Joint Content Caching, Recommendation, and Transmission Optimization for Next Generation Multiple Access Networks

Yaru Fu, Yue Zhang, Qi Zhu, Mingzhe Chen, Tony Q.S. Quek

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)


In this paper, we exploit a behavior-shaping proactive mechanism, namely, recommendation, in cache-assisted non-orthogonal multiple access (NOMA) networks, aiming at minimizing the average system's latency. Thereof, the considered latency consists of two parts, i.e., the backhaul link transmission delay and the content delivery latency. Towards this end, we first examine the expression of system latency, demonstrating how it is critically determined by content cache placement, personalized recommendation, and delivery associated NOMA user pairing and power control strategies. Thereafter, we formulate the minimization problem mathematically taking into account the cache capacity budget, the recommendation-oriented requirements, and the total transmit power constraint, which is a non-convex, multi-timescale, and mixed-integer programming problem. To facilitate the process, we put forth an entirely new paradigm named divide-and-rule. Specifically, we first solve the short-term optimization problem regarding user pairing as well as power allocation and the long-term decision-making problem with respect to recommendation and caching, respectively. On this basis, an iterative algorithm is developed to optimize all the optimization variables alternately. Particularly, for solving the short-timescale problem, graph theory enabled NOMA user grouping and efficient inter-group power control manners are invoked. Meanwhile, a dynamic programming approach and a complexity-controllable swap-then-compare method with convergence insurance are designed to derive the caching and recommendation policies, respectively. From Monte-Carlo simulation, we show the superiority of the proposed joint optimization method in terms of both system latency and cache hit ratio when compared to extensive benchmark strategies.

Original languageEnglish
Pages (from-to)1600-1614
Number of pages15
JournalIEEE Journal on Selected Areas in Communications
Issue number5
Publication statusPublished - 2022 May 1

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Joint Content Caching, Recommendation, and Transmission Optimization for Next Generation Multiple Access Networks'. Together they form a unique fingerprint.

Cite this