Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Hybrid Quantum-Classical Autoencoder Framework for End-to-End Communication Systems

Bolun Zhang, Gan Zheng, Nguyen Van Huynh

TL;DR

The paper tackles end-to-end wireless E2E communication under fading by introducing a hybrid quantum-classical autoencoder (QAE) that uses parallel PQCs at the transmitter to map one-hot messages into complex transmissions using amplitude embedding. It shows that QAE can achieve BLER comparable to classical DL-based autoencoders and conventional codes while cutting trainable parameters by roughly $50\%$, and it exhibits superior BLER convergence during training. The work highlights memory and convergence benefits from quantum representations and provides a practical-path demonstration using PennyLane, while acknowledging the current limitations of quantum simulation on classical hardware. It paves the way for hardware-aware quantum optimization and future integrations with advanced quantum architectures (e.g., quantum CNNs, OFDM, MIMO) in communications.

Abstract

This paper investigates the application of quantum machine learning to End-to-End (E2E) communication systems in wireless fading scenarios. We introduce a novel hybrid quantum-classical autoencoder architecture that combines parameterized quantum circuits with classical deep neural networks (DNNs). Specifically, we propose a hybrid quantum-classical autoencoder (QAE) framework to optimize the E2E communication system. Our results demonstrate the feasibility of the proposed hybrid system, and reveal that it is the first work that can achieve comparable block error rate (BLER) performance to classical DNN-based and conventional channel coding schemes, while significantly reducing the number of trainable parameters. Additionally, the proposed QAE exhibits steady and superior BLER convergence over the classical autoencoder baseline.

A Hybrid Quantum-Classical Autoencoder Framework for End-to-End Communication Systems

TL;DR

The paper tackles end-to-end wireless E2E communication under fading by introducing a hybrid quantum-classical autoencoder (QAE) that uses parallel PQCs at the transmitter to map one-hot messages into complex transmissions using amplitude embedding. It shows that QAE can achieve BLER comparable to classical DL-based autoencoders and conventional codes while cutting trainable parameters by roughly , and it exhibits superior BLER convergence during training. The work highlights memory and convergence benefits from quantum representations and provides a practical-path demonstration using PennyLane, while acknowledging the current limitations of quantum simulation on classical hardware. It paves the way for hardware-aware quantum optimization and future integrations with advanced quantum architectures (e.g., quantum CNNs, OFDM, MIMO) in communications.

Abstract

This paper investigates the application of quantum machine learning to End-to-End (E2E) communication systems in wireless fading scenarios. We introduce a novel hybrid quantum-classical autoencoder architecture that combines parameterized quantum circuits with classical deep neural networks (DNNs). Specifically, we propose a hybrid quantum-classical autoencoder (QAE) framework to optimize the E2E communication system. Our results demonstrate the feasibility of the proposed hybrid system, and reveal that it is the first work that can achieve comparable block error rate (BLER) performance to classical DNN-based and conventional channel coding schemes, while significantly reducing the number of trainable parameters. Additionally, the proposed QAE exhibits steady and superior BLER convergence over the classical autoencoder baseline.
Paper Structure (12 sections, 7 equations, 5 figures, 2 tables)

This paper contains 12 sections, 7 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. System Model
  3. Hybrid Quantum-Classical Autoencoder for End-to-End Communication Systems
  4. Quantum Neural Networks
  5. Hybrid Quantum-Classical Autoencoder Architecture
  6. Parameter and Complexity Analysis
  7. Simulation Results
  8. Experimental Settings
  9. Performance Evaluation
  10. BLER vs SNR
  11. BLER Convergence
  12. Conclusion

Figures (5)

  • Figure 1: Architectures of (a) traditional communication systems and (b) Autoencoder-based E2E communication systems.
  • Figure 2: Overview of hybrid quantum-classical autoencoder.
  • Figure 3: The $l$-th layer of quantum circuit with 4-qubit.
  • Figure 4: BLER comparisons of (a) (4,4), (b) (7,4), and (c) (8,8) under Rayleigh, Rician, 3GPP block fading.
  • Figure 5: BLER convergence (smoothed) comparisons between QAE and AE for Rayleigh block fading.