Improving on-demand data access efficiency in MANETs with cooperative caching

  • Authors:

  • Yu Du;Sandeep K. S. Gupta;Georgios Varsamopoulos

  • Affiliations:

  • IMPACT Laboratory, School of Computing and Informatics, Arizona State University, Tempe, AZ 85287-5406, United States;IMPACT Laboratory, School of Computing and Informatics, Arizona State University, Tempe, AZ 85287-5406, United States;IMPACT Laboratory, School of Computing and Informatics, Arizona State University, Tempe, AZ 85287-5406, United States

  • Venue:
  • Ad Hoc Networks
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Mobile ad hoc networks (MANETs) provide an attractive solution for networking in the situations where network infrastructure or service subscription is not available. Its usage can further be extended by enabling communications with external networks such as the Internet or cellular networks through gateways. However, data access applications in MANETs suffer from dynamic network connections and restricted resources. While most of the research focuses on media (or medium) access control (MAC) and routing layer solutions, we explore the possibility of making use of data locality and the commonality in users' interests at the application level. In this paper, we investigate how cooperative caching can be used to improve data access efficiency in MANETs. We propose COOP, a novel cooperative caching scheme for on-demand data access applications in MANETs. The objective is to improve data availability and access efficiency by collaborating local resources of mobile nodes. COOP addresses two basic problems of cooperative caching: cache resolution and cache management. To improve data availability and access efficiency, COOP discovers data sources which induce less communication overhead by utilizing cooperation zones, historical profiles, and hop-by-hop resolution. For cache management, COOP increases the effective capacity of cooperative caches by minimizing caching duplications within the cooperation zone and accommodating more data varieties. The performance of COOP is studied using mathematical analysis and simulations from the perspectives of data availability, time efficiency, and energy efficiency. The analysis and simulation results show that COOP significantly reduces response delay and improves data availability with proper settings of the cooperation zone radius.