Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

DRL-Based Maximization of the Sum Cross-Layer Achievable Rate for Networks Under Jamming

Abdul Basit, Muddasir Rahim, Tri Nhu Do, Nadir Adam, Georges Kaddoum

TL;DR

This work tackles channel-access in multi-cell quasi-static wireless networks facing malicious jamming. It formulates the problem as a partially observable Markov decision process (POMDP) and introduces a residual-network based deep Q-network (ResNet-DQN) for the intelligent UEs (iUEs) to learn robust transmission policies. The approach demonstrates near-optimal cross-layer performance by maximizing the sum cross-layer achievable rate (SCLAR) while mitigating collisions and jamming, validated through extensive simulations in a multi-cell, jamming-rich environment. The results indicate strong adaptability and scalability, with potential impact on secure and efficient MAC design in next-generation wireless networks.

Abstract

In quasi-static wireless networks characterized by infrequent changes in the transmission schedules of user equipment (UE), malicious jammers can easily deteriorate network performance. Accordingly, a key challenge in these networks is managing channel access amidst jammers and under dynamic channel conditions. In this context, we propose a robust learning-based mechanism for channel access in multi-cell quasi-static networks under jamming. The network comprises multiple legitimate UEs, including predefined UEs (pUEs) with stochastic predefined schedules and an intelligent UE (iUE) with an undefined transmission schedule, all transmitting over a shared, time-varying uplink channel. Jammers transmit unwanted packets to disturb the pUEs' and the iUE's communication. The iUE's learning process is based on the deep reinforcement learning (DRL) framework, utilizing a residual network (ResNet)-based deep Q-Network (DQN). To coexist in the network and maximize the network's sum cross-layer achievable rate (SCLAR), the iUE must learn the unknown network dynamics while concurrently adapting to dynamic channel conditions. Our simulation results reveal that, with properly defined state space, action space, and rewards in DRL, the iUE can effectively coexist in the network, maximizing channel utilization and the network's SCLAR by judiciously selecting transmission time slots and thus avoiding collisions and jamming.

DRL-Based Maximization of the Sum Cross-Layer Achievable Rate for Networks Under Jamming

TL;DR

This work tackles channel-access in multi-cell quasi-static wireless networks facing malicious jamming. It formulates the problem as a partially observable Markov decision process (POMDP) and introduces a residual-network based deep Q-network (ResNet-DQN) for the intelligent UEs (iUEs) to learn robust transmission policies. The approach demonstrates near-optimal cross-layer performance by maximizing the sum cross-layer achievable rate (SCLAR) while mitigating collisions and jamming, validated through extensive simulations in a multi-cell, jamming-rich environment. The results indicate strong adaptability and scalability, with potential impact on secure and efficient MAC design in next-generation wireless networks.

Abstract

In quasi-static wireless networks characterized by infrequent changes in the transmission schedules of user equipment (UE), malicious jammers can easily deteriorate network performance. Accordingly, a key challenge in these networks is managing channel access amidst jammers and under dynamic channel conditions. In this context, we propose a robust learning-based mechanism for channel access in multi-cell quasi-static networks under jamming. The network comprises multiple legitimate UEs, including predefined UEs (pUEs) with stochastic predefined schedules and an intelligent UE (iUE) with an undefined transmission schedule, all transmitting over a shared, time-varying uplink channel. Jammers transmit unwanted packets to disturb the pUEs' and the iUE's communication. The iUE's learning process is based on the deep reinforcement learning (DRL) framework, utilizing a residual network (ResNet)-based deep Q-Network (DQN). To coexist in the network and maximize the network's sum cross-layer achievable rate (SCLAR), the iUE must learn the unknown network dynamics while concurrently adapting to dynamic channel conditions. Our simulation results reveal that, with properly defined state space, action space, and rewards in DRL, the iUE can effectively coexist in the network, maximizing channel utilization and the network's SCLAR by judiciously selecting transmission time slots and thus avoiding collisions and jamming.
Paper Structure (51 sections, 52 equations, 10 figures, 4 tables, 1 algorithm)

This paper contains 51 sections, 52 equations, 10 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Research Question and Problem of Interest
  4. Contributions
  5. System Model
  6. Uplink Multi-Cell Multi-User Multiple Input Single Output (UL MC-MU-MISO) Network
  7. Slotted Time-Based MAC
  8. Quasi-Static Networking
  9. Stochastic pUEs
  10. Random Jammers
  11. DRL-Based iUE
  12. Signal Modelling in the Network
  13. Channel Modeling
  14. Signal Received at $\mathrm{C}^{[k]}$
  15. Linear Detection at $\mathrm{C}^{[k]}$ and SINR Formulation
  16. ...and 36 more sections

Figures (10)

  • Figure 1: Illustration of the considered network in the presence of malicious jammers.
  • Figure 2: Illustration of the time frame and transmission schedule of UEs and jammers in the considered network.
  • Figure 3: Extraction of $a^*$ from [$\pi^*$]
  • Figure 4: Multi-cell smart grid network deployment, with pUEs, jammers, and iUEs randomly distributed across cells, simulating realistic interference and diverse access protocols.
  • Figure 5: iUE's transmission actions (before training).
  • ...and 5 more figures

Theorems & Definitions (1)

  • Remark 1