Probabilistic Analysis of Operating Modes in Cache-Enabled Full-Duplex D2D Networks

TL;DR

This paper analyzes the probabilistic operating modes of cache-enabled full-duplex D2D networks, providing insights into their performance and mode probabilities under realistic caching policies.

Contribution

It introduces a comprehensive probabilistic model for all operating modes in cache-enabled FD-D2D networks considering realistic caching policies.

Findings

Quantifies probabilities of different operating modes.

Provides PMFs of nodes in each mode.

Offers insights into caching performance in FD-D2D networks.

Abstract

With the extensive acquisition of various mobile applications, cellular networks are facing challenges due to exponentially growing demand for high data rate, which causes a great burden on mobile core networks and backhaul links. Cache-enabled Device-to-Device (D2D) communication, which is recognized as one of the key enablers of the fifth generation (5G) cellular network, is a promising solution to alleviate this problem. However, conventional half-duplex (HD) communication may not be sufficient to provide fast enough content delivery over D2D links in order to meet strict latency targets of emerging D2D applications. In-band full-duplex (FD), with its capability of allowing simultaneous transmission and reception, can provide more content delivery opportunities, thus resulting improved spectral efficiency and latency reduction. However, given the random nature of the cached contents in user devices and users' random requests, it is unlikely to consider all involving nodes in content exchange collaborations as a pure HD or FD network. In this paper, we aim to analyze the caching perspective of a finite network of D2D nodes in which each node is endowed with FD capability and utilize a more realistic caching policy. We model and analyze all possible operating modes for an arbitrary device, which we compute the probability of occurrence of each mode along with the Probability Mass Functions (PMFs) of nodes that are operating in all possible modes. Our analysis concretely quantize all possible outcomes that strongly depend on the random nature of the caching parameters, yielding to have an accurate insight on the caching performance and all possible outcomes of the cache-enabled FD-D2D network.