High frequency counter

[Altera_Forum]

I am new on FPGA, then sorry for the little knowledge. 

 In my application I have to measure the transit-time of an ultrasonic signal, for that I will develop a counter, but I need the frequency as high as possible.   The ideal counter clock would be something around 1 GHz. Is a FPGA the best solution for this case? I was reading others posts related to this subject, and it seems that this clock rate is not reachable on a FPGA.  Checking the datasheet of Cyclone V, for example, I got the following information:  "PLL Specifications for Cyclone V Devices - PLL VCO operating range: 600~1600 MHz - Output frequency for internal global or regional clock: 550 MHz"  So, is it possible to use this 550 MHz as a clock for my counter? It's not 1 GHz, but would be something to start with...  Any other better solution?  Thank you, Yves

[Altera_Forum]

IIRC there was a post similar to this a while ago. You cannot have a single counter in the GHZ frequency, but I think the proposed solution was to use several counters on different phases of the same clock. I cant find the post at the moment, maybe someone else can...

[Altera_Forum]

Here are some links: 

http://www.alteraforum.com/forum/showthread.php?t=28472 (http://www.alteraforum.com/forum/showthread.php?t=28472) This one:http://www.alteraforum.com/forum/showthread.php?t=24720 (http://www.alteraforum.com/forum/showthread.php?t=24720) hints on using the serdes. Another idea discussed on this forum was to use a GX-receiver in basic mode as a sort of 'logic analyser'. That would go up to 3.25 GHz. How do you form your input pulse? How large are these transition times? I assume you have a comparator or something to detect the transition-time.  You may also want to take a look at time to digital convertors (http://en.wikipedia.org/wiki/time-to-digital_converter)

[Altera_Forum]

--- Quote Start ---   In my application I have to measure the transit-time of an ultrasonic signal   --- Quote End ---   Before jumping into implementing a counter to measure the round-trip distance, have you considered other options?  Radar and sonar systems transmit wide bandwidth signals as either chirps (swept frequencies) or pseudo-noise codes, and then cross-correlate the echos with the transmit code (this is called pulse compression). This results in an echo return with an impulse-like shape. You can detect the round trip distance via detecting that pulse, or you can Fourier transform and measure the phase slope.  FMCW radars (like the type the cops use to catch you speeding) transmit a chirp, mix the echo with the transmit signal, and the mixed signal can be used to determine the distance and speed of the target (something like that anyway ...)  Cheers, Dave