Short wave frequency band signal sorting method

Abstract

The invention discloses a short-wave frequency band signal sorting method, and relates to the field of electromagnetic spectrum sensing, and the method is based on morphological filtering of statistics and a bidirectional threshold background noise extraction algorithm, and a short-wave broadband signal extraction algorithm of multi-dimensional spectrum feature analysis, can better realize automatic extraction of signals, and is low in signal omission ratio and high in accuracy. According to the method, noise error extraction is reduced, the problems of inaccurate signal information, virtual signals and the like possibly caused by inaccurate center frequency and bandwidth are avoided, separation of signals and bottom noise is effectively realized, and the signal sorting accuracy is greatly improved.

Description

technical field [0001] The invention relates to the field of electromagnetic spectrum sensing, in particular to a method for sorting short-wave frequency band signals. Background technique [0002] Shortwave refers to radio waves with a frequency of 3 to 30 MHz. The short wave has a short wavelength, and the diffraction ability of the ground wave propagating along the surface of the earth is poor, and the effective distance of propagation is short. Short-wave signals mainly rely on ionospheric reflection (sky waves) to propagate, and can also rely on ground waves for short-distance propagation like long and medium waves. Ultrashort wave communication mainly relies on ground wave propagation and space wave line-of-sight propagation. When the communication distance is short, the whip antenna is usually used to propagate by ground wave; when the communication distance is long, an elevated antenna is used or the station is placed in a higher place to propagate by space wave; w...

AI Technical Summary

Problems solved by technology

[0003] At present, domestic short-wave signals still have the following problems when performing automatic sorting work. The background noise floor of short-wave monitoring fluctuates greatly, and statistics do not obey the Gaussian distribution problem; there are AM, CW and digital signals in the short-wave frequency band, among which AM is in the The spectrum shows multiple spectral peaks, which are easy to be extracted into multiple signals; the center frequency misalignment of the signal under the high-resolution bandwidth leads to the problem of signal repetition; in the above cases, it is easy to cause the problem of missing selection during signal sorting. At the same time, When the center frequency and bandwidth of the extracted signal are inaccurate, there are more redundant signals when sorting shortwave signals

Images