Research on Automatic Decoding Algorithm of Morse Telegraph

In the field of modern military communications, shortwave telegraphy is an important means of emergency communication in harsh conditions. But the existing automatic decoding method did not reach the practical level, so manual decoding is still used. To improve the decoding accuracy, this paper proposes a Morse automatic decoding algorithm based on time-frequency analysis theory that combined with digital image processing and machine learning. The manual decoding method is to determine the signal frequency at the same time to determine the duration of the signal, so the time-frequency analysis method is very suitable for the analysis of Morse telegraph. In the problem the recognition of dot, dash and intervals, the clustering algorithm is used to make the intelligent decision about which kind of one element belongs to. To converse code to text, the search algorithm based on binary tree is used to improve the efficiency of table search. The simulation results show that the proposed algorithm can restore the code information from the strong noise environment, and it shows strong adaptability to the manual code and robustness in the time-varying noise environment.