Table of Content

10 May 2018, Volume 38 Issue 5

Previous Issue    Next Issue

Review of human activity recognition based on wearable sensors

ZHENG Zengwei, DU Junjie, HUO Meimei, WU Jianzhong

2018, 38(5):  1223-1229.  DOI: 10.11772/j.issn.1001-9081.2017112715

Asbtract ( )   PDF (1238KB) ( )  

References | Related Articles | Metrics

Human Activity Recognition (HAR) has a wide range of applications in medical care, safety, and entertainment. With the development of sensor industry, sensors that can accurately collect human activity data have been widely used on wearable equipments such as wristband, watch and mobile phones. Compared with the behavior recognition method based on video images, sensor-based behavior recognition has the characteristics of low cost, flexibility and portability. Therefore, human activity recognition research based on wearable sensors has become an important research field. Data collection, feature extraction, feature selection and classification methods of HAR were described in detail, and the techniques commonly used in each process were analyzed. Finally, the main problems of HAR and the development directions were pointed out.

Nonparametric approximation policy iteration reinforcement learning based on Dyna framework

JI Ting, ZHANG Hua

2018, 38(5):  1230-1238.  DOI: 10.11772/j.issn.1001-9081.2017102531

Asbtract ( )   PDF (1297KB) ( )  

References | Related Articles | Metrics

In order to solve the problem that the approximator of the current approximation policy iteration reinforcement learning cannot be constructed completely automatically, a reinforcement learning algorithm of Nonparametric Approximation Policy Iteration based on Dyna Framework (NPAPI-Dyna) was proposed. Sampling cache and sampling change rate were introduced to design a two stage random sampling process to collect samples. By profile tolerance and K-means clustering, core state basis function was generated through trial-and-error process. Q-value function approximator was generated by using the complete coverage of sample as the target. Greedy strategy was applied to design action selector. Access frequency of the state basis function was used to describe environmental topology features and construct environment estimation model. Learning and planning processes were combined organically by identification of Dyna framework to accelerate the speed of learning.In the simulation experiments of single inverted pendulum balance control, when the reinforcement learning error rate is 0.01, the learning success rate of algorithm reaches 100%, the minimum number of successful attempts is only 2, the average number of attempts is only 7.73, and the mean absolute deviation of angle is 3.0538°, and the average oscillation range of angle is 2.759°. When reinforcement learning error rate is 0.1, 100 independent simulation operations are performed, to learn the control strategy, Online-LSPI and BLSPI (Batch Least-Squares Policy Iteration) have to try more than 150 times on average, however NPAPI-Dyna can succeed in 50 times of attempts. The experimental results show that NPAPI-Dyna can be completely automatically constructed and adjusted to enhance the learning structure, with high learning accuracy and rapid convergence ability.

Perturbation particle swarm optimization algorithm based on local far-neighbor differential enhancement

WANG Yonggui, HU Caiyun, LI Xin

2018, 38(5):  1239-1244.  DOI: 10.11772/j.issn.1001-9081.2017102557

Asbtract ( )   PDF (1070KB) ( )  

References | Related Articles | Metrics

To solve the problems that Particle Swarm Optimization (PSO) algorithm is easy to fall into the local extremum due to the lack of interaction between individuals in the search process, the diversity of the population is gradually lost, a Perturbation Particle Swarm Optimization algorithm based on Local Far-neighbor Differential Enhancement (LFDE-PPSO) was proposed. Firstly, in order to enlarge the population search space, the disturbance factor was introduced to make inertia weight and learning factor fluctuate within a small range. Secondly, the reconstruction probability was introduced, and the population with low fitness value was selected to reconstruct intermediate population. Finally, in order to increase the population diversity, the excellent individuals of poor individuals were retained, the irrelevant and far-neighbor individuals were introduced. The far-neighbors with large differences from differential individual genes were used for differential enhancement. The experimental results show that the proposed algorithm can preserve individuals with high fitness in the intermediate population, effectively increase the population diversity, make the population have strong ability to jump out of local extremum, speed up the particle approximation to the global aptimum, and have the advantages of fast convergence and high precision.

Hybrid feature selection algorithm fused Shapley value and particle swarm optimization

DENG Xiuqin, LI Wenzhou, WU Jigang, LIU Taiheng

2018, 38(5):  1245-1249.  DOI: 10.11772/j.issn.1001-9081.2017112730

Asbtract ( )   PDF (774KB) ( )  

References | Related Articles | Metrics

Concerning the problem that data often has irrelevant or redundant features which affect the classification accuracy in pattern classification problems, a hybrid feature selection method based on Shapley value and Particle Swarm Optimization (PSO) was proposed to obtain the best classification results with the fewest features. Firstly, the Shapley value of game theory was introduced into the local search of PSO algorithm. Then,by calculating the Shapley value of each feature in the particle (feature subset), the feature with the lowest Shapley value was gradually deleted to optimize the feature subset and update the particle, and enhance the global search ability of the algorithm at the same time. Finally, the improved particle swarm algorithm was applied to feature selection. The classification performance and the number of selected features of the support vector machine classifier were used as feature subset evaluation criteria. The classification experiments were performed on 17 medical data sets with different characteristic quantities of UCI machine learning data sets and gene expression data sets. The experimental results show that the proposed algorithm can remove more than 55% irrelevant or redundant features in the datasets effectively, especially more than 80% in the medium and large datasets, and the selected feature subset also has better classification ability,the classification accuracy can be increased by 2 to 23 percentage points.

Multi-objective decision making based on entropy weighted-Vague sets

ZHAO Qingqing, HUANG Tianmin

2018, 38(5):  1250-1253.  DOI: 10.11772/j.issn.1001-9081.2017112645

Asbtract ( )   PDF (540KB) ( )  

References | Related Articles | Metrics

In view of the subjective arbitrariness of objective weight in multi-objective decision making based on Vague sets and the monotong problem of evaluation function, a novel approach to multi-objective decision making based on entropy weighted-Vague sets was presented. Firstly, the decision matrix was transformed into the objective-grade-membership matrix. Then the objective weight of each objective was calculated by entropy coefficient method, and the weight vector interval of each objective was obtained by considering objective weight and subjective weight. Next the Vague evaluation was obtained by computing the sets of objectives being in favor, against and neutral. Finally, a new evaluation function was defined to sort the alternatives and select the optimal scheme. The rationality and effectiveness of the method were verified by an example.

Multi-population-based competitive differential evolution algorithm for dynamic optimization problem

YUAN Yichuan, YANG Zhou, LUO Tingxing, QIN Jin

2018, 38(5):  1254-1260.  DOI: 10.11772/j.issn.1001-9081.2017102552

Asbtract ( )   PDF (1051KB) ( )  

References | Related Articles | Metrics

To solve Dynamic Optimization Problems (DOP), a Differential Evolution algorithm with Competitive Strategy based on multi-population (DECS) was proposed. Firstly, one of the populations was chosen as a detection population. Whether the environment had changed was determined by monitoring the fitness values of all individuals in the population and dimension of the population. Secondly, the remaining populations were used as the search populations to search the optimal value independently. During the search, a exclusion rule was introduced to avoid the aggregation of multiple search populations in the same local optimal neighborhood. After the iteration of several generations, competitive operation was performed on all search populations. The population to which the optimal individual belong was retained and the next generation's individuals of the population were generated by using the quantum individual generation mechanism. Then other search populations were reinitialized. Finally, 49 dynamic change problems about 7 test functions were used to verify DECS, and the experimental results were compared with Artificial Immune Network for Dynamic optimization (Dopt-aiNet) algorithm, restart Particle Swarm Optimization (rPSO) algorithm, and Modified Differential Evolution (MDE) algorithm. The experimental results show that the average error mean of 34 problems for DECS is less than Dopt-aiNet and the average error mean of all problems for DECS was less than that for rPSO and MDE. Therefore, DECS is feasible to solve DOP.

Joint sentiment/topic model integrating user characteristics

XU Yinjie, SUN Chunhua, LIU Yezheng

2018, 38(5):  1261-1266.  DOI: 10.11772/j.issn.1001-9081.2017112709

Asbtract ( )   PDF (1024KB) ( )  

References | Related Articles | Metrics

The Joint Sentiment/Topic (JST) model can extract both the topic and the sentiment from the text, but the existing JST model mainly focuses on textual content, without considering the user characteristics, which may lead to demographic and event biases in sentiment mining reports. The Joint-User Sentiment/Topic (JUST) model was proposed. The main improvement of the JUST model was that the user characteristics were added to the model, a linear function of the user characteristics corresponding to the document was used as a priori of the document-emotional distribution, so the model could get emotional tendencies of different topics from customer with different characteristics. The validity of the JUST model was tested on the datasets of 13252 automobile review from autohome.com (www.autohome.com.cn). The experimental results show that the accuracy of the sentiment classification of the JUST model is higher than those of the JST model and TSMMF (Topic Sentiment Model based on Multi-feature Fusion) model. The topic and sentiment differences between users with different characteristics were also compared.

Identification of micro-blog advertising publisher based on clustering analysis

ZHAO Xingyu, ZHAO Zhihong, WANG Yepei, CHEN Songyu

2018, 38(5):  1267-1271.  DOI: 10.11772/j.issn.1001-9081.2017102478

Asbtract ( )   PDF (772KB) ( )  

References | Related Articles | Metrics

There is a large amount of advertising content in micro-blog space, which seriously affects user experience and related research work. Much of existing research on micro-blog process uses classification algorithm such as Support Vector Machine (SVM) and random forest algorithm. However, it is difficult to classify a large volume of data in the classification method manually. A micro-blog advertisement publisher identification method based on clustering analysis was proposed. For user dimension, a concept of core micro-blog was put forward to deal with the phenomenon that ordinary micro-blogs were posted to dilute advertising content. Then the extracted main themes of each user and corresponding micro-blog sequences could be used to calculate user characteristics as well as the text characteristics. After that, a clustering algorithm was used to cluster the features and identify the micro-blog advertisers. The experiment result shows that the precision is 93%, the recall is 97%, and the F value is 95%, which proves that the proposed method can accurately identify the micro-blog advertisement publisher under the condition that the content of the advertisement is artificially diluted. It provides theoretical support and practical methods for the recognition and cleaning work of micro-blog spam information.

Long text classification combined with attention mechanism

LU Ling, YANG Wu, WANG Yuanlun, LEI Zijian, LI Ying

2018, 38(5):  1272-1277.  DOI: 10.11772/j.issn.1001-9081.2017112652

Asbtract ( )   PDF (946KB) ( )  

References | Related Articles | Metrics

News text usually consists of tens to hundreds of sentences, which has a large number of characters and contains more information that is not relevant to the topic, affecting the classification performance. In view of the problem, a long text classification method combined with attention mechanism was proposed. Firstly, a sentence was represented by a paragraph vector, and then a neural network attention model of paragraph vectors and text categories was constructed to calculate the sentence's attention. Then the sentence was filtered according to its contribution to the category, which value was mean square error of sentence attention vector. Finally, a classifier base on Convolutional Neural Network (CNN) was constructed. The filtered text and the attention matrix were respectively taken as the network input. Max pooling was used for feature filtering. Random dropout was used to reduce over-fitting. Experiments were conducted on data set of Chinese news text classification task, which was one of the shared tasks in Natural Language Processing and Chinese Computing (NLP&CC) 2014. The proposed method achieved 80.39% in terms of accuracy for the filtered text, which length was 82.74% of the text before filtering, yielded an accuracy improvement of considerable 2.1% compared to text before filtering. The emperimental results show that combining with attention mechanism, the proposed method can improve accuracy of long text classification while achieving sentence level information filtering.

Joint Chinese word segmentation and punctuation prediction based on improved multilayer BLSTM network

LI Yakun, PAN Qing, WANG Feng

2018, 38(5):  1278-1282.  DOI: 10.11772/j.issn.1001-9081.2017112631

Asbtract ( )   PDF (903KB) ( )  

References | Related Articles | Metrics

The current mainstream sequence labeling is based on Recurrent Neural Network (RNN). Aiming at the problem of RNN and sequence labeling, an improved multilayer Bi-direction Long Short Term Memory (BLSTM) network for sequence labeling was proposed. Each layer of BLSTM had an operation of information fusion, and the output contained more contextual information. In addition, a method to perform Chinese word segmentation and punctuation prediction jointly was proposed. Experiments on the public datasets show that the improved multilayer BLSTM network model can improve the classification accuracy of Chinese segmentation and punctuation prediction. In the case of two tasks that need to be accomplished, the joint task method can greatly reduce the complexity of the system, and the new model and the joint task method can also be applied to solve other sequence labeling problems.

Reordering table reconstruction model for Chinese-Uyghur machine translation

PAN Yirong, LI Xiao, YANG Yating, MI Chenggang, DONG Rui

2018, 38(5):  1283-1288.  DOI: 10.11772/j.issn.1001-9081.2017102455

Asbtract ( )   PDF (934KB) ( )  

References | Related Articles | Metrics

Focused on the issue that lexicalized reordering models are faced with context independence and sparsity problems in machine translation, a reordering table reconstruction model based on semantic content for reordering orientation and probability prediction was proposed. Firstly, continuous distributed representation approach was employed to acquire the feature vectors of reordering rules. Secondly, Recurrent Neural Networks (RNN) were utilized to predict the reordering orientation and probability of each reordering rule that represented with dense vectors. Finally, the original reordering table was filtered and reconstructed with more reasonable reordering probability distribution for the purpose of improving the reordering information accuracy in reordering model as well as reducing the size of the reordering table to speed up subsequent decoding process. The experimental results show that the reordering table reconstruction model can provide BLEU point gains (+0.39) for Chinese to Uyghur machine translation task.

Improved explicit shape regression for face alignment algorithm

JIA Xiangnan, YU Fengqin, CHEN Ying

2018, 38(5):  1289-1293.  DOI: 10.11772/j.issn.1001-9081.2017102586

Asbtract ( )   PDF (862KB) ( )  

References | Related Articles | Metrics

To solve the problem that Explicit Shape Regression (ESR) has low precision in face alignment, an improved explicit shape regression for face alignment algorithm was proposed. Firstly, in order to get a more accurate initial shape, three-point face shape was used as an initial shape mapping standard to replace face rectangle. Then, pixel block feature was used against illumination variations instead of pixel feature, which improved the algorithm robustness. Finally, instead of average method, the accuracy of algorithm was further improved by multiple hypothesis fusion strategy which merged multiple estimations. Compared with explicit shape regression algorithm, the simulation experimental results show that the accuracy is improved by 7.96%, 5.36% and 1.94% respectively by using the proposed algorithm on LFPW, HELEN and 300-W face datasets.

Automatic image annotation based on multi-label discriminative dictionary learning

YANG Xiaoling, LI Zhiqing, LIU Yutong

2018, 38(5):  1294-1298.  DOI: 10.11772/j.issn.1001-9081.2017112650

Asbtract ( )   PDF (930KB) ( )  

References | Related Articles | Metrics

Concerning the problem of semantic gap between low-level visual features and high-level semantics in automatic image annotation, based on traditional dictionary learning, a multi-label discriminative dictionary learning method was proposed to automatic image annotation. First of all, multiple types of features for each image were extracted, and a combination of a variety of features was used as input information of the input feature space to the dictionary learning. Then, a label consistency regularization term was designed to integrate the label information of the original samples into the initial input feature data, and the dictionary of label consistency and the label consistency regularization term were combined to learn the dictionary. Finally, the label sparse coding vector was obtained by the dictionary and sparse coding matrix to implement the semantic annotation for an unknown image. The performance of the annotation was tested on the Corel 5K data set. The average precision and average recall could reach 35% and 48% respectively, compared with the traditional Sparse Coding Method (MSC), which were increased by 10 percentage points and 16 percentage points respectively, and increased by 3 percentage points and 14 percentage points respectively than the method of Distance Constraint Sparse/Group Sparse Coding (DCSC/DCGSC) for automatic image lableing. Compared with the current image annotation methods, the experimental results show the proposed method can predict the semantic information for an unknown image properly, and has better annotation performance.

Vehicle re-identification algorithm based on bag of visual words in complicated environments

WANG Qian, CHEN Yimin, DING Youdong

2018, 38(5):  1299-1303.  DOI: 10.11772/j.issn.1001-9081.2017102581

Asbtract ( )   PDF (758KB) ( )  

References | Related Articles | Metrics

To meet the demands of the public security department to search out specific target in complicated real environment, the target re-IDentification (re-ID) technique was introduced in vehicle identification field, and a vehicle re-ID algorithm based on the model of Bag of Visual Words (BoVW) was proposed. Firstly, in order to solve the probloms of occlusion pose change, target size and position difference in images, the improved scales and poses adaptive Part-based One-vs-One Feature (POOF) were extracted. Secondly, a set of visual words was clustered as a vocabulary by using k-means algorithm based on Euclidean distance, and the features of each image (or target) were expressed as the composition of visual vocabularies. Thirdly, the improved Keep It Simple and Straightforward Metric (KISSME) method followed with re-rank step was used to separate the between-classes and within-classes distances. Finally, the result was obtained by using K-Nearest Neighbor (KNN) method. The experimental results show that the algorithm has 3.85 percentage points increasement of identification rate in feature representation step compared with Bubble Bank (BB) and 3.14 percentage points increase in metric learning step compared with Bayesian face revisited. Furthermore, it is proved that the proposed algorithm is economical in time-consuming and has strong adaptability to target pose change and small portion of occlusion, which further domonstrates that it can adapt to complicated environments.

Aquatic animal image classification method based on transfer learning

WANG Keli, YUAN Hongchun

2018, 38(5):  1304-1308.  DOI: 10.11772/j.issn.1001-9081.2017102487

Asbtract ( )   PDF (949KB) ( )  

References | Related Articles | Metrics

Aiming at the problems that traditional aquatic animal image recognition methods have complex steps, low accuracy and poor generalization, and it is difficult to develop Deep Convolutional Neural Network (DCNN) model, a method based on parameter transfer strategy using fine-tune to retrain pre-trained model was proposed. Firstly, the image was preprocessed by data enhancement and so on. Secondly, on the basis of modifying the source model's fully connected classification layer, the weights of high-level convolution modules were set to be trained for adaptive adjustment. Finally, using training time and recognition accuracy on validation set as the evaluation indexes, the performance experiments were conducted on various network structures and different proportion of trainable parameters. The experimental results show that the highest retrained model classification accuracy can reach 97.4%, 20 percentage points higher than the source model, the ideal performance can be obtained when the proportion of trainable parameters is around 75%. It is proved that the fine-tune method can obtain a deep neural network image classification model with good performance under low-cost development condition.

Multi-perspective multi-region feature fusion for apple classification

LIU Yuanyuan, WANG Hui, GUO Gongde, JIANG Nanfeng

2018, 38(5):  1309-1314.  DOI: 10.11772/j.issn.1001-9081.2017102412

Asbtract ( )   PDF (965KB) ( )  

References | Related Articles | Metrics

Since manual sorting of apples is a huge project in our daily life, an apple image classification approach based on multi-perspective multi-region feature fusion was proposed. First of all, five classes of apples, containing 329 in total, were collected. For each apple, five images from five different perspectives were obtained:top, bottom, side1, side2 and side3. From each image, several (one to nine) small image regions were cut. Secondly, each region block was represented by color histogram vector, and the histogram vectors of region blocks were fused together end to end to generate a representation of the image. Finally, 12 classifiers were used to classify 329 samples. The experimental results show that the multi-perspective multi-region based method significantly outperforms single-perspective single-region based method, and the more the number of perspectives/regions, the better the result. In particular, classification performance reaches 97.87% by PLS (Partial Least Squares) even better than deep learning when using nine regions for each image cropped at five angles. The method is easy but efficient, whose computation complexity is 4n, where n is the total number of blocks in image cropping area. Thus, it can be applied to mobile applications and applied to more fruit and plant image classification.

Two-level confidence threshold setting method for positive and negative association rules

CHEN Liu, FENG Shan

2018, 38(5):  1315-1319.  DOI: 10.11772/j.issn.1001-9081.2017102469

Asbtract ( )   PDF (873KB) ( )  

References | Related Articles | Metrics

Aiming at the problem that traditional confidence threshold setting methods for positive and negative association rules are difficult to limit the number of low-reliability rules and easy to miss some interesting association rules, a new two-level confidence threshold setting method combined with the rule's itemset correlation was proposed, called PNMC-TWO. Firstly, taking into account the consistency, validity and interestingness of rules, under the framework of correlation-support-confidence, on the basis of the computation relationship between rule confidence and itemset support of the rule, the law of confidence of rule changing with support of itemsets of the rule was analyzed systematically. And then, combined with the user's requirement of high confidence and interesting rules in actual mining, a new confidence threshold setting model was proposed to avoid the blindness and randomness of the traditional methods when setting the threshold. Finally, the proposed method was compared with the original two-threshold method in terms of the quantity and quality of the rule. The experimental results show that the new two-level threshold method not only can ensure that the extracted association rules are more effective and interesting, but also can reduce the number of low-reliability rules significantly.

Fast label propagation algorithm based on node centrality and community similarity

GU Junhua, HUO Shijie, WANG Shoubin, TIAN Zhe

2018, 38(5):  1320-1326.  DOI: 10.11772/j.issn.1001-9081.2017102927

Asbtract ( )   PDF (1061KB) ( )  

References | Related Articles | Metrics

In order to reduce unnecessary update and solve the problem of low accuracy and poor stability of Label Propagation Algorithm (LPA), a Fast Label Propagation Algorithm based on Node Centrality and Community Similarity (FNCS_LPA) was proposed. According to the node centrality measure, the nodes of a network were sorted from low to high and added into node information list, which guided the update process to avoid unnecessary update and improve the stability of community detection. The accuracy of community detection was improved by a new update rule based on community similarity. Experiments were tested on a real social networks and LFR benchmarks. Compared with LPA and three improved LPA algorithms, the execution speed is improved by almost a dozen times, the modularities of the real social networks and the Normalized Mutual Information (NMI) of LFR (Lancichinetti Fortunato Radicchi) benchmark networks with more obscure community structure were significantly improved. The experimental results show that FNCS_LPA improves the accuracy and stability of community detection on the basis of improving execution speed.

Similarity search based on semantic features of bibliographic information network

QIU Qingyu, LI Jing, QUAN Bing, TONG Chao, ZHANG Lijun, ZHANG Haixian

2018, 38(5):  1327-1333.  DOI: 10.11772/j.issn.1001-9081.2017112623

Asbtract ( )   PDF (1169KB) ( )  

References | Related Articles | Metrics

Bibliography information network is a typical heterogeneous information network and the similarity search based on it is a hot topic of graph mining. However, current methods mainly adopt meta path or meta structure to search similar objects, do not consider semantic features of node itself which leads to a deviation in the search results. To fill this gap, a vector-based semantic feature extraction method was proposed, and a vector-based node similarity calculation method called VSim was designed and implemented. In addition, a similarity search algorithm VPSim (Similarity computation Based on Vector and meta Path) based on semantic features was designed by combining the meta-paths. In order to improve the execution efficiency of the algorithm, a pruning strategy based on the characteristics of bibliographic network data was designed. Experiments on real-world data sets demonstrate that VSim is applicative for searching entities with similar semantic features and VPSim is effective, efficient and extensible.

Multi-source point of interest fusion algorithm based on distance and category

XU Shuang, ZHANG Qian, LI Yan, LIU Jiayong

2018, 38(5):  1334-1338.  DOI: 10.11772/j.issn.1001-9081.2017102504

Asbtract ( )   PDF (748KB) ( )  

References | Related Articles | Metrics

In order to achieve effective integration and accurate fusion of multi-source Point of Interest (POI) data, a Mutually-Nearest Method considering Distance and Category (MNMDC) was proposed. Firstly, for spatial attributes, standardized weight algorithm was used to calculate the spatial similarity of the object to be fused, and the fusion set was obtained. Secondly, for non-spatial attributes, Jaro-Winkle algorithm was used to eliminate some objects with consistent categories by a low threshold, and exclude some objects with inconsistent categories by a high threshold. Finally, non-spatial Jaro-Winkle algorithm with distance constraint, category consistency constraint and high threshold was used to find out the missing objects in the spatial algorithm. The experimental results show that the average accuracy reaches 93.3%, compared with Combined Normal Weight and Title-similatity algorithm (COM-NWT) and the grid correction methods, the accuracy of MNMDC method in seven different groups of coincidence degree data, the average accuracy increases by 2.7 percentage points and 1.6 percentage points, the average recall increases by 2.3 and 1.4 percentage points. The MNMDC method allows more accurate fusion of POI data during actual fusion.

Storage method for flight delay platform based on HBase and Hive

WU Renbiao, LIU Chao, QU Jingyi

2018, 38(5):  1339-1345.  DOI: 10.11772/j.issn.1001-9081.2017102475

Asbtract ( )   PDF (1151KB) ( )  

References | Related Articles | Metrics

In the view of the problem that the portability and expansibility current flight delay platform in China can not adapt to the status of large data storage brought by rapid development of Chinese civil aviation, a flight delay big data platform with cross platform, high availability and high expansion was designed. The platform used a big data tool LeafLet as a visual carrier, displayed the flight trajectory in the map interface, and loaded trajectory data to HBase database. Message-Digest Algorithm (MD5) algorithm was used to redesign and optimize the rowkey of flight data table to solve its "hot spot" problem brought by incremental flight time. Considering the shortcomings of multi-level query of HBase filter, a query algorithm based on SolrCloud was proposed, which utilized SolrCloud to realize hierarchical storage of row and index fields, so as to realize HBase two-level fast indexing. Finally, based on historical flight data and flight plan data of HBase, a massive flight information data warehouse based on Hive was constructed. The experimental results show that the expensibility of large data platform for flight delays and the construction of flight information data warehouse can meet the demand of civil aviation for unified storage of data, and the response speed of the multi-condition query is improved by hundreds of times compared with the cluster without second index, and this advantage becomes more and more obvious as the flight data amount grows.

Outlier detection in time series data based on heteroscedastic Gaussian processes

YAN Hong, YANG Bo, YANG Hongyu

2018, 38(5):  1346-1352.  DOI: 10.11772/j.issn.1001-9081.2017102511

Asbtract ( )   PDF (1092KB) ( )  

References | Related Articles | Metrics

Generally, there are inevitable disturbances in time series data, such as inherent uncertainties and external interferences. To detect outlier in time series data with time-varying disturbances, an approach based on prediction model using Gaussian Processes was proposed. The monitoring data was decomposed into two components:the standard value and the deviation term. As the basis of model for the ideal standard value without any deviation, Gaussian processes were also employed to model the heteroscedastic deviations. The posterior distribution of predicted data which is analytically intractable after introducing deviation term was approximated by variational inference. The tolerance interval selected from posterior distribution was used for outlier detection. Verification experiments were conducted on the public time series datasets of network traffic from Yahoo. The calculated tolerance interval coincided with the actual range of reasonable deviation existing in labeled normal data at various time points. In the comparison experiments, the proposed model outperformed autoregressive integrated moving average model, one-class support vector machine and Density-Based Spatial Clustering of Application with Noise (DBSCAN) in terms of F1-score. The experimental results show that the proposed model can effectively describe the distribution of normal data at various time points, achieve a tradeoff between false alarm rate and recall, and avoid the performance problems caused by improper parameter settings.

Scheduling method of virtual cipher machine based on entropy weight evaluation in cryptography cloud

WANG Zewu, SUN Lei, GUO Songhui, SUN Ruichen

2018, 38(5):  1353-1359.  DOI: 10.11772/j.issn.1001-9081.2017102465

Asbtract ( )   PDF (1112KB) ( )  

References | Related Articles | Metrics

To balance load in cryptography cloud systems, a Virtual cipher machine Scheduling Method based on Entropy Weight Evaluation (VSMEWE) was proposed. In order to improve the quality of cryptography service and economize resources of cryptography cloud effectively, a virtual cipher machine migration selection solution was presented, according to the comparison results of comprehensive evaluation values of cloud cipher machine. To achieve the best comprehensive evaluation values, it evaluated the resource states of cloud cipher machine with the main indexes including the utilizations of resources, such as CPU, memory, network bandwidth and throughput bandwidth of cipher card. Finally, a migration selection scheme of virtual cipher machine was decided by the scheduling method. Compared with Entropy algorithm and Baseline algorithm, the experimental results show that the proposed algorithm has characteristics of wholeness and chronergy, the effect of load balancing is improved, and the execution efficiency is increased by 6.8% and 22.7% respectively.

Intrusion detection algorithm of industrial control network based on improved one-class support vector machine

LIU Wanjun, QIN Jitao, QU Haicheng

2018, 38(5):  1360-1365.  DOI: 10.11772/j.issn.1001-9081.2017102502

Asbtract ( )   PDF (1127KB) ( )  

References | Related Articles | Metrics

Since the intrusion detection method based on One-Class Support Vector Machine (OCSVM) can not detect internal abnormal points and outliers, which leads to the deviation of decision function from training samples. A new OCSVM anomaly detection function combining DBSCAN (Density-Based Spatial Clustering of Applications with Noise) and K-means was proposed. Firstly, the outliers in the training data were removed by DBSCAN algorithm to eliminate the influence of outliers. Then, K-means clustering method was used to classify normal data clusters, so that the internal abnormal points could be selected. Finally, a one-class classifier for each data cluster was created to detect exception data by OCSVM algorithm. The experimental results on industrial control networks show that the combined classifier can detect the intrusion attacks of the industrial control network by using normal data, and it can improve the detection effect of OCSVM algorithm. In intrusion detection experiment of gas pipeline, the overall detection rate of the proposed method is 91.81%, while the overall detection rate of OCSVM algorithm is 80.77%.

Dual game model of advanced persistent threat attack for distributed network structure

ZHANG Wei, SU Yang, CHEN Wenwu

2018, 38(5):  1366-1371.  DOI: 10.11772/j.issn.1001-9081.2017102448

Asbtract ( )   PDF (886KB) ( )  

References | Related Articles | Metrics

Considering the lack of theoretical analysis for distributed network structure under Advanced Persistent Threat (APT) attacks, a game model was proposed to solve the problem based on Nash equilibrium theory and node game theory. Firstly, a defensive framework of network security was established by analyzing the characteristics of APT attack and distributed network structure. Secondly, risky factor of vulnerability was calculated through node game model, and Oriented to APT (OAPG) was established on the basis of Nash equilibrium theory, the balance point of attack and defense was calculated, maximum return strategy of the attacker was analyzed, and then optimal defense strategy of the defender was proposed. Finally, an APT attack instance was used to verify the model. The calculation results show that the proposed model can analyze both the attack and the defense of the network from the APT attack path and provide a reasonable defense idea for the organizations using the distributed network.

Stateful group rekeying scheme with tunable collusion resistance

AO Li, LIU Jing, YAO Shaowen, WU Nan

2018, 38(5):  1372-1376.  DOI: 10.11772/j.issn.1001-9081.2017102413

Asbtract ( )   PDF (914KB) ( )  

References | Related Articles | Metrics

Logical Key Hierarchy (LKH) protocol has been proved that O(log n) is the lower bound of the communication complexity when resisting complete collusion attacks. However, in some resource-constrained or commercial application environments, user still require the communication overhead below O(log n). Although Stateful Exclusive Complete Subtree (SECS) protocol has the characteristic of constant communication overhead, but it can only resist single-user attacks. Considering the willingness of users to sacrifice some security to reduce communication overhead, based on LKH which has the characteristic of strict confidentiality, and combined with SECS which has constant communication overhead, a Hybrid Stateful Exclusive Complete Subtree (H-SECS) was designed and implemented. The number of subgroups was configured by H-SECS according to the security level of application scenario to make an optimal tradeoff between communication overhead and collusion resistance ability. Theoretical analysis and simulation results show that, compared with LKH protocol and SECS protocol, the communication overhead of H-SECS can be regulated in the ranges between O(1) and O(log n).

Fully homomorphic encryption scheme based on learning with errors under multi-attribute environment

BAI Ping, ZHANG Wei

2018, 38(5):  1377-1382.  DOI: 10.11772/j.issn.1001-9081.2017102568

Asbtract ( )   PDF (927KB) ( )  

References | Related Articles | Metrics

Learning With Errors (LWE)-based fully homomorphic encryption scheme was presented by Gentry, Sahai and Waters (GENTRY C, SALAHAI A, WATERS B. Homomorphic encryption from learning with errors:conceptually-simpler, asymptotically-faster, attribute-based[C]//Proceedings of the 33rd Annual Cryptology Conference. Berlin:Springer, 2013:75-92), namely GSW scheme, can only work under single-attribute settings. Aiming at this problem and introducing the concept of fully system, a fully homomorphic encryption scheme under multi-attribute settings was constructed. In the proposed scheme, whether a user was legitimate was determined through a conditional equation. Then, a new ciphertext matrix that meeting the requirements of GSW13 was constructed by using ciphertext expansion algorithm. Finally fuzzy system technology was used to complete the construction. INDistinguishability-X-Chosen Plain Attack (IND-X-CPA) security was proved under the standard model. The advantage of the proposed scheme lies in that it can be used in multi-attribute environment. The disadvantage is that the computational complexity is increased.

Password strength estimation model based on ensemble learning

SONG Chuangchuang, FANG Yong, HUANG Cheng, LIU Liang

2018, 38(5):  1383-1388.  DOI: 10.11772/j.issn.1001-9081.2017102516

Asbtract ( )   PDF (850KB) ( )  

References | Related Articles | Metrics

Focused on the issue that the existing password evaluation models cannot be used universally, and there is no evaluation model applicable from simple passwords to very complex passwords. A password evaluation model was designed based on multi-model ensemble learning. Firstly, an actual password training set was used to train multiple existing password evaluation models as the sub-models. Secondly, a multiple trained evaluation sub-models were used as the base learners for ensemble learning, and the ensemble learning strategy which designed to be partial to weakness, was used to get all advantages of sub-models. Finally, a common password evaluation model with high accuracy was obtained. Actual user password set that leaked on the network was used as the experimental data set. The experimental results show that the multi-model ensemble learning model used to evaluate the password strength of different complexity passwords, has a high accuracy and is universal. The proposed model has good applicability in the evaluation of passwords.

Portable operating system interface of UNIX compatibility technology in mass small distributed file system

CHEN Bo, HE Lianyue, YAN Weiwei, XU Zhaomiao, XU Jun

2018, 38(5):  1389-1392.  DOI: 10.11772/j.issn.1001-9081.2017102934

Asbtract ( )   PDF (791KB) ( )  

References | Related Articles | Metrics

Focused on the issue that the mass small file system developed based on HDFS (Hadoop Distributed File System), SMDFS (Mass Small Distributed File System), is not compatible with POSIX (Portable Operating System Interface of UNIX) constraints, a POSIX compatible technology based on local cache and an efficient metadata management technology based on temporary data cache were proposed. Firstly, the data storage area was set to realize the redirection of the file flow in the read-write mode, and then an asynchronous thread pool model was established to synchronize the data in temporary cache, thereby completing all POSIX-related file operations from the user layer to the storage layer. In addition, with the help of the metadata cache of the skip list structure, the efficiency of metadata operations such as the List directory was optimized. The test results show that, compared to the Linux client of HDFS, the performance of random read improves ten times more, the sequential read and sequential write improves about three to four times. The performance of random write can reach 20% of the local file system. Besides, the List operation efficiency of the directory improves about 10 times. However, due to the additional switching of kernel-mode and user-mode introduced by FUSE (Filesystem in Userspace), the Linux client of SMDFS3.0 has a performance penalty of about 50% compared to Java interface.

Implementation of deterministic routing fault-tolerant strategies for K-Ary N-Bridge system

XU Jiaqing, WAN Wen, CAI Dongjing, TANG Fuqiao, HE Jie, ZHANG Lei

2018, 38(5):  1393-1398.  DOI: 10.11772/j.issn.1001-9081.2017103024

Asbtract ( )   PDF (956KB) ( )  

References | Related Articles | Metrics

The leaf switch failure would seriously affect the use of high performance computer system with K-Ary N-Bridge topology. In order to improve the usability and maintainability of that topology, a routing fault-tolerant strategy based on misrouting algorithm was proposed. The basic idea was to bypass the failed leaf switch leveraging misrouting, jump to other leaf switches in the same dimension, and then reached the destination node through the normal route. The proposed fault-tolerant strategy could shield the failed leaf switch without affecting the system usage. A fault-tolerant experiment was carried out in a practical K-Ary N-Bridge topology. The result shows that this fault-tolerant strategy can quickly shield the failed leaf switch as expected and can effectively improve the efficiency of system maintenance.

Dynamic weighted scheduling strategy based on Docker swarm cluster

HUANG Kai, MENG Qingyong, XIE Yulai, FENG Dan, QIN Leihua

2018, 38(5):  1399-1403.  DOI: 10.11772/j.issn.1001-9081.2017102789

Asbtract ( )   PDF (830KB) ( )  

References | Related Articles | Metrics

As the built-in scheduling strategy of Docker swarm cannot implement load balance of cluster very well and the utilization rate of cluster resource is not very high, a dynamic weighted scheduling algorithm was proposed. The weight coefficient was set on the resource, and the parameter bias was introduced to dynamically adjust the resource weight for different services. According to the actual resource utilization of each node, the node weight was calculated to reflect node load, and was used for scheduling. Compared with the original Docker scheduling strategy and the weighted scheduling strategy without parameter adjustment, the proposed algorithm makes all the resource utilization of each node in the cluster more balanced. At the same time, the proposed algorithm can achieve faster service running speed under the condition of high cluster load.

Cache optimization for compressed databases in various storage environments

ZHANG Jiachen, LIU Xiaoguang, WANG Gang

2018, 38(5):  1404-1409.  DOI: 10.11772/j.issn.1001-9081.2017102861

Asbtract ( )   PDF (1124KB) ( )  

References | Related Articles | Metrics

In recent years, the amount of data in various industries grows rapidly, which results in the increasing of optimization demands in database storage system. Relational databases are I/O-intensive, take use of relatively free CPU time, data compression technology could save data storage space and I/O bandwidth. However, the compression features of current database systems were designed for traditional storage and computing environments, without considering the impact of virtualized environments or the use of Solid State Drive (SSD) on system performance. To optimize the cache performance of database compression system, a database compression system performance model was proposed, and the impact on the I/O performance of various system environments was analyzed. Take the open source database MySQL as an example, the corresponding cache optimization methods were given based on analysis. Evaluation results on Kernel-based Virtual Machine (KVM) and MySQL database show that the optimized version has an increase of more than 40% in performance under some configurations, even close to superior physical machine performance.

Adaptive image matching algorithm based on SIFT operator fused with maximum dissimilarity coefficient

CHEN Hong, XIAO Yue, XIAO Chenglong, SONG Hao

2018, 38(5):  1410-1414.  DOI: 10.11772/j.issn.1001-9081.2017102562

Asbtract ( )   PDF (809KB) ( )  

References | Related Articles | Metrics

As the traditional Scale Invariant Feature Transform (SIFT) image matching algorithm has high false matching rate and eliminating the condition of mismatching points is unitary, an adaptive image matching method based on SIFT operator fused with maximum dissimilarity coefficient was proposed. Firstly, On the basis of Euclidean distance measurement, the optimal maximum dissimilarity coefficients values of the 128-dimensional feature vectors in SIFT algorithm were obtained. Then, the matching points were selected according to the obtained optimal values. Random Sample Consensus (RANSAC) was used to calculate the correct rate of matching. Finally, the stereo matching images of Daniel Scharstein and Richard Szeliski were used to verify the algorithm. The experimental results show that the correct matching rate of the improved algorithm is about 10 percentage points higher than that of the traditional SIFT algorithm. The improved algorithm effectively reduces the mismatches and is more suitable for image matching applications with similar regions. In terms of runtime, the proposed method has an average time of 1.236 s, which can be applied to the systems with low real-time requirements.

Adaptive scale bilateral texture filtering method

WANG Hui, WANG Yue, LIU Changzu, ZHUANG Shanna, CAO Junjie

2018, 38(5):  1415-1419.  DOI: 10.11772/j.issn.1001-9081.2017102589

Asbtract ( )   PDF (901KB) ( )  

References | Related Articles | Metrics

Almost all of existing works on structure-preserving texture smoothing utilize the statistical features of pixels within local rectangular patches to distinguish structures from textures.However, the patch sizes of the rectangular regions are single-scale, which may lead to texture over-smoothed or non-smoothed for images with sharp structures or structures at different scales. Thus, an adaptive scale bilateral texture filtering method was proposed. Firstly, the patch size of rectangular region for each pixel was chosen adaptively from some given candidate sizes based on statistical analysis of local patches, where larger patch sizes were chosen for the homogeneous texture regions and smaller ones for regions near the structure edges. Secondly, guided image were computed via the adaptive patch sizes. Finally, the guided bilateral filtering was operated on the original image. The experimental results demonstrate that the proposed method can better preserve image structures and smooth textures.

Single image dehazing algorithm based on traffic scene region enhancement

LIANG Zhonghao, PENG Dewei, JIN Yanxu, GUO Liang

2018, 38(5):  1420-1426.  DOI: 10.11772/j.issn.1001-9081.2017112663

Asbtract ( )   PDF (1224KB) ( )  

References | Related Articles | Metrics

For the current dehazing algorithm easily results in low brightness of near road area and distant sky area with strong dehazing, and high brightness of middle and distant area with weak dehazing, based on a depth learning dehazing algorithm, a dehazing algorithm combined with image scene depth and road image characteristics of fog and sky roads was proposed. Firstly, based on the principle of dehazing algorithm of deep learning, a convolution neural network was constructed to calculate the scene transmittance. And then the image depth map was estimated based on the transmittance and atmospheric scattering model. Two parameters were constructed, the upper threshold and the lower threshold, to divide the depth map into middle, far, and near areas. Based on the enhancement function constructed by the different parts of the depth map, the enhancement amplitude of image processing was determined. Finally, based on the traditional atmospheric scattering model, the intensified illumination intensity was used to adjust the recovery intensity of different areas to obtain the optimized image. The experimental results show that the proposed algorithm is as good as other representative dehazing algorithms and enhance the middle and distant areas of the road image better. It effectively solves the color distortion and low contrast ratio of the near road surface and distant sky in the foggy road image, improves the visual effect of the reconstructed image, and has better image sharpening effect than dark channel prior algorithm, vision enhancement algorithm for homogeneous and heterogeneous fog, and typical dehazing algorithm based on deep learning.

Foggy image enhancement based on adaptive Riesz fractional differential

LEI Sijia, ZHAO Fengqun

2018, 38(5):  1427-1431.  DOI: 10.11772/j.issn.1001-9081.2017102480

Asbtract ( )   PDF (794KB) ( )  

References | Related Articles | Metrics

In order to improve clarity of foggy images and solve the problem of unicity of fractional order, a new adaptive fractional differential image enhancement method was proposed. Based on an approximate formula of Riesz fractional differential with six order accuracy, a new high precision fractional differential mask:RH operator (Riesz Higher order operator) was constructed, and the IRH operator (Improved Riesz Higher order operator) was proposed by improving RH operator. A fractional differential function was established based on image local features, and a selection criterion of fractional differential order was proposed, and then the adaptive selection method of order point by point was implemented. Combining with IRH operator, an adaptive fractional differential image enhancement algorithm was formed. For color images, due to low independence among components in RGB space, color distortion may occur after enhancement of each channel. Therefore, the image was converted from the RGB space to the HSV space and only the luminance channel was enhanced. A group of foggy images was selected compared with Tiansi operator, the segmentation-based adaptive fractional differential image enhancement algorithm and the adaptive fractional-differential compound bilateral filtering algorithm. The results show that the proposed method has obvious enhancement effect by calculating the information entropy and average gradient in comparison with methods in the reference, which further demonstrates the effectiveness of the proposed algorithm.

RGB-D saliency detection based on improved local background enclosure feature

YUAN Quan, ZHANG Jianfeng, WU Lizhi

2018, 38(5):  1432-1435.  DOI: 10.11772/j.issn.1001-9081.2017102587

Asbtract ( )   PDF (625KB) ( )  

References | Related Articles | Metrics

Focusing on the issue that the LBE (Local Background Enclosure) algorithm is over dependent on depth information and difficult to fully detect the object with complex structure, a RGB-D saliency detection algorithm based on the improved LBE features was proposed. Firstly, a set of segmentations was obtained by multi-level segmentation. Then, the depth saliency map was obtained by computing and merging the LBE features on each level segmentation map. Finally, a saliency map was obtained by adjusting the depth saliency map with color information and prior information. The experimental results show that compared with LBE algorithm, the precision of the proposed algorithm is slightly decreased and the recall is significantly improved, and the obtained saliency maps are much more close to the true values.

Efficient and fast dual-channel MAC protocol for terahertz wireless personal area networks

ZHOU Xun, ZHOU Haidong, REN Zhi, ZOU Mingrui, LI Guangbin

2018, 38(5):  1436-1441.  DOI: 10.11772/j.issn.1001-9081.2017102542

Asbtract ( )   PDF (981KB) ( )  

References | Related Articles | Metrics

To address the problems that data transmission delay is high and the channel utilization rate is low in the TAB-MAC (Assisted Beamforming MAC (Medium Access Control) protocol for terahertz communication network) for present terahertz Wireless Personal Area Network (T-WPAN), an Efficient and Fast dual-channel MAC protocol for T-WPAN (EF-MAC) was proposed. A test frame was sent to the source node through destination node to reduce an acknowledgment frame, so as to reduce control overhead and test delay. And then a sending and receiving mechanism of node location information was adaptively concelled, the source or destination node obtained the location information of the other node through interaction process previously of RTS/CTS (Request To Send/Clear To Send) frame, and the position of other node had not changed, thus the location information of RTS or CTS frame could be omitted to reduce control overhead. The theoretical analysis and simulation results show that compared with TAB-MAC protocol, the proposed protocol can effectively reduce data transmission delay and improve network throughput.

Simulation and implementation of physical random access channel signal detection in long term evolution advanced system

ZHANG Yajing, LIU Yulin, ZHANG Zhizhong

2018, 38(5):  1442-1446.  DOI: 10.11772/j.issn.1001-9081.2017102600

Asbtract ( )   PDF (805KB) ( )  

References | Related Articles | Metrics

According to the influence of Doppler frequency shift on the detection of Physical Random Access Channel (PRACH) signals, the signal detection algorithms were were divided into medium speed, high speed, ultra-high speed modes, and they were improved respectively. In the medium speed mode, a preamble detection algorithm based on frequency offset correction was proposed. In the high speed mode, an multi-sliding window peak detection algorithm was proposed. In the ultra-high speed mode, a frequency offset compensation preamble detection algorithm based on integer subcarrier was proposed. The simulation results show that, in different scenarios, when the PRACH signals are transmitted through the Additive White Gaussian Noise (AWGN) channel, and the false alarm rate performance of the receiver is improved by at least 3.8 dB, when the PRACH signals are transmitted through the Extend Typical Urban model (ETU) channel, the false alarm rate performance is improved by at least 1 dB. Compared with the frequency domain correlation detection algorithm, the proposed algorithms can improve the probability of successful detection of preamble signals and reduce the random access delay.

Pilot optimization and channel estimation in massive multiple-input multiple-output systems based on compressive sensing

JIN Feng, TANG Hong, ZHANG Jinyan, YIN Lixin

2018, 38(5):  1447-1452.  DOI: 10.11772/j.issn.1001-9081.2017112677

Asbtract ( )   PDF (922KB) ( )  

References | Related Articles | Metrics

Aiming at the problem that pilot overhead required by downlink channel estimation of FDD (Frequency-Division Duplexing) massive MIMO (Multiple-Input Multiple-Output) was unaffordable, a pseudo-random pilot optimization scheme based on Compressive Sensing (CS) techniques with non-orthogonal pilot at the base station and the objective to minimize the cross correlation of the measurement matrix was proposed firstly. Then, a crossover and mutation judgment mechanism and an inner loop and outer loop mechanism were introduced to ensure the optimization of pilot sequence. Secondly, a Channel State Information (CSI) estimation algorithm based on CS techniques by utilizing the spatially common sparsity and temporal correlation in wireless MIMO channels was presented. Matrix estimation is performed by using LMMSE (Linear Minimum Mean Square Error) algorithm to accurately obtain CSI. Analysis and simulation results show that compared with random search pilot optimization scheme, location-based optimization scheme, local common support algorithm, Adaptive Structured Subspace Pursuit (ASSP) algorithm, Orthogonal Matching Pursuit (OMP) algorithm and Stepwise Orthogonal Matching Pursuit (StOMP) algorithm, the proposed algorithm can significantly achieve good channel estimation performance in the case of low pilot overhead ratio and low Signal-to-Noise Ratio (SNR).

Cache placement optimization scheme in D2D networks with heterogeneous cache capacity

LONG Yanshan, WU Dan, CAI Yueming, WANG Meng, GUO Jibin

2018, 38(5):  1453-1457.  DOI: 10.11772/j.issn.1001-9081.2017112710

Asbtract ( )   PDF (885KB) ( )  

References | Related Articles | Metrics

Limited and heterogeneous cache capacity is one of the key parameters which affect the cache efficiency in Device-to-Device (D2D) caching networks. However, most of existing literatures assume all users have homogeneous cache capability. In this regard, cache placement optimization is necessary for practical scenarios with heterogeneous cache capacities. Considering the mobility and random distribution of terminal users, users with different cache capacities were modeled as mutually independent homogeneous Poisson point processes with stochastic geometry. Moreover, the average cache hit ratio was derived with considering both self-offloading and D2D-offloading cases. Finally, a Joint Cache Placement (JCP) algorithm based on coordinate gradient optimization was proposed to obtain the optimal cache placement scheme which can maximize the cache hit ratio. Simulation results show that the proposed JCP can achieve larger cache hit ratio than the existing cache placement schemes.

Two-dimensional space code index modulation algorithm

JIANG Zhilin, GE Lijia, XING Fengying, YANG Qin

2018, 38(5):  1458-1462.  DOI: 10.11772/j.issn.1001-9081.2017102612

Asbtract ( )   PDF (767KB) ( )  

References | Related Articles | Metrics

To deal with the problem that the number of antennas increases when Quadrature Spatial Modulation (QSM) improves the transmission rate, which requires a lot of resources and makes it difficult to achieve, a two-dimensional Space Code Orthogonal Index Modulation (SCOIM) was proposed. The transmitter information bits were mapped to the Pseudo Noise (PN) code index, the antenna index and the modulation symbol respectively. The in-phase part and orthogonal part of the modulation symbol spreaded spectrum through selecting activated PN code respectively, and were transmitted by activated antennas respectively. Analysis and simulation results show that in the comparison experiments with QSM, the index resource of SCOIM can be saved at least half at the same transmission rate and the saving increases exponentially with the increase of transmission rate. What's more, SCOIM has a performance advantage of about 5 dB when the bit error rate is 10^-4.

Video region detection algorithm for virtual desktop protocol

HOU Wenhui, WANG Junfeng

2018, 38(5):  1463-1469.  DOI: 10.11772/j.issn.1001-9081.2017102610

Asbtract ( )   PDF (1194KB) ( )  

References | Related Articles | Metrics

At present, there are some problems when video is played on virtual desktop protocol with partitioning mechanism, such as the video is not smooth and the bandwidth is highly occupied. In this paper, a Video Area Detection Algorithm (VRDA) was proposed based on virtual desktop protocol, called SPICE (Simple Protocol for Independent Computing Environment). Video regions were detected in the process of playing video on virtual desktop protocol, each of which was intercepted as a complete video frame, and decompressed by MPEG4 (Moving Pictures Experts Group-4) video compression algorithm instead of the original compression algorithm MJPEG (Motion JPEG) with lower efficiency. A evaluation metric named DAETD (Difference between Actual and Expected Display Time) was proposed to test the fluency of the improved SPICE, meanwhile, the bandwidth consumption of SPICE was also tested. The experimental results show that the proposed algorithm can improve the video fluency and reduce the network bandwidth consumption.

Adaptive unicast routing algorithm for vertically partially connected 3D NoC

SUN Meidong, LIU Qinrang, LIU Dongpei, YAN Binghao

2018, 38(5):  1470-1475.  DOI: 10.11772/j.issn.1001-9081.2017102411

Asbtract ( )   PDF (876KB) ( )  

References | Related Articles | Metrics

Traditional TSV (Through Silicon Via) table in vertically partially connected three-Dimensional Network-on-Chip (3D NoC) only stores TSV address information, which easily causes network congestion. In order to solve this problem, a record table architecture was proposed. The record table stored not only the nearest four TSV addresses to the router, but also the input-buffer occupancy and fault information of the corresponding router. Based on the record table, a novel adaptive unicast routing algorithm for the shortest transmission path was proposed. Firstly, the coordinates of current node and destination node were calculated to determine the transmission mode of packets. Secondly, by using the proposed algorithm, whether the transmission path was faulty and got information of buffer occupancy was obtained simultaneously. Finally, the optimal transmission port was determined and the packets were transmitted to the neighboring router. The experimental results under two network sizes show that the proposed algorithm has obvious advantages in average delay and throughput compared with Elevator-First algorithm. Additionally, the rates of losing packet under Random model and Shuffle traffic model are 25.5% and 29.5% respectively when the network fault rate is 50%.

Multi-channel scheduling strategy in smart distribution network

BAO Xingchuan, PENG Lin

2018, 38(5):  1476-1480.  DOI: 10.11772/j.issn.1001-9081.2017102444

Asbtract ( )   PDF (870KB) ( )  

References | Related Articles | Metrics

In order to effectively improve the Quality of Service (QoS) of wireless sensor network-based distribution network and further enhance the real-time and reduce the delay in a distribution network, a multi-channel scheduling strategy based on priority was proposed. First of all, a Link routing algorithm Based on Minimum Hop Spanning Tree (LB-MHST) was proposed to overcome the radio frequency interference and ensure the service quality of the smart grid according to the information of real-time channel state. Then, according to the different delay requirements of different data packets in the distribution network, the priority of data transmission was considered, which effectively improved the data transmission efficiency of the sensing node and further satisfied the QoS requirements in the distribution network. The experimental results show that the proposed algorithm can improve the real-time performance by 12 percentage points, 15.2 percentage points and 18 percentage points, compared with the Minimum Hop Spanning Tree (MHST) algorithm, under the cases with one channel, 8 channels and 16 channels.

Unscented Kalman filtering method with nonlinear equality constraint

TANG Qi, HE Lamei

2018, 38(5):  1481-1487.  DOI: 10.11772/j.issn.1001-9081.2017102472

Asbtract ( )   PDF (1004KB) ( )  

References | Related Articles | Metrics

A new constrained filtering method based on Unscented Kalman Filter (UKF) and pseudo-observation called SPUKF (Sub-system Parallel Unscented Kalman Filter) was proposed for state estimation of nonlinear system with nonlinear constraints. In the proposed method, the system and constraint equations were fictitiously divided and reconstructed into two sub-systems, and then the state estimation was obtained from two concurrent filtering processes which were established on these two sub-systems alternately. Compared to sequential processing method of pseudo-measurement, SPUKF did not need to determine the processing order between measurement and constraint, but achieved better performances, so as to address the problem of deciding processing order beforehand in sequential processing method. In the simulation of pendulum motion, it is verified that SPUKF gets better estimation performance and less running time than the two forms of sequential processing method, and enhances the estimation error improvement ratio by 22 percentage points than UKF. Furthermore, it obtains comparable estimation results with batch processing way.

Population model of giant panda ecosystem based on population dynamics P system

TIAN Hao, ZHANG Gexiang, RONG Haina, Mario J. PÉREZ-JIMÉNEZ, Luis VALENCIA-CABRERA, CHEN Peng, HOU Rong, QI Dunwu

2018, 38(5):  1488-1493.  DOI: 10.11772/j.issn.1001-9081.2017102551

Asbtract ( )   PDF (1014KB) ( )  

References | Related Articles | Metrics

Giant panda pedigree data is an important data base for studying the population dynamics of giant pandas. Therefore, it is of great significance for data modeling of giant panda ecosystems from the perspective of panda conservation. Focused on this issue, a data modeling method of giant panda ecosystem based on population dynamics P system was proposed. Based on the giant panda pedigree data released by Chinese Association of Zoological Gardens, the population characteristics of captive pandas were simulated and researched in China Giant Panda Conservation Research Center from individual behavior. The change rules of reproductive parameters were analyzed in detail, and added to the field released module. Eventually, a population dynamic P system for giant panda was designed releasing-to-the-wild with a two-layer nested membrane structure, a collection of objects and a series of evolution rules which is inline with the characteristics of giant panda. For all giant panda, the maximum relative error between the simulation results and the actual data was within ±4.13% and basically controlled within ±2.7% of P system. The experimental results verify the effectiveness and soundness of the proposed model. It can simulate the population dynamic change trend of giant panda and provide the basis for management decision-making.

Wolves optimization algorithm based on Cell-DEVS for forest fire-fighting resource scheduling

LI Bin, CHEN Aibin, ZHOU Guoxiong, ZHOU Tao

2018, 38(5):  1494-1499.  DOI: 10.11772/j.issn.1001-9081.2017102603

Asbtract ( )   PDF (1010KB) ( )  

References | Related Articles | Metrics

In view of the difficulty of forest fire-fighting dispatching force organization and low degree of refinement,a wolves optimization algorithm based on Cell-DEVS(Discrete Event System Specification) for forest fire-fighting resource scheduling was proposed. Firstly, Rothermel forest fire spread model was used to classify forest fire spread rate. Secondly, according to the principle of control the key in forest fire control, the forest fire-fighting resource scheduling model was built by a way of coupling drive modular designing,and a Wolves Strong Survival Update Mechanism (WSSUM) based on the Cell-DEVS model was proposed. Finally, aiming at the problem that the scheduling of forest fire-fighting resources was not fine enough in the local search of unit time step, an modified Wolves Optimization Algorithm (WOA) based on improved local search strategy was adopted to schedule the forest fire fighting resources in the local walk interactively. In the comparison experiments with WSSUM algorithm, the WOA improved the local search performance while reducing the task execution time. The experimental results show that the convergence speed is improved by 10.1% compared with that before improvement. The study adapts to the individual fire-fighting command system equipped with locating equipment to realize differentiated fine force dispatching.

Distribution analysis method of industrial waste gas for non-detection zone based on bi-directional error multi-layer neural network

WANG Liwei, WANG Xiaoyi, WANG Li, BAI Yuting, LU Yutian

2018, 38(5):  1500-1504.  DOI: 10.11772/j.issn.1001-9081.2017102606

Asbtract ( )   PDF (893KB) ( )  

References | Related Articles | Metrics

Industrial waste gas has accounted for about 70% of the atmospheric pollution sources. It is crucial to establish a full-scale and reasonable monitoring mechanism. However, the monitoring area is so large and monitoring devices can not be set up in some special areas. Besides, it is difficult to model the gas distribution according with the actual. To solve the practical and theoretical problems, an analysis method of industrial waste gas distribution for non-detection zone was proposed based on a Bi-directional Error Multi-Layer Neural Network (BEMNN). Firstly, the monitoring mechanism was introduced in the thought of "monitoring in boundary and inference of dead zone", which aimed to offset the lack of monitoring points in some areas. Secondly, a multi-layer combination neural network was proposed in which the errors propagate in a bi-directional mode. The network was used to model the gas distribution relationship between the boundary and the dead zone. Then the gas distribution in the dead zone could be predicted with the boundary monitoring data. Finally, an experiment was conducted based on the actual monitoring data of an industrial park. The mean absolute error was less than 28.83 μg and the root-mean-square error was less than 45.62 μg. The relative error was between 8% and 8.88%. The results prove the feasibility of the proposed method, which accuracy can meet the practical requirement.

Online portfolio selection based on autoregressive moving average reversion

YU Shunchang, HUANG Dingjiang

2018, 38(5):  1505-1511.  DOI: 10.11772/j.issn.1001-9081.2017102572

Asbtract ( )   PDF (996KB) ( )  

References | Related Articles | Metrics

Focused on the issue that noisy data, single period hypothesis and nonstationary prediction are not fully considered in the existing mean reversion strategy, an efficient OnLine Autoregressive moving average Reversion (OLAR) algorithm based on multi-period was proposed. Firstly, a stock price forecasting model was given by using the autoregressive moving average algorithm, and it was converted into an autoregressive model by a reasonable assumption. Then, an objective function was given by combining the loss function and a regular term, and a closed solution was obtained by using the second-order information of the loss function. The portfolio's closed-form update was obtained by using the online Passive Aggressive (PA) algorithm. Theoretical analysis and experimental results show that, compared with Robust Median Reversion (RMR), the accumulated profits of OLAR increase by 455.6%, 221.5%, 11.2% and 50.3% on NYSE (N), NYSE (N), Dow Jones Industrial Average (DJIA) and MSCI datasets respectively. Meanwhile, the results of statistical test show that the superior performance of OLAR is not caused by random factors. In addition, compared with algorithms such as RMR and Online Moving Average Reversion (OLMAR), OLAR achieves the highest annualized percentage yield, Sharpe ratio and Calmar ratio. Finally, the running time of OLAR is almost the same as that of RMR and OLMAR, therefore OLAR is suitable for large-scale real-time applications.

Selective ensemble algorithm for gene expression data based on diversity and accuracy of weighted harmonic average measure

GAO Huiyun, LU Huijuan, YAN Ke, YE Minchao

2018, 38(5):  1512-1516.  DOI: 10.11772/j.issn.1001-9081.2017102464

Asbtract ( )   PDF (708KB) ( )  

References | Related Articles | Metrics

The diversity between base classifiers and the accuracy of single base classifiers itself are two important factors that affect the generalization performance of ensemble system. Aiming at the problem that the diversity and accuracy are difficult to balance, a selective ensemble algorithm for gene expression data based on Diversity and Accuracy of Weighted Harmonic Average (D-A-WHA) was proposed. The Kernel Extreme Learning Machine (KELM) was used as the base classifier, and the diversity and accuracy of base classifiers were adjusted by D-A-WHA measure. Finally, a set of classifiers with high accuracy and high diversity with other base classifiers were selected to ensemble. The experimental results on UCI gene dataset show that compared with traditional Bagging, Adaboost and other ensemble algorithms, the classification accuracy and stability of the selective ensemble algorithm based on D-A-WHA measure are improved significantly,and it can be applied to the classification of cancer gene expression data effectively.

Active disturbance rejection control for mobile robot with skidding and slipping

LUO Rui, SHI Wuxi, LI Baoquan

2018, 38(5):  1517-1522.  DOI: 10.11772/j.issn.1001-9081.2017102505

Asbtract ( )   PDF (796KB) ( )  

References | Related Articles | Metrics

The trajectory tracking of wheeled mobile robots with skidding and slipping disturbance was studied. Firstly, based on the kinematics model of the robot, an auxiliary kinematic controller was designed to make the auxiliary speed of the robot asymptotically converge to desired speed. Then based on dynamics model, a first-order Linear Active Disturbance Rejection Control (LADRC) was proposed by using back stepping technique, an Extended State Observer (ESO) was used to estimate and compensate for the skidding and slipping disturbance during operation, so that the actual speed of the robot converged to auxiliary speed, which could make the trajectory error to asymptotically converge to zero. The effectiveness of the proposed approach to reject skidding and slipping disturbance of wheeled mobile robot was verified by simulation and experiment.

The shortest path planning for mobile robots using improved A^* algorithm

WANG Wei, PEI Dong, FENG Zhang

2018, 38(5):  1523-1526.  DOI: 10.11772/j.issn.1001-9081.2017102446

Asbtract ( )   PDF (623KB) ( )  

References | Related Articles | Metrics

Aiming at the poor real-time performance of mobile robot path planning in complex indoor environment, a further improvement on A^* algorithm was proposed by analyzing and comparing Dijkstra algorithm, traditional A^* algorithm and some improved A^* algorithms. Firstly, the estimated path cost of the current node and its parent node were weighted in exponentially decreasing way. In this way, when the current code was far away from the target, the improved algorithm could search towards to the target quickly instead of searching around the start node. While the current code was near to the target, the algorithm could search the target carefully to ensure that the target was reachable. Secondly, the generated path was smoothed by quintic polynomia to further shorten the path and facilitate robot control. The simulation results show that compared with the traditional A^* algorithm, the proposed algorithm can reduce the searching time by 93.8% and reduce the path length by 17.6% and get the path without quarter turning point, so that the robot could get to the destination along the planned path without a break. The proposed algorithm is verified in different scenarios, and the results show that the proposed algorithm can adapt to different environments and has good real-time performance.