Reducing travel time in VANETs using MACO with CUDA on GPU

Authors

  • Vinita Jindal
  • Punam Bedi

Keywords:

Parallelization; MACO; Ant Colony Optimizati

Abstract

Over the past few years, with the advent of Compute Unified Device Architecture (CUDA), Graphics Processing Units (GPUs) have evolved into highly parallel programmable architecture and provide the solution to many real-world applications. In this paper, we deal with a GPU implementation of Modified Ant Colony Optimization (MACO) algorithm that reduces the overall travel time of the journey by avoiding congestion enroute. The GPU implementation provides the faster computation in order to further reduce the travel time for the vehicles on move. We are providing the parallel implementation for all the phases of MACO approach. The implementation is done using parallel architecture on the GPU at NVIDIA GeForce 710M using C++ language running with CUDA toolkit 7.5 on Microsoft Visual Studio 2013. The accuracy of the proposed parallel implementation of MACO algorithm is validated both graphically as well as statistically using hypothesis testing by comparing it with both, the parallel implementation of the standard Dijkstra algorithm and that of the existing MACO algorithm on a real world North-West Delhi map with an increased number of vehicles. The obtained results for the parallel implementation of MACO illustrate the significance in the reduction of travel time with 99% confidence.

Downloads

Download data is not yet available.

Downloads

Published

2016-07-01

How to Cite

Vinita Jindal, & Punam Bedi. (2016). Reducing travel time in VANETs using MACO with CUDA on GPU. Journal of Network and Innovative Computing, 4, 9. Retrieved from https://cspub-jnic.org/index.php/jnic/article/view/121

Issue

Section

Original Article