短时傅里叶变换
卷积神经网络
计算机科学
稳健性(进化)
傅里叶变换
人工智能
模式识别(心理学)
算法
时频分析
图形
循环神经网络
语音识别
人工神经网络
雷达
数学
傅里叶分析
电信
数学分析
生物化学
化学
理论计算机科学
基因
作者
Xuguang Xu,Cunqian Feng
标识
DOI:10.1109/lawp.2023.3267238
摘要
Signal-to-noise ratio (SNR) is an important prior information for many micro-motion echo processing approaches and techniques. Estimating SNR in advance can effectively enhance the performance of such technologies. This letter proposes a long-term recurrent convolutional network (LRCN)-based SNR estimation method for cone-shaped target. First, mathematical expression of the echo is derived by studying the cone-shaped target micro-motion model, and hence the characteristics of the echo and its short-time fourier transform are analyzed. Second, the motivation for the proposed are analyzed, and the echo is converted as an time-frequency graph with short-time Fourier transform (STFT), so that an estimation technique based on LRCN (a hybrid of convolutional neural network and recurrent neural networks) is designed to estimate the SNR. Experiments indicate that the proposed method outperforms other already existing methods with better accuracy and robustness.
科研通智能强力驱动
Strongly Powered by AbleSci AI