点击次数：

论文类型：会议论文

发表时间：2017-01-01

收录刊物：EI、CPCI-S

页面范围：1-9

摘要：Neutrality is an essential indicator to elucidate the quality of service of Internet Service Providers (ISPs). In regards of crowdsourcing method which relies on Traceroute, some neutrality inference mechanisms, especially System 4, are facing the problem that traffic information is precious and scarce. In this paper, we present a novel mechanism named GTR to patch the Traceroute using geographic information in neutrality inference crowdsourcing. We design a method to take advantage of widely distributed and accessible Web servers to implement practical measurement to demystify the regularities of routing in geolocation-level network mapping. Besides, we prove that GTR could improve the recognition rate of overlaps in System 4 up to 75% in practice.