An OFDM Intelligent Receiver Based on an Adaptive Cooperative Network

• 1、 Beijing University of Posts and Telecommunications
• 2、Beijing University of Posts and Telecommunications

Abstract：To address the limited robustness of conventional OFDM receivers in complex channel environments and the fixed computational complexity of existing deep learning-based receivers, an intelligent OFDM receiver based on an Adaptive Cooperative Network (ACNet) is investigated. The proposed approach constructs subnetworks with different functional focuses and introduces a gating mechanism to dynamically select inference paths according to channel conditions. This design enables the receiver to utilize more network capacity to improve detection performance under challenging channel conditions while reducing computational complexity when the channel quality is favorable. In addition, a self-attention mechanism is employed to collaboratively model the outputs of different subnetworks and achieve adaptive fusion of multi-branch features. A simulation dataset is constructed based on an OFDM communication system, and the performance of the proposed model is evaluated under various channel conditions. Simulation results demonstrate that the proposed method achieves improved bit error rate performance in different channel environments while effectively reducing computational complexity, indicating good robustness and application potential.

Keywords： wireless communication technology OFDM receiver deep learning receiver adaptive cooperative network