 147-154. https://doi.org/10.3969/j.issn.1000-3428.2017.12.028Traditional evolutionary algorithms usually use the macro-level interaction between individuals or populations to co-evolution,with less consideration in the individual gene encoding in micro level when interacting local optimization.Aiming at this situation,this paper proposes Dynamic Matrix Evolutionary Algorithm(DMEA) based on population gene distribution structure.It uses binary gene matrix to construct individual genes.It evaluates population individuals according to the difference of genes encoding and fitness and decide the adjustment position of individual gene structure by comparing the population gene series.According to structure of the dominant population,it generates the next generation of individuals and guides the evolution of the population by the synergistic effect of micro-level gene location.Experimental results show that the proposed algorithm has good performance in the optimization of mid-dimensional or high-dimensional functions,and it also can better balance the relationship between the microscopic optimization and the macroscopic optimization.
-
Computer Engineering. 2017, 43(12): 155-159. https://doi.org/10.3969/j.issn.1000-3428.2017.12.029In the calculation of user influence,the PageRank algorithm considers only the following relation among users,which leads to the low accuracy of the calculation results.Therefore,a URank algorithm combining user behavior factors with PageRank algorithm is proposed.By using the factors such as forwarding rate,comment rate and authentication,the user’s quality can be obtained by combining the quality of users and the quality of followers.Experimental results show that based on the SIR propagation model,URank algorithm is superior to PageRank algorithm in computational accuracy.
-
Computer Engineering. 2017, 43(12): 160-164,172. https://doi.org/10.3969/j.issn.1000-3428.2017.12.030In order to improve the accuracy of micro-blog emotional classification recognition,regarding the network microblog data as research object,this paper proposes a choice method of emotional basic word based on graph.Combined with similarity knowledge of HowNet,the method builds a graph model to choose high quality and high coverage emotion basic words according to node betweenness centrality in graph.It builds emotional dictionary for sentiment analysis according to selected basic words.The polarity of emotional words is also given.The emotional words are applied to mine short sentence emotional features.Those features will join into traditional Support Vector Machine(SVM) model.More semantic information is mined on micro-blog sentences to obtain a more reasonable semantic composition function.The sentence emotional changes are captured to better grasp the micro-blog emotion of whole sentence.Conditional Random Field(CRF) model that has characteristics of feature constraint is used to classify short sentences.Experimental results verify the effectiveness of CRF model on short sentences.Compared with SVM classification methods with different features,it also has a greater effect on different data sets.
-
Computer Engineering. 2017, 43(12): 165-172. https://doi.org/10.3969/j.issn.1000-3428.2017.12.031Semantic information for Web tables is not understood by machines.Traditional entity column detection methods find entity columns with header information and knowledge base.They are not applicable for tables without headers.This paper proposes an entity column discovery algorithm of Web table based on column value of approximate functional dependencies and normalization,which is used to annotate entity column for tables that have no header or cannot restore a full header even multiple entity column tables.The approximate function dependency relations between Web table attributes are detected according to attribute values in Web tables.The noisy function dependency relations are filtered according to the characteristics of Web tables.The entity columns of the Web table are obtained by normalization of the function dependency set.Compared with entity column detection algorithm based on knowledge base,the proposed algorithm is independent of header information,3%~5% higher in precision and recall,and can be applied in more scenes.
-
Computer Engineering. 2017, 43(12): 173-178. https://doi.org/10.3969/j.issn.1000-3428.2017.12.032The recommendation method based on Denoising Auto-encoder(DAE)lacks of items co-occurrence analysis ability,and the model exists the problem of cold start of parameters.In order to solve these problems,this paper proposes an information recommendation method based on DAE network and word vector.A training corpus is built by mapping users into documents and mapping items into words.The word vector model is used to train the corpus to generate item vectors which contain implicit context information.All item vectors are used as the initial weights to reconstruct the DAE neural network,and the model parameters are obtained through training.The model is used to predict ratings to complete top-N recommendation.Experimental results on standard datasets show that the proposed method improves the accuracy of recommendation,and the training speed is better than that of DAE,Singular Value Decomposition(SVD) and Collaborative Filtering(CF) methods.
-
Computer Engineering. 2017, 43(12): 179-183. https://doi.org/10.3969/j.issn.1000-3428.2017.12.033In order to effectively use domain knowledge to improve the effect of Chinese-Vietnamese metallurgy field translation,this paper proposes a method of statistical machine translation,which uses the terms and topics’ distribution relation to constrain the choice of terms.It constructs term translation model in the process of phrase probability table extraction,the topic information of the source language document for the term pair is incorporated into the translation probability table,and the relationship between the term and the topic is described by topic distribution.Based on the translation model of metallurgy field corpus training,the two models are integrated into the decoding process of Chinese-Vietnamese phrase-based translation system to guide the selection of terms and translations that best match the domain characteristics.The contrastive experiments are carried out before and after the fusion model.The results show that,compared with the phrase-based machine translation method,the proposed method can increase translation performance in the Chinese-Vietnamese metallurgy field effectively.
-
Computer Engineering. 2017, 43(12): 184-191. https://doi.org/10.3969/j.issn.1000-3428.2017.12.034Word embedding can capture the semantic information of words from the large corpus,and its combination with the probabilistic topic model can solve the problem of lack of semantic information in the standard topic model.So in this paper,Word-Topic Mixture(WTM) model is proposed to improve word representation and topic model simultaneously.Firstly,external corpus is introduced into the Topic Word Embedding(TWE) model to get the initial topic and word representation.Then the word embedding feature representation and topic vector are integrated in the topic model by redefining the probability conditional distribution of topic vectors and word embedding,meanwhile the KL divergence of the new word-topic distribution function and the original distribution function are minimized.Experimental results prove that the WTM model performs better on word representation and topic detection compared with Word2vec,TWE,Latent Dirichlet Allocation(LDA) and LFLDA model.
-
Computer Engineering. 2017, 43(12): 192-196,202. https://doi.org/10.3969/j.issn.1000-3428.2017.12.035In text sentiment analysis,feature extraction method based on Chi-square statistic (CHI) is easy to ignore single text word frequency which leads to text feature accuary is low,a feature extraction method based on hybrid chi-square statistics is proposed.By adding the characteristic frequency,inverse document frequency and negative correlation coefficients,it selects the feature words that are concentrated in a particular category and reduces the interference of negative correlation.Logistic regression method based on stochastic gradient descent to realize text sentiment classfication.Simulated annealing principle is used to select step length adaptivly to solve the problem in determining step length when using stochastic gradient descent algorithm.The experimental results show that the proposed method has higher sentiment analysis quality than feature extraction method based on chi-square statistic.
-
Computer Engineering. 2017, 43(12): 197-202. https://doi.org/10.3969/j.issn.1000-3428.2017.12.036Spiking neural network uses temporal coding for data processing,which is an effective tool for complex spatial and temporal information processing.In view of this,this paper applies multiple sequence composite kernel into the spiking sequence processing and proposes an online supervised learning algorithm.It uses accumulation and cumulative mechanisms for supervised learning and makes an experiment compared with online PSD algorithm based on single kernel.Experimental results show the better performance of the proposed algorithm.Especially in the performance of larger data sample,it can maintain this excellent performance more significantly.The results also show that the combination of multiple kernel functions can get more stable and efficient spiking sequence composite kernel representation.
-
Computer Engineering. 2017, 43(12): 203-210. https://doi.org/10.3969/j.issn.1000-3428.2017.12.037Aiming at the slow convergence speed and solution accuracy of Fireworks Algorithm(FWA),by introducing putting reverse learning and dynamic memory feedback mechanism,this paper proposes an optimized algorithm.The proposed algorithm uses reverses learning strategy to generate initial population,so as to ensure the diversity of population.Then it adds feedback layer into the structure of FWA to memory location information for last iteration optimal fireworks and extracts the change of characteristics about position information from the feedback layer in the memory,so as to dynamic update the next iteration of fireworks populations.Simulation is carried out with 10 typical benchmark functions,and the results show that,compared with FWA,Standard Particle Swarm Optimization(SPSO) algorithm and Enhanced FWA(EFWA),the proposed algorithm has better performance in terms of convergence speed,accuracy and stability.
-
Computer Engineering. 2017, 43(12): 211-215. https://doi.org/10.3969/j.issn.1000-3428.2017.12.038The traditional binocular 3D reconstruction methods have several disadvantages such as the rebuild point cloud sparse,high false match rates for the target with poor texture information and so on.This paper proposes a novel 3D reconstruction method.It combines the the digital speckle and dense matching algorithm based on Bayes theory,which achieves high-precision dense reconstruction for texture-deficient objects.The Bayes stereo matching algorithm and digital speckle method are introduced respectively.It verifies the feasibility of combining two methods.Meanwhile,it deduces theoretically the multi-probe system calibration,stitching the point cloud to a full shape.And the multi-probe system consists of multiple pairs of binocular.Experimental results show that the 3D reconstruction accuracy of the Bayes stereo matching algorithm is up to 0.08 mm at 500 mm distance between the object and the multi-probe system.It means that the point cloud number and point cloud accuracy improve significantly without loss of reconstruction efficiency.
-
Computer Engineering. 2017, 43(12): 216-221. https://doi.org/10.3969/j.issn.1000-3428.2017.12.039Aiming at the problem of two-pixel distortion and incomplete refinement in ZS refinement algorithm,an improved refinement algorithm is proposed.The ZS refinement algorithm is used to refine the binary image,and the result is refined.And the 4-connected refinement template is used to refine the incomplete image.Based on the recognition result,at the same time,the breakpoint problem in the refinement process is solved by constructing the reserved template,so as to achieve the effective refinement of the target image.This paper uses the oil scale as an example,it combines with improved refinement algorithm to achieve instrument readings.Simulation results show that the improved refinement algorithm effectively solves the problem of two-pixel distortion and pixel redundancy,and completes the identification of pointer readings accurately,while keeping the original ZS refinement algorithm fast.
-
Computer Engineering. 2017, 43(12): 222-226. https://doi.org/10.3969/j.issn.1000-3428.2017.12.040Considering the time-consuming and unmodifiable problems in manual paper cuts,this paper proposes a new synthesis method for facial makeup of Peking Opera which can simulate the real effects of paper cuts.Based on the analysis on the paper cuts structure of traditional makeup types of Peking opera,this paper constructs the pattern of features,decorations and outlines.In the creation of Peking Opera facial makeup process,user can select patterns from the bank of design patterns,and combine the designed facial makeup with the texture synthesized by Ashikhmin algorithm after adjusting the size and distribution of design patterns.Experimental results show that this method can effectively simulate the hollow effects of manual paper cuts for the facial makeup of Peking Opera and improve the work efficiency,so as to facilitate artists to change the design patterns and revise the works.Therefore,it can be widely applied in interior decoration,advertising design and paper cuts exhibition.
-
Computer Engineering. 2017, 43(12): 227-230,236. https://doi.org/10.3969/j.issn.1000-3428.2017.12.041Camera reference pattern noise is computed by averaging sensor pattern noise values of all images in the traditional camera reference pattern noise extraction algorithm.However,the traditional method cannot filter the interpolation noise and not consider the proportion of different types of images.To improve the extraction effect of the camera reference pattern noise,this paper proposes an improved camera reference pattern extraction algorithm based on the characteristics of noise variance and texture and lightness differences among images.Firstly,it extracts all image sensor pattern noise and removes interpolation point to extract real image sensor pattern noise.Then it uses the product of image texture and lightness as the weight of each image sensor pattern noise in the camera reference pattern noise.Finally,it uses a weighted average to estimate camera reference pattern noise.The experimental results show that this algorithm achieves higher accuracy in image source detection compared with traditional camera reference pattern noise extraction algorithm.
-
Computer Engineering. 2017, 43(12): 231-236. https://doi.org/10.3969/j.issn.1000-3428.2017.12.042Aiming at the problem of low accuracy for the gait recognition in video surveillance,the bilateral Fourier calibration point estimation category Haar feature template matching based multiple lens gait recognition algorithm is proposed.According to the characteristics of video monitoring,the point estimation algorithm is used for image reconstruction and a markerless Haar feature template matching based feature extraction algorithm is designed to realize the automatic recognition and extraction of gait feature.According to the angles difference of feature extraction in motion,the bilateral Fourier is used for observation angle correction,and the adaptive sequential forward selection is designed for high dimensional feature extraction.Simulation results in the Southampton test database show that the correct classification rate of the proposed algorithm can reach about 96.3%,and through the observation angle correction,the gait motion angle can reach basically the same in different gait cycles.
-
Computer Engineering. 2017, 43(12): 237-241. https://doi.org/10.3969/j.issn.1000-3428.2017.12.043In order to improve the vanishing point detection effect of curved road,this paper proposes a novel vanishing point detection method in the unstructured road.This method uses the vertical projection rectification method to rectify the dominant Gabor texture orientation of a voter,then a voter utilizes its dominant texture orientation and edge information to increase weights of some voters above it,and realizes dominant orientation voting with its weight,chooses a candidate vanishing point who accepts the largest sum of voting values from votes to be the vanishing point to detect the vanishing point.The experimental results show that,compared with the existing vanishing point detection methods of unstructured road,the proposed method has lower running time and higher detection accuracy for the winding road.
-
Computer Engineering. 2017, 43(12): 242-247. https://doi.org/10.3969/j.issn.1000-3428.2017.12.044In order to solve image matching problem in 3D reconstruction of shoelast,this paper proposes an improved shoelast digitalization method.Shoelast surface texture is firstly gridded to form feature points,which are extracted by combining Harris and Laplace operators.The extracted feature points are precisely matched by an improved method of removing false match.The 3D coordinates of the matched feature points are obtained through binocular vision triangulation.Then,an Iterative Closest Point(ICP) algorithm is used on the point set with different directions for coordinate fusion and 3D mosaic.Finally,based on the whole point cloud data,the methods of triangulation and surface fitting are used for the 3D reconstruction of showlast.Experimental results show that this method can be conveniently and efficiently used for digital reconstruction of shoelast,which can be improved for personalized shoe-making demand.
-
Computer Engineering. 2017, 43(12): 248-254. https://doi.org/10.3969/j.issn.1000-3428.2017.12.045In order to overcome the low efficiency shortcoming of traditional Principal Component Analysis(PCA) feature extraction,this paper proposes a new face recognition method based on fast PCA dimensionality reduction algorithm which is able to accelerate the process of the eigenvalues and eigenvectors of calculating the sample covariance matrix.In the sense of cross-validation,this paper takes the recognition accuracy of Support Vector Machine(SVM) training model as the fitness value of the Particle Swarm Optimization(PSO),searches globally for the optimal values of penalty parameter and kernel function parameter of SVM,obtains the global optimal values of the parameters,and takes the values to train a final classifier model.Experimental results on the face images of ORL and Yale library show that the new method has higher feature extraction efficiency and recognition accuracy compared with the recognition method based on the traditional PCA and SVM algorithm.
-
Computer Engineering. 2017, 43(12): 255-260,266. https://doi.org/10.3969/j.issn.1000-3428.2017.12.046In order to avoid network congestion,and reduce the data flow,based on bilinear mapping,this paper proposes short proxy signature scheme with only one group element,including the signature definition,the construction of new scheme and the security analysis.A rigorous security proof is given under the random oracle model,which is based on the discrete logarithm,computational Diffie-Hellman and reverse computational Diffie-Hellman hardness assumption.Compared with the other existing security proxy signature schemes,the performance of the scheme is analyzed.Analysis results show that the proposed scheme doesn’t require any bilinear pairing operation in signature generation phase and requires just two bilinear pairing operations in signature verification phase,and it has the advantages of short signature,high efficiency,safety and so on.
-
Computer Engineering. 2017, 43(12): 261-266. https://doi.org/10.3969/j.issn.1000-3428.2017.12.047The autonomous vehicles make decisions according to their own information and lack effective interaction with human driving vehicles.When the autonomous vehicle is in complex traffic environment,its advantage is hard to play and even causes accidents.Based on the thought of intelligent connected vehicle,aiming at the lane changing behaviors in the driving process,this paper puts forward a lane-changing model by using game theory.Firstly,it gives full consideration to the behavior characteristics of the human lane changing.After the vehicle makes lane changing decision,it observes and analyzes the driving style of other vehicles,which affects the final decision.Secondly,the payoff function is determined according to the accelerating space in the premise of safety while adding the parameters expressing the driving style and quantifying the driving style scores.Finally,through the simulation by SUMO,the experimental results show that,by using game lane-changing model,the vehicles passing through increases by 5.60% and driving time decreases by 8.44% on average under different traffic densities,and no accident happens in the test.With the ability of perception and judgments of human driving style,traffic efficiency under the premise of safety is optimized.
-
Computer Engineering. 2017, 43(12): 267-273,277. https://doi.org/10.3969/j.issn.1000-3428.2017.12.048
The existing link estimation methods cannot guarantee the reliability of prediction and limitations large.To solve this problem,according to the similar current link information between the similar nodes of the source node and the similar nodes of the target nodes,the homophily connection principle is introduced,and a relatedness measure between different types of objects is designed to compute the existence probability of a link.It also extends conventional proximity measures to heterogeneous links.Furthermore,the labeled and unlabeled data in heterogeneous information networks are combined,and a heterogeneous collective link prediction algorithm is proposed to predict multiple types of links collectively by capturing the diverse and complex relationships among different types of links and leveraging the complementary prediction information.Empirical studies on real-world tasks demonstrate that the proposed collective link prediction approach can effectively boost link prediction performances in heterogeneous information networks.
-
Computer Engineering. 2017, 43(12): 274-277. https://doi.org/10.3969/j.issn.1000-3428.2017.12.049In Uniform Embedding Revisited Distortion(UERD) algorithm,the distortion function only considers the number of nonzero coefficients and does not take into account the influence of the statistical distribution on the embedding position,so this paper designs a new JPEG uniform embedding steganography algorithm,which combines with the concept of block entropy and improves the definition of distortion function.The secret information can be better embedded in the image regions with high complexity,such as edges and texture.Experimental results demonstrate that compared with BCH,EBS,UERD and SI-UNI algorithms,the embedding capacity and steganalysis capability of the proposed algorithm are greatly improved.
-
Computer Engineering. 2017, 43(12): 278-282,291. https://doi.org/10.3969/j.issn.1000-3428.2017.12.050Emotional speech synthesis is expected to make the synthesized speech more expressive.In order to synthesis more natural emotional speech signals,this paper proposes a new emotional speech synthesis method combining Pitch Synchronous Overlap Add(PSOLA) and Discrete Cosine Transform(DCT).The research builds up emotional rules for happy,sad,neutral speech.Through analyzing the prosody parameters,it can modify the each syllable of neutral speech’s fundamental frequency,energy and duration based on the emotional rules.The combination method adjusts pitch frequency for which marked pitch through DCT method,and then adjusts the pitch frequency to approach the target emotional fundamental frequency by the PSOLA algorithm.Experimental results show that the proposed method is more sensitive than the PSOLA algorithm.The subjective emotion recognition rate is higher,and the synthesized emotion speech quality is better.
-
Computer Engineering. 2017, 43(12): 283-291. https://doi.org/10.3969/j.issn.1000-3428.2017.12.051An arrival distribution fitting method of elevator passenger flow by using Hyper-Erlang Distribution(HErD) is proposed based on the maximum likelihood estimation.The relationship among the parameters estimated of HErD is analyzed based on regression.It is obtained that mixing ratio α and the scale parameter λ are both related to the squared coefficient of variation c2x.As a result,the parameters of HErD are reduced to one scale parameter determined by the first moment and c2x.Arrival time interval distributions of morning peak elevator passenger in Chengdu’s CBD is fitted by using HErD,exponential distribution,normal distribution,etc.The results show that HErD before and after reduction performs best and stable compared with other distributions.
-
Computer Engineering. 2017, 43(12): 292-295,302. https://doi.org/10.3969/j.issn.1000-3428.2017.12.052Bandwidth transfer is often required and deadline missing occurs when a real-time system changes its running mode.In order to solve this problem,based on the real-time task set scheduled with the earliest deadline first algorithm,the bandwidth transfer caused by new tasks’ insertion and current tasks’ acceleration is studied.According to the three stages of the old mode,the transition process,and the new mode,a model of multitask bandwidth transfer is presented.Some current tasks are compressed to release bandwidth.Even the system runs at full load,it is proved that deadline missing is only possible before the transition process as long as the sum of the utilizations(bandwidths) of all the tasks in the new mode is not greater than one,and thus the starting and ending points of the transition process are defined.Simulations results show that this model gives an overall description of the bandwidth transfer of multitask.
-
Computer Engineering. 2017, 43(12): 296-302. https://doi.org/10.3969/j.issn.1000-3428.2017.12.053
In order to improve the computational efficiency and performance of the community discovery algorithm,a semi supervised community discovery gradient descent algorithm based on latent space mapping is proposed.Firstly,a semi supervised community discovery framework based on latent space mapping is constructed based on the latent space representation,then the similarity of the latent space is evaluated by use of the square distance or KL divergence;Secondly,the semi supervised gradient descent optimization rules community discovery algorithm based on the norm of matrices and Frobenius is constructed to achieve the objective function of the local minimum quick access points,which improves the algorithm in large scale community found in practical value.Finally,the theoretical analysis of the computational complexity of the proposed algorithm is presented.Experimental results show that compared with the finding local community structure in networks method,Girvan-Newman networks,Label propagation and other 6 algorithms,the performance of the proposed algorithm is found to have a better performance of community discovery,and the effectiveness of the algorithm is verified.
-
Computer Engineering. 2017, 43(12): 303-308,314. https://doi.org/10.3969/j.issn.1000-3428.2017.12.054In order to solve the design problem of contact spring of circuit breaker,a Cloud Enhancement Differential Evolution (ECDE) optimization design method is proposed.Taking HSW6 series universal circuit breakers as an example,the design and analysis of A,B and C’s contact springs are carried out.The mathematical model of contact spring is constructed and solved by cloud ECDE.Experimental results show that the proposed method can obtain the optimal solution of contact spring model quickly and effectively compared with the differential evolutionary algorithm.
-
Computer Engineering. 2017, 43(12): 309-314. https://doi.org/10.3969/j.issn.1000-3428.2017.12.055In order to obtain the navigation information of orchard mobile robot and display in real time,a method to determine the type of road centerline is proposed and an orchard path recognition system is designed.The image binarization threshold is determined by analyzing the H component of the sixteen hue ring in the HSV color space.For pre-processed binary image,the line dislocation point is obtained by line scanning point detection algorithm and the straight line is fitted.The mean and variance distribution of the discrete point to the fitted straight line distance is analyzed when straight road and curved road are used as samples.The criterion for identifying the type of center line of orchard road is determined and the center line is fitted according to the criterion.In order to realize the visualization of the system,a hybrid programming of Matlab and VB is used to display the navigation information in the system interface of VB design in real time.Experimental results show that compared with the Matlab-GUI visual navigation system using the vertical projection method,the navigation information of the system is faster and the road recognition success rate is higher.
-
Computer Engineering. 2017, 43(12): 315-320. https://doi.org/10.3969/j.issn.1000-3428.2017.12.056Among the existing short-wave communication systems with predistortion,the predistortion processing is commonly highly integrated with the baseband signal processing of the exciter and cannot be applied directly in the existing short-wave communication systems,or it requires drastic changes on the existing systems.Aiming at the problem,this paper proposes a plug-in predistortion technology,and with the hardware platform of DSP+FPGA,the Plug-in Predistortion Device (PIPD) is designed and realized.The PIPD is cascaded between the exciter and Power Amplifier (PA),The DSP controls the entire system as well as the predistortion algorithm portion.FPGA completes the digital mixer digital filtering and predistortion compensation in the normal communication and measurement both modes.The ARM control panel integrated Ethernet and PC plug-in functions achieves predistortion short-wave communications network port.Experimental results show that the plug-in short-wave predistortion device can improve intermodulation index to beneath full-band -36 dB at only 1/3 frequency points.
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