Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Timely NextG Communications with Decoy Assistance against Deep Learning-based Jamming

Maice Costa, Yalin E. Sagduyu

TL;DR

This work tackles the challenge of delivering time-sensitive information in NextG environments under a deep-learning-powered jammer constrained by average power. It introduces a decoy-based anti-jamming approach, where decoy transmissions confuse the jammer and reduce disruption to real messages, and analyzes its impact on information freshness (AoI) and reliability. The study compares two status-update policies—a buffered M/D/1 queue and a bufferless just-in-time (JIT) scheme—and derives closed-form PAoI expressions along with guidelines for optimal update rates. Numerical results indicate that decoys can lower average jamming power and improve AoI by up to 8–11%, validating the practicality of decoy-assisted anti-jamming for timely and covert NextG communications in adversarial settings.

Abstract

We consider the transfer of time-sensitive information in next-generation (NextG) communication systems in the presence of a deep learning based eavesdropper capable of jamming detected transmissions, subject to an average power budget. A decoy-based anti-jamming strategy is presented to confuse a jammer, causing it to waste power when disrupting decoy messages instead of real messages. We investigate the effectiveness of the anti-jamming strategy to guarantee timeliness of NextG communications in addition to reliability objectives, analyzing the Age of Information subject to jamming and channel effects. We assess the effect of power control, which determines the success of a transmission but also affects the accuracy of the adversary's detection, making it more likely for the jammer to successfully identify and jam the communication. The results demonstrate the feasibility of mitigating eavesdropping and jamming attacks in NextG communications with information freshness objectives using a decoy to guarantee timely information transfer.

Timely NextG Communications with Decoy Assistance against Deep Learning-based Jamming

TL;DR

This work tackles the challenge of delivering time-sensitive information in NextG environments under a deep-learning-powered jammer constrained by average power. It introduces a decoy-based anti-jamming approach, where decoy transmissions confuse the jammer and reduce disruption to real messages, and analyzes its impact on information freshness (AoI) and reliability. The study compares two status-update policies—a buffered M/D/1 queue and a bufferless just-in-time (JIT) scheme—and derives closed-form PAoI expressions along with guidelines for optimal update rates. Numerical results indicate that decoys can lower average jamming power and improve AoI by up to 8–11%, validating the practicality of decoy-assisted anti-jamming for timely and covert NextG communications in adversarial settings.

Abstract

We consider the transfer of time-sensitive information in next-generation (NextG) communication systems in the presence of a deep learning based eavesdropper capable of jamming detected transmissions, subject to an average power budget. A decoy-based anti-jamming strategy is presented to confuse a jammer, causing it to waste power when disrupting decoy messages instead of real messages. We investigate the effectiveness of the anti-jamming strategy to guarantee timeliness of NextG communications in addition to reliability objectives, analyzing the Age of Information subject to jamming and channel effects. We assess the effect of power control, which determines the success of a transmission but also affects the accuracy of the adversary's detection, making it more likely for the jammer to successfully identify and jam the communication. The results demonstrate the feasibility of mitigating eavesdropping and jamming attacks in NextG communications with information freshness objectives using a decoy to guarantee timely information transfer.

Paper Structure

This paper contains 11 sections, 11 equations, 6 figures.

Table of Contents

  1. Introduction
  2. System Model
  3. Communication Model
  4. Adversary Model
  5. Status Updating Model
  6. Performance Analysis
  7. Outage Probabilities
  8. Signal Detection at Adversary
  9. Communication Timeliness
  10. Numerical Results
  11. Conclusion

Figures (6)

  • Figure 1: Network model.
  • Figure 2: Classifier accuracy vs. SNR between T and J, and effect of number of I/Q symbols per packet.
  • Figure 3: Sawtooth curve - A sample path for AoI.
  • Figure 4: Jamming average power vs. T and D activity.
  • Figure 5: PAoI vs. decoy and transmit probabilities.
  • ...and 1 more figures