Collaborative Hierarchical Caching Over 5G Edge Computing Mobile Wireless Networks Conference Paper uri icon

abstract

  • © 2018 IEEE. Edge computing techniques have been developed to support the exponentially increasing service demands in the fifth generation (5G) networks by bringing the data contents and their corresponding computations/communications to the edge of the wireless networks, which is the area near to mobile users. As one of the promising and enabling techniques in edge computing wireless networks, in- network caching stores the data contents close to mobile users to efficiently reduce the transmission delay for time-sensitive multimedia data contents. However, one of the main challenges for implementing in-network caching techniques lies in how to develop the collaborative caching mechanisms among all caches in the edge of wireless networks to upper-bound the data transmission delay while maximizing the cache hitting rate. In this paper, we propose the inter-tier and intra-tier collaborative hierarchical caching mechanisms over 5G edge computing multimedia mobile wireless networks, where the popular multimedia data contents are selectively cached at different wireless network caching tiers (e.g., at routers, cellular- base stations/WiFi-access-point, and mobile devices, respectively). The inter-tier collaborative hierarchical caching minimizes the average number of hops (including wireless hops and wireline hops) through the collaborative caching across three wireless network caching tiers, and the intra-tier collaborative hierarchical caching maximizes the overall cache hitting rate within the same wireless network caching tier. To optimize the intra-tier collaborative hierarchical caching mechanism, we derive the upper-bound and lower- bound of maximum numbers of device-to-device (D2D) pairs at the bottom tier of wireless network caching. Finally, we use numerical analyses to evaluate and validate our proposed collaborative hierarchical caching mechanisms over edge computing mobile wireless networks.

author list (cited authors)

  • Zhang, X. i., & Zhu, Q.

citation count

  • 11

publication date

  • May 2018

publisher