A Novel Content Caching and Delivery Scheme for Millimeter Wave Device-to-Device Communications

TL;DR

The paper tackles content caching for cache-enabled D2D networks operating in mmWave, with fallback to sub-6 GHz cellular links, by aiming to maximize the average successful LoS reception probability $ASLP$. It develops a convexified optimization for caching probabilities $q_i$ under a cache budget, using a KKT-based closed-form solution, and couples this with a LoS-aware user association policy to boost offloading. The performance is analyzed through stochastic geometry, and validated numerically, showing superior $ASLP$, offloading, and energy efficiency compared with a cache-hit–maximizing baseline, particularly in ultra-dense, high-blockage environments. The framework provides a practical approach to improving QoS-driven content delivery in next-generation wireless networks with mmWave D2D capabilities.

Abstract

A novel content caching strategy is proposed for a cache enabled device-to-device (D2D) network where the user devices are allowed to communicate using millimeter wave (mmWave) D2D links (> 6 GHz) as well as conventional sub 6 GHz cellular links. The proposed content placement strategy maximizes the successful content delivery probability of a line of sight D2D link. Furthermore, a heuristic algorithm is proposed for efficient content delivery. The overall scheme improves the successful traffic offloading gain of the network compared to conventional cache-hit maximizing content placement and delivery strategies. Significant energy efficiency improvements can also be achieved in ultra-dense networks.

Figures (3)

• Figure 1: The behavior of the overall ${\rm SP}$ of the system with $\lambda_u$ for different [$D_{L}$,R,$\epsilon]$ combinations.
• Figure 2: Offloading and the successful offloading probabilities of the D2D network for 1 Gbps data rate, where $\epsilon =1.2$.
• Figure 3: The Energy Efficiency of the D2D link with $\lambda_u$ for different [$D_{L}$,R,$\epsilon]$ combinations.