一种混合模式的神经网络自动调制识别器
doi: 10.3724/SP.J.1146.2007.00515
An Automatic Modulation Recognizer Using Neural Networks Based on the Hybrid Mode
-
摘要: 数字信号自动调制识别(AMR)有基于决策论和统计模式两种方法,该文提出一种将两者相结合的自动调制识别系统,利用提取决策论特征向量集和统计特征向量集相结合的特征参数,使用带动量项的自适应权重的BP神经网络对MASK,MFSK,MPSK,MQAM等4类信号进行分类识别。当信噪比在0-10dB,在估计载频与实际载频相差0-100Hz的情况下正确识别率仍高达97%以上,实验证明这种分类识别方法的鲁棒性和实用性。Abstract: On automatic modulation there are two approaches, decision-theoretic and statistical pattem. An automatic modulation recognition system to recognize four digital signal classes as MASK, MFSK, MPSK, MQAM is proposed in this paper, which using decision-theoretic based feature set addition to statistical pattem based feature set with momentum auto-adapted weight BP neural network. Performance is generally good when Signal to Noise Ratios (SNR) in 0-10dB, and the estimated carrier frequency differs from the actual carrier frequency of 0-100Hz, simulations show the results even larger than 97%, that confirm the robustness and practicability of this recognition method.
-
Nandi A K and Azzouz E E. Algorithms for automaticmodulation recognition of communication signals[J]. IEEETrans. on Communications, 1998, 46(4): 431-436.[2]Azzouz E E and Nandi A K. Procedure for automaticrecognition of analogue and digital modulations [J]. IEEProceedings, Communications, 1996, 143(5): 259-266.[3]Wong M L D and Nandi A K. Automatic digital modulationrecognition using spectral and statistical features withmulti-layer perceptrons[C]. ISSPA. KualaLumpur, Malaysia,2001-08, 2: 390-393.[4]李春辉. 调制体制识别算法综述[J]. 数字通信世界, 2005,20(11): 53-56.[5]Swami Anonthram and Sadler Brain M. Hierarchical digitalmodulation classification using cumulants[J].IEEE Trans. onCommunications.2000, 48(3):416-429[6]吕铁军, 肖先赐. 基于神经网络最佳分类器通信信号的调制识别[J]. 系统工程与电子技术, 2001, 23(5): 44-46.[7]陈玉芳, 雷霖. 提高BP 网络收敛速率的又一种算法[J]. 计算机仿真, 2004, 21(11): 74-76. -
计量
- 文章访问数: 3254
- HTML全文浏览量: 80
- PDF下载量: 960
- 被引次数: 0
下载:
