Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Cache-Aided Communications in MISO Networks with Dynamic User Behavior

Milad Abolpour, MohammadJavad Salehi, Antti Tölli

TL;DR

This work tackles the challenge of implementing coded caching in dynamic MISO networks where users freely enter and depart. It introduces a universal shared-cache framework with content placement independent of user count and two CC delivery strategies that accommodate any user-to-profile association, including regimes not covered by prior work. Closed-form DoF expressions quantify the caching and spatial multiplexing gains and reveal DoF losses due to nonuniform user distribution, while a SIC-free data delivery method plus fast iterative beamforming optimizes finite-SNR performance. Numerical results demonstrate substantial finite-SNR gains over unicasting and confirm the scheme's universality across various network parameters, making cache-aided dynamic networks more practically viable. The combination of universal placement, two flexible delivery strategies, and efficient beamforming offers a robust, implementable approach for next-generation wireless networks with dynamic user behavior.

Abstract

Coded caching (CC) can substantially enhance network performance by leveraging memory as an additional communication resource. However, the use of CC is challenging in various practical applications due to dynamic user behavior. The existing solutions, based on shared caching, cannot directly handle all scenarios where users freely enter and depart the network at any time as they are constrained by specific conditions on network parameters. This paper proposes a universally applicable shared-caching scheme for dynamic setups without any restriction on network parameters. The closed-form expressions for the achievable degrees of freedom (DoF) are computed for the resulting generalized scheme, and are shown to achieve the existing optimal bounds of the shared-cache model. Furthermore, a successive-interference-cancellation-free extension based on a fast iterative optimized beamformer design is devised to optimize the use of excess spatial dimensions freed by cache-aided interference cancellation. Extensive numerical experiments are carried out to assess the performance of the proposed scheme. In particular, the results demonstrate that while a dynamic setup may achieve a DoF substantially lower than the optimal DoF of shared caching, our proposed scheme significantly improves the performance at the finite signal-to-noise ratio compared to unicasting, which only benefits from the local caching gain.

Cache-Aided Communications in MISO Networks with Dynamic User Behavior

TL;DR

This work tackles the challenge of implementing coded caching in dynamic MISO networks where users freely enter and depart. It introduces a universal shared-cache framework with content placement independent of user count and two CC delivery strategies that accommodate any user-to-profile association, including regimes not covered by prior work. Closed-form DoF expressions quantify the caching and spatial multiplexing gains and reveal DoF losses due to nonuniform user distribution, while a SIC-free data delivery method plus fast iterative beamforming optimizes finite-SNR performance. Numerical results demonstrate substantial finite-SNR gains over unicasting and confirm the scheme's universality across various network parameters, making cache-aided dynamic networks more practically viable. The combination of universal placement, two flexible delivery strategies, and efficient beamforming offers a robust, implementable approach for next-generation wireless networks with dynamic user behavior.

Abstract

Coded caching (CC) can substantially enhance network performance by leveraging memory as an additional communication resource. However, the use of CC is challenging in various practical applications due to dynamic user behavior. The existing solutions, based on shared caching, cannot directly handle all scenarios where users freely enter and depart the network at any time as they are constrained by specific conditions on network parameters. This paper proposes a universally applicable shared-caching scheme for dynamic setups without any restriction on network parameters. The closed-form expressions for the achievable degrees of freedom (DoF) are computed for the resulting generalized scheme, and are shown to achieve the existing optimal bounds of the shared-cache model. Furthermore, a successive-interference-cancellation-free extension based on a fast iterative optimized beamformer design is devised to optimize the use of excess spatial dimensions freed by cache-aided interference cancellation. Extensive numerical experiments are carried out to assess the performance of the proposed scheme. In particular, the results demonstrate that while a dynamic setup may achieve a DoF substantially lower than the optimal DoF of shared caching, our proposed scheme significantly improves the performance at the finite signal-to-noise ratio compared to unicasting, which only benefits from the local caching gain.
Paper Structure (23 sections, 3 theorems, 65 equations, 9 figures, 4 tables)

This paper contains 23 sections, 3 theorems, 65 equations, 9 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Prior Art
  3. Our Contributions
  4. System Model
  5. Resource Allocation and Data Delivery
  6. System operation for $\alpha \leq \hat{\eta}$
  7. System operation for $\alpha > \hat{\eta}$
  8. Transmission Strategy A
  9. Transmission Strategy B
  10. Unicast (UC) Data Delivery
  11. SIC-Free Data Delivery
  12. DoF Analysis
  13. Beamforming Design
  14. Numerical Results
  15. DoF Analysis
  16. ...and 8 more sections

Key Result

Theorem 1

Consider a dynamic MISO network with the spatial multiplexing gain of $\alpha$, cache ratio $\gamma$ and the delivery parameter $\hat{\eta}$. If the system operates with Strategy A in the CC delivery step, the DoF is given by: where $\beta^{\prime}=\beta \binom{P-{\Bar{t}}-1}{Q-{\Bar{t}}-1}$ and $D(\delta_{r})= \phi_{r}$ if $\delta_{r} \neq 0$; otherwise, $D\left( \delta_{r} \right)=0$. If Strate

Figures (9)

  • Figure 1: A dynamic cache-aided network with $P=3$ profiles, where users $\left\lbrace 1,2,\cdots, 7 \right\rbrace$ are present and do not depart the network, users $\left\lbrace 8,9,10 \right\rbrace$ and $\left\lbrace 11,12,13,14 \right\rbrace$ are departing and entering the network, respectively.
  • Figure 2: System model for a dynamic coded caching setup, where $P=2$, $\gamma=\frac{1}{2}$, ${\Bar{t}}=1$, $\alpha=4$, $\hat{\eta}=4$, $Q=2$ and $\beta=3$. During the placement phase, each user assigned to profiles $A$ and $B$ stores the cache content associated with those profiles. For the delivery phase, user $5$ is served via unicasting and other users are served via $3$ multicast transmissions.
  • Figure 3: The elevation process to create the set of served users during the transmission triple ${\mathsf{A}}_{i}$ for Strategy A in a network with $\gamma=\frac{2}{3}$, ${\Bar{t}} =2$, $P=3$, $\alpha = 2$, $\hat{\eta} =3$, $\eta_{1} = \delta_{1} = 3$, $\eta_{2} = \delta_{2} = 2$, $\eta_{3} = \delta_{3} = 1$, $Q=3$, $\beta = 2$, ${\mathsf{A}}_{1} = (1,1,1)$, ${\mathsf{A}}_{2} = (1,2,1)$ and ${\mathsf{A}}_{3} = (1,3,1)$.
  • Figure 4: The DoF versus $\hat{\eta}$ with different values of $Q$, $K=30$, $\alpha=8$, $P=5$, $\gamma=0.2$ and ${\Bar{t}}=1$ for: (a) $\sigma=0$, (b) $\sigma=0.63$ and (c) $\sigma=2.19$. Here, A and B show that the system operates with Strategy A and Strategy B, respectively. The number of user profiles is fixed in all subfigures (a), (b), and (c).
  • Figure 5: The average of the maximum achievable DoF $\left( {\rm DoF}_{M} \right)$ versus the standard deviation $(\sigma)$ with $K=30$, $\gamma=0.2$, $P=5$, ${\Bar{t}}=1$ and $\alpha \in \left\lbrace 5,8 \right\rbrace$.
  • ...and 4 more figures

Theorems & Definitions (10)

  • Remark 1
  • Example 1
  • Example 2
  • Example 3
  • Theorem 1
  • Corollary 1
  • Remark 2
  • Remark 3
  • Remark 4
  • Lemma 1