Abstract
Given a set of moving objects as well as their friend relationships, a time-aware road network, and a time threshold per friend pair, the proximity detection problem in time-aware road networks is to find each pair of moving objects such that the time distance (defined as the shortest time needed for two moving objects to meet each other) between them is within the given threshold. The problem of proximity detection is often encountered in autonomous driving and traffic safety related applications, which require low-latency, real time proximity detection with relatively low communication cost. However, (i) most existing proximity detection solutions focus on the Euclidean space which cannot be used in road network space, (ii) the solutions for road networks focus on static road networks and do not consider time distance and thus cannot be applied in time-aware road networks, and (iii) there are no works aiming to simultaneously reduce the communication cost, the communication latency, and computational cost. Motivated by these, we first design a low-latency proximity detection architecture based on Mobile Edge Computing (MEC) with the purpose of achieving low communication latency, then propose a proximity detection method including a client-side algorithm and a server-side algorithm, aiming at reducing the communication cost, and subsequently propose server-side computational cost optimization techniques to reduce the computational cost. Experimental results show that our MEC enhanced proximity detection architecture, our proximity detection method, and the server-side computational cost optimization techniques can reduce the communication latency, the communication cost, and the computational cost effectively.
Highlights
In a road network, how to effectively detect whether the moving users are within proximity or not, is referred to as the problem of proximity detection in road networks
Given a large number of mobile users and time distance T, time-aware proximity detection problem in road networks is to find a solution which can continuously detect which pairs of users among all users are within proximity based on time distance, and can achieve the objectives of reducing communication cost, communication latency, and computational cost, so as to save the network bandwidth and improve the reliability and efficiency of proximity detection
We propose server-side computational cost optimization techniques, which can reduce the computational time at the server side to a large extent
Summary
How to effectively detect whether the moving users are within proximity or not, is referred to as the problem of proximity detection in road networks. Y. Liu et al.: MEC-Enhanced Proximity Detection in Time-Aware Road Networks absolutely distributed peer-to-peer (P2P) architecture, based on geographical distance (Euclidean distance or road network distance), facing the bottlenecks of long communication latency, long computational time, and large communication cost. Traditional C/S architecture can provide a global view on the proximity states of all moving users, but it faces the problems of long communication latency and large computational time. We adopt the Mobile Edge Computing (MEC) paradigm to design the proximity detection architecture, which can reduce the communication latency between the clients and the server greatly. We propose a time-aware mobile region based proximity detection method, namely, TMRBD, including a client-side algorithm and a server-side algorithm, to solve the time-aware proximity detection problem, which can reduce the communication cost effectively.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
