面向无人机航拍图像传输的信道弹性编解码方法OA

Unmanned aerial vehicle(UAV)is recognized as the core component of the low-altitude economy network.It is crucial to achieve efficient image transmission in air-to-ground communication.During mission execution,rapid channel fluctuations are induced by changes in the flight area,including variations in signal-to-noise ratio(SNR)and transmission rate.To adapt to these changes,a channel elastic encoding and decoding method for UAV image transmis-sion was proposed.This method included a lightweight feature extraction module,a channel feature enhancement module,a transmission rate adaptive module,and symmetric decoding modules.The lightweight feature extraction module con-tained residual blocks and the mobile Mamba,which could quickly extract local detail features and global context.It could effectively obtain rich semantic features from aerial images.The feature enhancement module enhanced key seman-tic features according to the channel state SNR.The transmission rate adaptive module adjusted the size of transmitted fea-tures based on the channel transmission rate.Extensive experiments on the aerial image dataset(AID)demonstrate that the proposed method achieves higher transmission accuracy than the state-of-the-art methods.