Infinium discrimination question

Hobbyist, in a PI detector, the timing of the transmitted signal is delayed and compared to the timing of the received signal. Any variation is seen as metal in the field. The timing is altered by an RLC coupled receiver and the two waveforms can be compared and analyzed providing fairly accurate discrimination. The processing power required to do this is huge and this is what makes most people say it is impossible. I will say that the waveform distortion is so precisely repeatable for each type of metal that when you do discriminate, it is nearly 100% positive.

The processor required to do this in real time is normally a 32 bit processor running at 60mhz or above with dynamic, not static or flash, RAM. Fourrier transforms and complex Calculus is required to do the analysis. Basically, you need an embedded Pentium to do it with ease. It's the processing power that makes it such a hard thing to do. There's no easy notch or anything. The exact distortion of the waveform must be analyzed bit by bit in real time.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,21119.msg142669.html#msg142669

It's being done right now in some of the more modern machines but to a lesser degree than you would expect. Processor development takes time. As much as 2 years sometimes and when they select what is a fast processor today, by the end of 6 months, is obsolete or outdated. It's like the mouse being chased by the cat who keeps grabbing just the tip of his tail but never getting the mouse. Processors tend to stay just a little ahead of practical application.

It won't be long before a really good PI with discrimination is on the market. This "dig everything" guy might have to change his tune.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,21119.msg142672.html#msg142672