A new real-time map matching algorithm based on fuzzy logic is proposed. 3 main factors affecting the reliability of map matching, including the distance between the vehicle location and the matching road segment, the...A new real-time map matching algorithm based on fuzzy logic is proposed. 3 main factors affecting the reliability of map matching, including the distance between the vehicle location and the matching road segment, the angle between the vehicle direction and the road segment direction and the road connectivity are discussed. Fuzzy rules for the distance, angle and connectivity are presented to calculate the matching reliability. 2 indicators for estimating the matching reliability are then derived, one is the lower limit of the reliability, and the other is the limit error of the difference between the maximal value and the second-maximal value of the reliability. A real-time map-matching system based on fuzzy logic is therefore developed. Using the real data of global positioning system(GIS) based navigation and geographic information system(GPS) based road map, the method is verified and the (results) prove the effectiveness of the proposed method.展开更多
To improve the accuracy of short text matching,a short text matching method with knowledge and structure enhancement for BERT(KS-BERT)was proposed in this study.This method first introduced external knowledge to the i...To improve the accuracy of short text matching,a short text matching method with knowledge and structure enhancement for BERT(KS-BERT)was proposed in this study.This method first introduced external knowledge to the input text,and then sent the expanded text to both the context encoder BERT and the structure encoder GAT to capture the contextual relationship features and structural features of the input text.Finally,the match was determined based on the fusion result of the two features.Experiment results based on the public datasets BQ_corpus and LCQMC showed that KS-BERT outperforms advanced models such as ERNIE 2.0.This Study showed that knowledge enhancement and structure enhancement are two effective ways to improve BERT in short text matching.In BQ_corpus,ACC was improved by 0.2%and 0.3%,respectively,while in LCQMC,ACC was improved by 0.4%and 0.9%,respectively.展开更多
Individual identification of dairy cows is the prerequisite for automatic analysis and intelligent perception of dairy cows'behavior.At present,individual identification of dairy cows based on deep convolutional n...Individual identification of dairy cows is the prerequisite for automatic analysis and intelligent perception of dairy cows'behavior.At present,individual identification of dairy cows based on deep convolutional neural network had the disadvantages in prolonged training at the additions of new cows samples.Therefore,a cow individual identification framework was proposed based on deep feature extraction and matching,and the individual identification of dairy cows based on this framework could avoid repeated training.Firstly,the trained convolutional neural network model was used as the feature extractor;secondly,the feature extraction was used to extract features and stored the features into the template feature library to complete the enrollment;finally,the identifies of dairy cows were identified.Based on this framework,when new cows joined the herd,enrollment could be completed quickly.In order to evaluate the application performance of this method in closed-set and open-set individual identification of dairy cows,back images of 524 cows were collected,among which the back images of 150 cows were selected as the training data to train feature extractor.The data of the remaining 374 cows were used to generate the template data set and the data to be identified.The experiment results showed that in the closed-set individual identification of dairy cows,the highest identification accuracy of top-1 was 99.73%,the highest identification accuracy from top-2 to top-5 was 100%,and the identification time of a single cow was 0.601 s,this method was verified to be effective.In the open-set individual identification of dairy cows,the recall was 90.38%,and the accuracy was 89.46%.When false accept rate(FAR)=0.05,true accept rate(TAR)=84.07%,this method was verified that the application had certain research value in open-set individual identification of dairy cows,which provided a certain idea for the application of individual identification in the field of intelligent animal husbandry.展开更多
目前电磁时间反演(electromagnetic time reversal,EMTR)多应用在单一线路故障定位,且现有判据在高阻抗接地情况下效果不理想。针对上述问题,基于EMTR故障定位原理和均匀传输线理论推导了传播过程中线路故障信号与测量信号的传递函数,...目前电磁时间反演(electromagnetic time reversal,EMTR)多应用在单一线路故障定位,且现有判据在高阻抗接地情况下效果不理想。针对上述问题,基于EMTR故障定位原理和均匀传输线理论推导了传播过程中线路故障信号与测量信号的传递函数,根据传递函数的相关性提出了P范数判据。利用ATP-EMTP搭建10 kV配电网线路,对比了2范数与P范数判据在复杂配电网中的定位性能,并验证了所提判据在混合配电网线路的适用性。最后,分析了配电网发生低阻抗及高阻抗接地故障下P范数判据的鲁棒性。仿真结果表明,该方法在过渡电阻高达3 kΩ的情况下能准确定位,且定位精度高,受噪声、故障类型和采样频率的影响小。展开更多
A robust digital receiver based on a matched filter （MF） is proposed for the radio frequency identification （RFID） reader system to enhance the reliability of signal processing in the electronic product code （EPC...A robust digital receiver based on a matched filter （MF） is proposed for the radio frequency identification （RFID） reader system to enhance the reliability of signal processing in the electronic product code （EPC） sensor network （ESN）. The performance of the proposed receiver is investigated by examining the anti-collision algorithm in the EPC global Class1 Generation2 protocol. The validity and usefulness are demonstrated by both computer simulations and experiments. Based on the verification results, comparing with the conventional zero crossing detector （ZCD） based receiver, the proposed receiver is very robust against strong amplitude distortions and considerable frequency deviations happening on the backscattered signal from a passive tag.展开更多
For the problem of large network load generated by the Gnutella resource-searching model in Peer to Peer （P2P） network, a improved model to decrease the network expense is proposed, which establishes a duster in P2P...For the problem of large network load generated by the Gnutella resource-searching model in Peer to Peer （P2P） network, a improved model to decrease the network expense is proposed, which establishes a duster in P2P network, auto-organizes logical layers, and applies a hybrid mechanism of directional searching and flooding. The performance analysis and simulation results show that the proposed hierarchical searching model has availably reduced the generated message load and that its searching-response time performance is as fairly good as that of the Gnutella model.展开更多
Networked customization had gradually become one of t he import manufacturing and marketing patterns in the information age. Product d escription had an important effect on realizing networked customization. From th i...Networked customization had gradually become one of t he import manufacturing and marketing patterns in the information age. Product d escription had an important effect on realizing networked customization. From th is point of view, this paper firstly analyzed four requirements of adaptability, stability, conciseness and definitude that product description methods for netw orked customization must satisfy. According to these requirements, the paper com mented on current methods of product description, pointing out their advantages and disadvantages. As none of current methods could satisfy the four requirement s simultaneously, they were not fully applicable to networked customization. The n a plug-in based product description method was brought forward and discussed in detail. In this method, product description for networked customization w as abstracted as three component parts of basic description, basic feature and c ustomization feature. These parts had different characteristics. Basic descripti on was applicable to all products types. Basic feature was stable as concerning specific product type. Variants of certain product type all possessed identi cal basic features. Customization features were variable. Different combinations of customization features would give birth to different product variants. Apply ing object-oriented method, three standard product description plug-ins could be got by encapsulating these component parts. As "basic feature" and "customiza tion feature" plug-in had similar structures, they could be further abstracted as "feature" plug-in. When describing a type of product, a "basic description" plug-in would be instantiated firstly. Then according to product’s specific ch aracteristics, necessary number of "basic feature" and "customization feature" p lug-ins would be instantiated and appended, such as "color", "material", etc. E ach "basic feature" plug-in related to an instantiated "feature value" plug-in , while each "customization feature" plug-in related to one or multiple instant iated "feature value" plug-ins. Finally, "associate" and "exclude" relation ship should be specified among associated "customization feature" plug-ins. Thr ough appending and combining various plug-ins, this method was capable of descr ibing any type of product while meeting the four requirements simultaneously. Th is paper also gave an example to clarify the application of plug-in based produ ct description method. This example firstly described a kind of customizable TV provided by certain company. Then it showed how to describe an individual co nfiguration of this customizable TV in XML. Furthermore, suppose a consumer want ed to search for relevant information of a TV with game function, this example explained how to utilize XML query language to benefit consumer’s searching for product information. From analysis, we could draw the conclusion that plug -in based product description method was superior to current description method s and fully adapted to networked customization environment.展开更多
Networked manufacturing is a primary management mod e of virtual enterprise. It takes advantage of advanced network technology to pass information and devise and assemble necessary components so as to reach a certa in...Networked manufacturing is a primary management mod e of virtual enterprise. It takes advantage of advanced network technology to pass information and devise and assemble necessary components so as to reach a certa in specific purpose. The characteristics of it are dynamic, international and pr ompt. In contrast with traditional manufacturing, networked manufacturing utilizes external predominant resources, and breaks through tangible limits of enterpris e, so it can optimize the combination of social resources and improve the enterp rise′capability of adapting to the constantly changing market. Networked manufacturing produces some problems while bringing enterprise infin ite commercial chance. Compared with traditional manufacturing, stealing intelle ctual achievements and encroaching the related property rights not only becomes easier and quicker, but also leaves no traces when enterprises transfer informat ion through network. It′s well known that competition among corporations in the future is the on e of intellectual property in the final analysis. Therefore, recently, the statu s of intellectual property protection is promoted grandly. In the following parts, several proposals on protecting intellectual property ar e put forward in detail. We can protect the intellectual property by the ways of strengthening the consci ousness of protection. We can protect the intellectual property by perfecting the law. We will know which behaviors on the network are legitimate and which one are illegitimate. As a result, it can decrease and resolve the conflicts partially. Besides, the article puts forward that the conflicts can be resolved through a m edi-institution, an authoritative governmental organization. Moreover, we should strengthen the education on ethics. At last, the paper presents that we can protect our intellectual property throug h international cooperation.展开更多
A model is built to analyze the performance of service location based on greedy search in P2P networks. Hops and relative QoS index of the node found in a service location process are used to evaluate the performance ...A model is built to analyze the performance of service location based on greedy search in P2P networks. Hops and relative QoS index of the node found in a service location process are used to evaluate the performance as well as the probability of locating the top 5% nodes with highest QoS level. Both model and simulation results show that, the performance of greedy search based service location improves significantly with the increase of the average degree of the network. It is found that, if changes of both overlay topology and QoS level of nodes can be ignored during a location process, greedy-search based service location has high probability of finding the nodes with relatively high QoS in small number of hops in a big overlay network. Model extension under arbitrary network degree distribution is also studied.展开更多
基金Projects(40301043 and 40171078) supported by the National Natural Science Foundation of China
文摘A new real-time map matching algorithm based on fuzzy logic is proposed. 3 main factors affecting the reliability of map matching, including the distance between the vehicle location and the matching road segment, the angle between the vehicle direction and the road segment direction and the road connectivity are discussed. Fuzzy rules for the distance, angle and connectivity are presented to calculate the matching reliability. 2 indicators for estimating the matching reliability are then derived, one is the lower limit of the reliability, and the other is the limit error of the difference between the maximal value and the second-maximal value of the reliability. A real-time map-matching system based on fuzzy logic is therefore developed. Using the real data of global positioning system(GIS) based navigation and geographic information system(GPS) based road map, the method is verified and the (results) prove the effectiveness of the proposed method.
文摘To improve the accuracy of short text matching,a short text matching method with knowledge and structure enhancement for BERT(KS-BERT)was proposed in this study.This method first introduced external knowledge to the input text,and then sent the expanded text to both the context encoder BERT and the structure encoder GAT to capture the contextual relationship features and structural features of the input text.Finally,the match was determined based on the fusion result of the two features.Experiment results based on the public datasets BQ_corpus and LCQMC showed that KS-BERT outperforms advanced models such as ERNIE 2.0.This Study showed that knowledge enhancement and structure enhancement are two effective ways to improve BERT in short text matching.In BQ_corpus,ACC was improved by 0.2%and 0.3%,respectively,while in LCQMC,ACC was improved by 0.4%and 0.9%,respectively.
基金Supported by the National Key Research and Development Program of China(2019YFE0125600)China Agriculture Research System(CARS-36)。
文摘Individual identification of dairy cows is the prerequisite for automatic analysis and intelligent perception of dairy cows'behavior.At present,individual identification of dairy cows based on deep convolutional neural network had the disadvantages in prolonged training at the additions of new cows samples.Therefore,a cow individual identification framework was proposed based on deep feature extraction and matching,and the individual identification of dairy cows based on this framework could avoid repeated training.Firstly,the trained convolutional neural network model was used as the feature extractor;secondly,the feature extraction was used to extract features and stored the features into the template feature library to complete the enrollment;finally,the identifies of dairy cows were identified.Based on this framework,when new cows joined the herd,enrollment could be completed quickly.In order to evaluate the application performance of this method in closed-set and open-set individual identification of dairy cows,back images of 524 cows were collected,among which the back images of 150 cows were selected as the training data to train feature extractor.The data of the remaining 374 cows were used to generate the template data set and the data to be identified.The experiment results showed that in the closed-set individual identification of dairy cows,the highest identification accuracy of top-1 was 99.73%,the highest identification accuracy from top-2 to top-5 was 100%,and the identification time of a single cow was 0.601 s,this method was verified to be effective.In the open-set individual identification of dairy cows,the recall was 90.38%,and the accuracy was 89.46%.When false accept rate(FAR)=0.05,true accept rate(TAR)=84.07%,this method was verified that the application had certain research value in open-set individual identification of dairy cows,which provided a certain idea for the application of individual identification in the field of intelligent animal husbandry.
文摘目前电磁时间反演(electromagnetic time reversal,EMTR)多应用在单一线路故障定位,且现有判据在高阻抗接地情况下效果不理想。针对上述问题,基于EMTR故障定位原理和均匀传输线理论推导了传播过程中线路故障信号与测量信号的传递函数,根据传递函数的相关性提出了P范数判据。利用ATP-EMTP搭建10 kV配电网线路,对比了2范数与P范数判据在复杂配电网中的定位性能,并验证了所提判据在混合配电网线路的适用性。最后,分析了配电网发生低阻抗及高阻抗接地故障下P范数判据的鲁棒性。仿真结果表明,该方法在过渡电阻高达3 kΩ的情况下能准确定位,且定位精度高,受噪声、故障类型和采样频率的影响小。
基金supported by the Korea Evaluation Institute of Industrial Technology(KEIT),under the R&D Support Program of Ministry of Knowledge Economy,Korea
文摘A robust digital receiver based on a matched filter （MF） is proposed for the radio frequency identification （RFID） reader system to enhance the reliability of signal processing in the electronic product code （EPC） sensor network （ESN）. The performance of the proposed receiver is investigated by examining the anti-collision algorithm in the EPC global Class1 Generation2 protocol. The validity and usefulness are demonstrated by both computer simulations and experiments. Based on the verification results, comparing with the conventional zero crossing detector （ZCD） based receiver, the proposed receiver is very robust against strong amplitude distortions and considerable frequency deviations happening on the backscattered signal from a passive tag.
文摘For the problem of large network load generated by the Gnutella resource-searching model in Peer to Peer （P2P） network, a improved model to decrease the network expense is proposed, which establishes a duster in P2P network, auto-organizes logical layers, and applies a hybrid mechanism of directional searching and flooding. The performance analysis and simulation results show that the proposed hierarchical searching model has availably reduced the generated message load and that its searching-response time performance is as fairly good as that of the Gnutella model.
文摘Networked customization had gradually become one of t he import manufacturing and marketing patterns in the information age. Product d escription had an important effect on realizing networked customization. From th is point of view, this paper firstly analyzed four requirements of adaptability, stability, conciseness and definitude that product description methods for netw orked customization must satisfy. According to these requirements, the paper com mented on current methods of product description, pointing out their advantages and disadvantages. As none of current methods could satisfy the four requirement s simultaneously, they were not fully applicable to networked customization. The n a plug-in based product description method was brought forward and discussed in detail. In this method, product description for networked customization w as abstracted as three component parts of basic description, basic feature and c ustomization feature. These parts had different characteristics. Basic descripti on was applicable to all products types. Basic feature was stable as concerning specific product type. Variants of certain product type all possessed identi cal basic features. Customization features were variable. Different combinations of customization features would give birth to different product variants. Apply ing object-oriented method, three standard product description plug-ins could be got by encapsulating these component parts. As "basic feature" and "customiza tion feature" plug-in had similar structures, they could be further abstracted as "feature" plug-in. When describing a type of product, a "basic description" plug-in would be instantiated firstly. Then according to product’s specific ch aracteristics, necessary number of "basic feature" and "customization feature" p lug-ins would be instantiated and appended, such as "color", "material", etc. E ach "basic feature" plug-in related to an instantiated "feature value" plug-in , while each "customization feature" plug-in related to one or multiple instant iated "feature value" plug-ins. Finally, "associate" and "exclude" relation ship should be specified among associated "customization feature" plug-ins. Thr ough appending and combining various plug-ins, this method was capable of descr ibing any type of product while meeting the four requirements simultaneously. Th is paper also gave an example to clarify the application of plug-in based produ ct description method. This example firstly described a kind of customizable TV provided by certain company. Then it showed how to describe an individual co nfiguration of this customizable TV in XML. Furthermore, suppose a consumer want ed to search for relevant information of a TV with game function, this example explained how to utilize XML query language to benefit consumer’s searching for product information. From analysis, we could draw the conclusion that plug -in based product description method was superior to current description method s and fully adapted to networked customization environment.
文摘Networked manufacturing is a primary management mod e of virtual enterprise. It takes advantage of advanced network technology to pass information and devise and assemble necessary components so as to reach a certa in specific purpose. The characteristics of it are dynamic, international and pr ompt. In contrast with traditional manufacturing, networked manufacturing utilizes external predominant resources, and breaks through tangible limits of enterpris e, so it can optimize the combination of social resources and improve the enterp rise′capability of adapting to the constantly changing market. Networked manufacturing produces some problems while bringing enterprise infin ite commercial chance. Compared with traditional manufacturing, stealing intelle ctual achievements and encroaching the related property rights not only becomes easier and quicker, but also leaves no traces when enterprises transfer informat ion through network. It′s well known that competition among corporations in the future is the on e of intellectual property in the final analysis. Therefore, recently, the statu s of intellectual property protection is promoted grandly. In the following parts, several proposals on protecting intellectual property ar e put forward in detail. We can protect the intellectual property by the ways of strengthening the consci ousness of protection. We can protect the intellectual property by perfecting the law. We will know which behaviors on the network are legitimate and which one are illegitimate. As a result, it can decrease and resolve the conflicts partially. Besides, the article puts forward that the conflicts can be resolved through a m edi-institution, an authoritative governmental organization. Moreover, we should strengthen the education on ethics. At last, the paper presents that we can protect our intellectual property throug h international cooperation.
文摘A model is built to analyze the performance of service location based on greedy search in P2P networks. Hops and relative QoS index of the node found in a service location process are used to evaluate the performance as well as the probability of locating the top 5% nodes with highest QoS level. Both model and simulation results show that, the performance of greedy search based service location improves significantly with the increase of the average degree of the network. It is found that, if changes of both overlay topology and QoS level of nodes can be ignored during a location process, greedy-search based service location has high probability of finding the nodes with relatively high QoS in small number of hops in a big overlay network. Model extension under arbitrary network degree distribution is also studied.
