Machine Learning System Uses Images To Teach Itself Morse Code

Conventional wisdom holds that the best way to learn a new language is immersion: just throw someone into a situation where they have no choice, and they’ll learn by context. Militaries use immersion language instruction, as do diplomats and journalists, and apparently computers can now use it to teach themselves Morse code.

The blog entry by the delightfully callsigned [Mauri Niininen (AG1LE)] reads like a scientific paper, with good reason: [Mauri] really seems to know a thing or two about machine learning. His method uses curated training data to build a model, namely Morse snippets and their translations, as is the usual approach with such systems. But things take an unexpected turn right from the start, as [Mauri] uses a Tensorflow handwriting recognition implementation to train his model.

Using a few lines of Python, he converts short, known snippets of Morse to a grayscale image that looks a little like a barcode, with the light areas being the dits and dahs and the dark bars being silence. The first training run only resulted in about 36% accuracy, but a subsequent run with shorter snippets ended up being 99.5% accurate. The model was also able to pull Morse out of a signal with -6 dB signal-to-noise ratio, even though it had been trained with a much cleaner signal.

Other Morse decoders use lookup tables to convert sound to text, but it’s important to note that this one doesn’t. By comparing patterns to labels in the training data, it inferred what the characters mean, and essentially taught itself Morse code in about an hour. We find that fascinating, and wonder what other applications this would be good for.

Thanks to [Gordon Shephard] for the tip.

Zijn we onlangs – zonder het te beseffen – een nieuwe zonnecyclus ingerold?

The RFI Hunter: Looking For Noise In All The Wrong Places

Next time you get a new device and excitedly unwrap its little poly-wrapped power supply, remember this: for every switch-mode power supply you plug in, an amateur radio operator sheds a tear. A noisy, broadband, harmonic-laden tear.

The degree to which this fact disturbs you very much depends upon which side of the mic you’re on, but radio-frequency interference, or RFI, is something we should all at least be aware of. [Josh (KI6NAZ)] is keenly aware of RFI in his ham shack, but rather than curse the ever-rising noise floor he’s come up with some helpful tips for hunting down and eliminating it – or at least reducing its impact.

Attacking the problem begins with locating the sources of RFI, for which [Josh] used the classic “one-circuit-at-a-time” approach – kill every breaker in the panel and monitor the noise floor while flipping each breaker back on. This should at least give you a rough idea of where the offending devices are in your house. From there, [Josh] used a small shortwave receiver to locate problem areas, like the refrigerator, the clothes dryer, and his shack PC. The family flat-screen TV proved to be quite noisy too. Remediation techniques include wrapping every power cord and cable around toroids or clamping ferrite cores around them, both on the offending devices and in the shack. He even went so far as to add a line filter to the dryer to clamp down on its unwanted interference.

Judging by his waterfall displays, [Josh]’s efforts paid off, bringing his noise floor down from S5 to S1 or so. It’s too bad he had to take matters into his own hands – it’s not like the FCC and other spectrum watchdogs don’t know there’s a problem, after all.

Mogelijk gemaakt door Tempera & WordPress.