Scholarlink

<p>Orthogonal frequency division multiplexing (OFDM) is a widely adopted modulation scheme in modern wireless communication systems, particularly in fifth-generation (5G) networks. Orthogonal time–frequency space (OTFS) is an emerging modulation technique currently under active investigation for future sixth-generation (6G) networks due to its promising performance in high-mobility, doubly dispersive channels. Despite the benefits of both approaches, using them independently may lead to less-than-ideal efficiency because there is no common processing framework. This research presents a novel neural network (NN)-based architecture that combines the processing of both modulation techniques into a single system to address this problem. This integrated framework enables the development of intelligent and flexible communication systems that can operate reliably across a wide range of operational conditions. The proposed NN model maintains minimal computing complexity while managing many signal-processing tasks concurrently. Simulation results demonstrate improved bit-error-rate (BER) performance with the integration of OTFS and NN, and comparisons with OFDM-based results highlight the advantages and reliability of the proposed unified strategy.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A unified neural network model for OFDM and OTFS modulations in wireless systems

  • 期刊论文

  • 发表日期：

    2026年4月21日

查看全文
  • Sneha Chennamsetty,
  • Subbarao Boddu

摘要

Orthogonal frequency division multiplexing (OFDM) is a widely adopted modulation scheme in modern wireless communication systems, particularly in fifth-generation (5G) networks. Orthogonal time–frequency space (OTFS) is an emerging modulation technique currently under active investigation for future sixth-generation (6G) networks due to its promising performance in high-mobility, doubly dispersive channels. Despite the benefits of both approaches, using them independently may lead to less-than-ideal efficiency because there is no common processing framework. This research presents a novel neural network (NN)-based architecture that combines the processing of both modulation techniques into a single system to address this problem. This integrated framework enables the development of intelligent and flexible communication systems that can operate reliably across a wide range of operational conditions. The proposed NN model maintains minimal computing complexity while managing many signal-processing tasks concurrently. Simulation results demonstrate improved bit-error-rate (BER) performance with the integration of OTFS and NN, and comparisons with OFDM-based results highlight the advantages and reliability of the proposed unified strategy.