協調運転支援のためのMECにおける超低遅延イベント処理方式の評価

Abstract

Intelligent Transportation Systems (ITS) is the application of advanced information and communications technology to surface transportation in order to achieve enhanced safety and mobility. We posit that event processing with ultra low latency (less than 1msec) enabled in the fifth generation (5G) mobile network enhance ITS especially in terms of safety driving. One of the requirements for safety driving support systems is communication delay less than 100msec, but it is difficult to achieve this in the current mobile communication infrastructure, e.g., Long Term Evolution (LTE). Therefore, Mobile Edge Computing (MEC) is proposed to to reduce communication delay. However, we point out following two challenges in applying MEC to ITS. First, the information that can be collected within short time in edge servers may be geographically limited. Second, the support for mobility in sensing information and actuating on it is challenging. To solve these problems, we propose a system with a hierarchical structure utilizing not only cloud and edge servers but also extreme edge nodes near UEs to deal with geographical limitation in information sensing and with publish-subscribe architecture for event processing. With a prototype system and an example application of dangerous driver detection, we evaluate communication delay and transaction delay in our proposed system and show that it is challenging to achieve communication delay of less than 100 msec in LTE. As a result, we posit that it is desirable to reduce the latency between UEs and MEC network nodes, especially radio part, in the order of 1msec for enabling mission critical applications in 5G mobile network such as supporting safety driving in ITS.