4 ms·
Cool story! > Modern radars sample at 50 MHz, giving a 3m inter-sample distance. Getting that 50 MBytes/sec data down a 19.2k bits/sec channel was a challenge
by mjb 12y ago
Cool story!
> Modern radars sample at 50 MHz, giving a 3m inter-sample distance. Getting that 50 MBytes/sec data down a 19.2k bits/sec channel was a challenge
Surely you would do the processing for target extraction on the radar's side of that link, and send much less frequent samples from the estimated track over the slow link? Tracking on scanning radars can be tricky, but I suspect that would something so big and so slow the processing would be quite simple. It's also a bit surprising that ports don't do this with secondary radar (like airports do). Using the ship's signals with PCL or similar technology would definately be feasible.
Radar is a technology with great promise. The advances in computing power, signal processing, and radio technology over the last decade or so is a radar revolution waiting to happen.
- ColinWright 12y ago> Surely you would do the processing for target > extraction on the radar's side of that link, > and send much less frequent samples from the > estimated track over the slow link? That's the obvious thing to try, and it works well in ideal conditions when there are only large targets, and no clutter. Lots of people track at the radar head and only send what they believe to be the stuff that matters. We work in contexts where there can be personal water-craft in areas of moderate clutter, and plot extractors have a very poor record and even worse reputation for finding the real targets that are most at risk of being run down by vessels. We transmit the actual radar video so the operators can put the Mark I eyeball on the image. Customers like that, and we regularly out-perform our competitors at getting customers to find and identify small targets at risk.