View in EDS

Bi-directional Recurrent Neural Network with Conditional Iterative Learning in End-to-End Wireless Systems.

Saved in:
Bibliographic Details
Title: Bi-directional Recurrent Neural Network with Conditional Iterative Learning in End-to-End Wireless Systems.
Authors: Vijayarani, Ramachandran1 (AUTHOR) vijee.ram@gmail.com, R., Aparna1 (AUTHOR), K., Divya1 (AUTHOR), E., Sharmila1 (AUTHOR)
Source: Mobile Networks & Applications. Aug2026, Vol. 31 Issue 3/4, p306-313. 8p.
Subjects: Recurrent neural networks, Sequential learning, Bit error rate, Artificial neural networks, Wireless communications performance, Wireless channels, Generative adversarial networks, Wireless communications
Abstract: The existing methods for end-to-end (E2E) wireless communication systems are simple and advantageous in compared to traditional wireless communication methods. But the lack of a channel model with differentiable parameters presents a challenge in training transmitters, hindering the attainment of optimal performance. To solve this problem, conditional generative adversarial networks (CGANs) and deep neural networks (DNN) are incorporated, however which have lower performance measured by the rate of errors and also prone to loss and delay. The proposed model harnesses the power of Bi-RNN architecture, renowned for its capacity to collect temporal dependencies in data presented sequentially, while incorporating conditional iterative learning techniques to enhance its adaptability and efficiency in wireless communication scenarios. By leveraging the bidirectional nature of Bi-RNNs, the system aims to comprehensively understand the evolution of the channel over time, facilitating more accurate predictions and adaptive responses. Furthermore, the proposed learning-based method operates more effectively in terms of bit error rate and block error rates, loss compared to traditional methods. Furthermore, the proposed method can be used in wireless communication by making use of shared communication resources, so improving the overall dependability and performance of the system. [ABSTRACT FROM AUTHOR]
Copyright of Mobile Networks & Applications is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
Description
Abstract:The existing methods for end-to-end (E2E) wireless communication systems are simple and advantageous in compared to traditional wireless communication methods. But the lack of a channel model with differentiable parameters presents a challenge in training transmitters, hindering the attainment of optimal performance. To solve this problem, conditional generative adversarial networks (CGANs) and deep neural networks (DNN) are incorporated, however which have lower performance measured by the rate of errors and also prone to loss and delay. The proposed model harnesses the power of Bi-RNN architecture, renowned for its capacity to collect temporal dependencies in data presented sequentially, while incorporating conditional iterative learning techniques to enhance its adaptability and efficiency in wireless communication scenarios. By leveraging the bidirectional nature of Bi-RNNs, the system aims to comprehensively understand the evolution of the channel over time, facilitating more accurate predictions and adaptive responses. Furthermore, the proposed learning-based method operates more effectively in terms of bit error rate and block error rates, loss compared to traditional methods. Furthermore, the proposed method can be used in wireless communication by making use of shared communication resources, so improving the overall dependability and performance of the system. [ABSTRACT FROM AUTHOR]
ISSN:1383469X
DOI:10.1007/s11036-026-02517-5