Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Cooperative Caching Towards Efficient Spectrum Utilization in Cognitive-IoT Networks

Nadia Abdolkhani, Walaa Hamouda

TL;DR

This work addresses spectrum scarcity in cognitive IoT by integrating cooperative caching with DRL to jointly optimize content placement and channel access. It introduces a discrete-time MDP solved by a Double Deep Q-Network enhanced with a Zipf-aware UCB exploration (UCBZ) to navigate uncertain channel occupancy and dynamic content popularity. Empirical results show substantial gains in average sum rate and reduced delay compared to non-cooperative, epsilon-greedy, and LRU baselines, highlighting the value of edge caching and cooperative spectrum sharing. The approach offers a novel perspective on CIoT network design, leveraging DRL-guided caching to improve throughput under constrained resources. Practical implications include improved QoS for CIoT applications and more efficient spectrum utilization in dense IoT deployments.

Abstract

In cognitive Internet of Things (CIoT) networks, efficient spectrum sharing is essential to address increasing wireless demands. This paper presents a novel deep reinforcement learning (DRL)-based approach for joint cooperative caching and spectrum access coordination in CIoT networks, enabling the CIoT agents to collaborate with primary users (PUs) by caching PU content and serving their requests, fostering mutual benefits. The proposed DRL framework jointly optimizes caching policy and spectrum access under challenging conditions. Unlike traditional cognitive radio (CR) methods, where CIoT agents vacate the spectrum for PUs, or relaying techniques, which merely support spectrum sharing, caching brings data closer to the edge, reducing latency by minimizing retrieval distance. Simulations demonstrate that our approach outperforms others in lowering latency, increasing CIoT and PU cache hit rates, and enhancing network throughput. This approach redefines spectrum sharing, offering a fresh perspective on CIoT network design and illustrating the potential of DRL-guided caching to highlight the benefits of collaboration over dynamic spectrum access scenarios, elevating CIoT performance under constrained resources.

Cooperative Caching Towards Efficient Spectrum Utilization in Cognitive-IoT Networks

TL;DR

This work addresses spectrum scarcity in cognitive IoT by integrating cooperative caching with DRL to jointly optimize content placement and channel access. It introduces a discrete-time MDP solved by a Double Deep Q-Network enhanced with a Zipf-aware UCB exploration (UCBZ) to navigate uncertain channel occupancy and dynamic content popularity. Empirical results show substantial gains in average sum rate and reduced delay compared to non-cooperative, epsilon-greedy, and LRU baselines, highlighting the value of edge caching and cooperative spectrum sharing. The approach offers a novel perspective on CIoT network design, leveraging DRL-guided caching to improve throughput under constrained resources. Practical implications include improved QoS for CIoT applications and more efficient spectrum utilization in dense IoT deployments.

Abstract

In cognitive Internet of Things (CIoT) networks, efficient spectrum sharing is essential to address increasing wireless demands. This paper presents a novel deep reinforcement learning (DRL)-based approach for joint cooperative caching and spectrum access coordination in CIoT networks, enabling the CIoT agents to collaborate with primary users (PUs) by caching PU content and serving their requests, fostering mutual benefits. The proposed DRL framework jointly optimizes caching policy and spectrum access under challenging conditions. Unlike traditional cognitive radio (CR) methods, where CIoT agents vacate the spectrum for PUs, or relaying techniques, which merely support spectrum sharing, caching brings data closer to the edge, reducing latency by minimizing retrieval distance. Simulations demonstrate that our approach outperforms others in lowering latency, increasing CIoT and PU cache hit rates, and enhancing network throughput. This approach redefines spectrum sharing, offering a fresh perspective on CIoT network design and illustrating the potential of DRL-guided caching to highlight the benefits of collaboration over dynamic spectrum access scenarios, elevating CIoT performance under constrained resources.

Paper Structure

This paper contains 7 sections, 9 equations, 5 figures, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Optimal Joint Channel Access Coordination and caching policy strategy
  4. Double Deep Q-Network
  5. Proposed Upper Confidence Bound Strategy
  6. Simulation Model and Results
  7. Conclusions

Figures (5)

  • Figure 1: The proposed DDQN-UCBZ algorithm
  • Figure 2: The CIoT agent's ASR during episodes of training.
  • Figure 3: The CIoT agent's service delay across different values of $L$ the number of PU occupying time slots.
  • Figure 4: The CIoT agent's service delay across different CIoT transmission power $P_s^t$.
  • Figure 5: The CIoT agent's cache-hit rate across different zipf parameters $\gamma_s$.