Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 internet speed test! ×

Submission + - Machine Learning Algorithms to Crack Morse Code (blogspot.com)

mni12 writes: "Morse code has been used since early 1840's and is still a very popular mode of communication especially among ham radio operators. While it takes some effort for humans to learn Morse code it is a very efficient way in communicating short messages over radio waves, especially under noise, interference, propagation fading or other adverse conditions. Experienced human operators can easily outperform any publicly available Morse decoding software.
I have done some experiments with machine learning algorithms, especially with Self Organizing Maps (SOM) applied to real-time decoding Morse code in real world noise & interference filled signals. Early test results look promising but I would like to turn to Slashdot community for some advice and ideas.

What kind of machine learning algorithms would be applicable for real time Morse decoder when signals contain a lot of noise, interference from other stations, fading, irregular timing and other problematic features?"

This discussion was created for logged-in users only, but now has been archived. No new comments can be posted.

Machine Learning Algorithms to Crack Morse Code

Comments Filter:

Someday somebody has got to decide whether the typewriter is the machine, or the person who operates it.

Working...