论文部分内容阅读
文中介绍了一种基于ELMAN型神经网络的振幅移位键控(ASK)信号解调器,并研究了该解调器的相关性能。与传统解调器相比,它有一些很重要的特点:第一,神经网络算法用于解调处理,其抗干扰性能优于传统方法;第二,神经网络解调器有和传统解调器相似的处理单元,但在神经网络中,这些功能被整合在多个神经元中,无需对每个处理单元和功能进行单独设计,这些处理功能都是在其学习过程中自己获得的;第三,解调系统为并行结构,所以处理速度比传统速度更快。由MATLAB的仿真结果可以看到,该解调器有效且性能优良。
In this paper, an amplitude shift keying (ASK) signal demodulator based on ELMAN neural network is introduced and the performance of the demodulator is studied. Compared with the traditional demodulator, it has some important features: First, the neural network algorithm is used for demodulation, and its anti-interference performance is superior to the traditional method. Second, the neural network demodulator has the advantages of traditional demodulation In neural networks, however, these functions are integrated in multiple neurons without the need for a separate design of each processing unit and function, all of which are themselves acquired during their learning. Third, the demodulation system for the parallel structure, so the processing speed faster than the traditional. It can be seen from the simulation results of MATLAB that the demodulator is effective and excellent in performance.