CAJ | 학술논문

由于车联网是物联网在现代城市交通网络中的具体应用,车联网的数据融合已成为物联网信息感知领域中一个重要的研究课题.针对传统DS证据理论在证据合成过程中存在冲突证据分配不合理、融合方法收敛效果差以及多BBM的证据推理等问题,提出了一种基于等距映射的证据推理方法（isometric mapping evidential reasoning,IMER）.IMER,方法根据多BBM证据体间相似关系求取每一个证据体的低维嵌入向量,并计算出低维证据体相似度,以实现对车联网中多BBM证据体的证据推理.实验结果表明本文方法可以合理地分配冲突证据,同时具有较好的收敛性和有效性.
유우차련망시물련망재현대성시교통망락중적구체응용,차련망적수거융합이성위물련망신식감지영역중일개중요적연구과제.침대전통DS증거이론재증거합성과정중존재충돌증거분배불합리、융합방법수렴효과차이급다BBM적증거추리등문제,제출료일충기우등거영사적증거추리방법（isometric mapping evidential reasoning,IMER）.IMER,방법근거다BBM증거체간상사관계구취매일개증거체적저유감입향량,병계산출저유증거체상사도,이실현대차련망중다BBM증거체적증거추리.실험결과표명본문방법가이합리지분배충돌증거,동시구유교호적수렴성화유효성.
As vehicular Ad-hoc NETworks（VANET） is regarded as practical application of internet of things（IOT） in modern urban traffic network,data fusion of VANET has become an important research subject in the field of IOT information sensing.Aiming at the problems of irrational assignment of conflict evidence,bad convergence effect of fusion method and multi-BBM evidence reasoning in evidence combination of the traditional DS evidence theory,an isometric mapping evidential reasoning（IMER）method is proposed.In IMER method,the low-dimensional embedding vector of each evidence body is obtained by using the similar relations between multi-BBM evidence bodies,and then the similarity of lowdimensional evidence bodies is calculated to achieve evidential reasoning of multi-BBM evidence bodies in VANET.The experimental results show that the proposed method can reasonably assign conflict evidence and has good convergence and effectiveness.