Description: 现有数字信号自动调制识别方法大多只适用于无记忆信号，如PSK、ASK、FSK信号等。将有记忆
信号(MSK信号)和无记忆信号一起考虑，提出了一种改进的数字信号自动识别方法。该方法采用信号的瞬时统
计量作为特征参数，采用多层神经网络作为分类器。计算机仿真表明：当噪声采用高斯白噪声，并且信噪比大于
l5 dB时，识别率高于96％ ；当信噪比不低于l0 dB时，识别率不低于90％。-Existing digital signal automatic modulation recognition methods are mostly applied only to memoryless signals, such as PSK, ASK, FSK signals. Will have a memory signal (MSK signal) and non-memory signal into consideration, an improved digital signal automatic identification method. This method is the use of the instantaneous signal statistics as a characteristic parameter, the use of multi-layer neural network as classifier. Computer simulation shows that: When the noise using Gaussian white noise, and signal to noise ratio greater than l5 dB, the recognition rate is higher than 96 when the signal to noise ratio not less than l0 dB, the recognition rate of not less than 90. Platform: |
Size: 185344 |
Author:happyuan |
Hits:
Description: This a very good paper on the implementation of modulation classifier for SDR-This is a very good paper on the implementation of modulation classifier for SDR Platform: |
Size: 525312 |
Author:canbruce luwang |
Hits:
Description: 该文提出一种基于广义分数阶傅里叶变换和分数低阶Wigner-Ville 分布的数字调制识别新方法，该方法提取广义分数阶傅里叶变换的零
中心归一化瞬时幅度谱密度的最大值和分数低阶Wigner-Ville 分布幅度的最大值作为识别特征参数，并采用判决树分类器，实现了非高斯噪声下数字调制信号识别。-This paper presents a generalized fractional Fourier transform and fractional lower order Wigner-Ville distribution of new digital modulation recognition method, the method to extract generalized fractional Fourier transform, the center is one of the instantaneous amplitude spectral density The maximum amplitude values and fractional lower order Wigner-Ville distribution as an identifying characteristic parameters, and decision tree classifier, to achieve a non-Gaussian noise under digital modulation signal recognition. Platform: |
Size: 217088 |
Author:forevr |
Hits:
Search - modulation classifier