Mobile Device-to-Device (D2D) content delivery networking: A design and optimization framework

Hye Joong Kang, Chung Gu Kang

    Research output: Contribution to journalArticlepeer-review

    52 Citations (Scopus)

    Abstract

    We consider a mobile content delivery network (mCDN) in which special mobile devices designated as caching servers (caching-server device: CSD) can provide mobile stations with popular contents on demand via device-to-device (D2D) communication links. On the assumption that mobile CSD's are randomly distributed by a Poisson point process (PPP), an optimization problem is formulated to determine the probability of storing the individual content in each server in a manner that minimizes the average caching failure rate. Further, we present a low-complexity search algorithm, optimum dual-solution searching algorithm (ODSA), for solving this optimization problem. We demonstrate that the proposed ODSA takes fewer iterations, on the order of O(log N) searches, for caching N contents in the system to find the optimal solution, as compared to the number of iterations in the conventional subgradient method, with an acceptable accuracy in practice. Furthermore, we identify the important characteristics of the optimal caching policies in the mobile environment that would serve as a useful aid in designing the mCDN.

    Original languageEnglish
    Pages (from-to)568-577
    Number of pages10
    JournalJournal of Communications and Networks
    Volume16
    Issue number5
    DOIs
    Publication statusPublished - 2014 Oct 1

    Bibliographical note

    Publisher Copyright:
    © 2014 KICS.

    Keywords

    • Caching probability
    • Caching server device
    • Device-to-device (D2D) communication
    • Mobile contents distributed network (mCDN)
    • Poisson point process (PPP)

    ASJC Scopus subject areas

    • Information Systems
    • Computer Networks and Communications

    Fingerprint

    Dive into the research topics of 'Mobile Device-to-Device (D2D) content delivery networking: A design and optimization framework'. Together they form a unique fingerprint.

    Cite this