MELOC-X: Extended Memory and Location Optimized Caching for Large Mobile Ad Hoc Networks of UAVs
Effective caching in mobile ad hoc network increases data availability. However, caching at strategic locations with reduced (controlled) number of copies is needed for many military applications involving UAVs to address security concerns, less maintenance overhead and maintaining availability. In general, existing cooperative caching approaches are deficient in finding the reduced number of strategic cache locations. One such technique to reduce the number of strategic cache locations without affecting the efficacy of data access for a small network topology of UAVs is called “memory and location optimized caching scheme (MELOC)”. However, having a single broker and metadata broadcast across the whole network in MELOC lead to severe performance hindrance in case of a large network topology of UAVs. Moreover, frequent cache replacements due to a change in network topology do not favor cache hit and bandwidth conservation in case of large mobile networks consisting of UAVs. In this paper, we design and evaluate an extended version of “MELOC called MELOC-X”, which suits large network topologies of UAVs by overcoming the above challenges. Our comparison with one such recent scheme with similar objectives showcased a significant improvement in performance. We also evaluate the impact of this scheme with respect to different metrics including the average number of hops, the average roundtrip time (i.e., average query latency), cache hits and mobility to access cached data through extensive simulations.
L. Chidambaram et al., "MELOC-X: Extended Memory and Location Optimized Caching for Large Mobile Ad Hoc Networks of UAVs," Distributed and Parallel Database Journal, vol. 34, no. 2, pp. 217-258, Springer, Jan 2016.
The definitive version is available at https://doi.org/10.1007/s10619-015-7181-2
Intelligent Systems Center
Keywords and Phrases
Caching; Cache location; Mobile ad hoc networks; Availability; UAV network
International Standard Serial Number (ISSN)
Article - Journal
© 2016 Springer, All rights reserved.