| This paper discussed the method of automatic modulation recognition of digital communication signals. It has decision theory method and statistical pattern method, The decision theory method, analyzing the statistics characteristic of signals, because it has considered the influence of noise jamming, has a better performance under the low SNR. But this kind of method usually aims at some kind of concrete modulation signals'statistical characteristic which the analysis obtains; therefore the recognition scope is narrow. The statistical pattern recognition method usually extract features based on the non-noise jamming signals, therefore, under high SNR condition, recognition performance is generally good, under low SNR condition, the recognition performance is generally bad.Neural networks compare to the tradition technology have the ability of resolving the complicated classification problem's rapidly. Addition to its fault-tolerant ability,insensitive to noise and incomplete data though its train or self-taught, let us choose neural networks to solve the problem of automatic digital modulation.In this paper we proposed an automatic modulation recognition system to recognize four digital signal classes as MASK,MFSK,MPSK,MQAM using decision-theoretic based feature set addition to statistical pattern based feature set with momentum auto-adapted weight BP neural network. In order to verify the performance of the system, we carried out large amount of emulation experiment from the neural networks structure parameter and signal treatment way. In the neural network target matrix aspect, we use the different target matrix to the different modulation signals. 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. |
PDF Full Text Request