音频传输中的瞬态频域扰动信号调制技术研究  

Research on transient frequency domain disturbance signal modulation technology in audio transmission

作　　者：褚倢 王琼[1] CHU Jie;WANG Qiong(North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学,山西太原030051

出　　处：《现代电子技术》2023年第23期75-78,共4页Modern Electronics Technique

摘　　要：音频信号传输时受到因噪声引起的瞬态频域扰动信号影响,导致整体调制效果较差。为此,文中研究具备去噪能力的音频瞬态频域扰动信号调制技术。此技术基于傅里叶变换去除音频信号传输中的噪声信息后,使用基于卷积过程神经网络的瞬态频域扰动信号识别方法,以信号识别的方式提取音频传输中的瞬态频域扰动信号。将提取到的瞬态频域扰动信号输入基于DDS的瞬态频域扰动信号调制器中,调制信号频率满足期望状态,完成瞬态频域扰动信号调制。实验结果显示:所提技术对数字电视系统传输的音频信号进行去噪、瞬态频域扰动信号识别和调制后,有效滤除了音频信号中的噪声信息,瞬态频域扰动信号识别结果准确,且具备有效调制音频传输中瞬态频域扰动信号数据传输速率的能力。The transmission of audio signals is affected by transient frequency domain disturbance signals caused by noise,which results in poor overall modulation performance.Therefore,this research focuses on audio transient frequency domain disturbance signal modulation technology with denoising capabilities.In this technology,the noise information in audio signal transmission is removed according to Fourier transform.And then,a transient frequency domain disturbance signal recognition method based on convolutional process neural network is used to extract transient frequency domain disturbance signal in audio transmission by means of signal recognition.The extracted transient frequency domain disturbance signal is input into the DDS⁃based transient frequency domain disturbance signal modulator,and the signal frequency is modulated to meet the expected state to complete the signal modulation.The experimental results show that the proposed technology effectively filters out noise information in the audio signal transmitted by the digital television system after denoising,transient frequency domain disturbance signal recognition and modulation,the recognition results of the transient frequency domain disturbance signal are accurate,and the proposed technology has the ability to effectively modulate the data transmission rate of transient frequency domain disturbance signals in audio transmission.

关 键 词：音频传输 瞬态频域 扰动信号 调制技术 小波消噪 卷积过程神经网络 信号分类 直接数字合成技术 

分 类 号：TN92-34[电子电信—通信与信息系统]