[hpsdr] Questions about Ethernet and Bandwidth

[Timothy S. Powell]

Jeremy, 
Thank you, your answers have help me understand a little better. I appreciate it. 


Timothy




On Aug 24, 2011, at 9:33 PM, Jeremy McDermond <mcdermj at xenotropic.com> wrote:

> On Aug 24, 2011, at 6:00 PM, Timothy Powell wrote:
>> Here is what I think I understand, please correct me where I’m wrong, thanks.  After reading the website about Mercury it’s my understanding that the ADC can sample up to 55 MHz of radio spectrum, more if you use under sampling which I don’t understand yet.  That information is sent to the FPGA which is used as a digital down convert that takes the real signal and converts it to a complex (I,Q) baseband signal centered at zero frequency (yes, no) and it also lowers the sampling rate (yes, no).  Then this data can be sent to a Metis board that sends it over the network to the PC and then the PC does its work.
> 
> That's pretty much correct.
> 
>> Some questions I have are:
>> 
>> Knowing that data sent over a network are wrapped in protocols like UDP/IP which uses bits, what is the maximum amount of sample data that can be sent to the computer using a 1Gbs network connection assuming a Metis board is connected directly to a computer?  How much of the radio spectrum (bandwidth) would that be?  How could it be calculated or estimated?
> 
> Each packet coming out of Metis has an IP header and a UDP header.  That's 8 octets for the UDP header and 12 octets IP header.  Ethernet frame overhead is 38 octets.  Metis also has a packet header of 8 octets.  Each Metis packet holds 2x512 byte HPSDR frames.  That means that each packet is 1090 octets -- that's 1720 bits.  That means that you can fit 581,395 Metis packets per second down the pipe.  Each one of those packets holds 126 I/Q samples.  That's 73,255,770 samples per second.  Nyquist says that we should have about twice the sample rate of the highest frequency we want to represent.  That means that you'll get about half the bandwidth as your sample rate.  That means that you'll be able to see about 36.627 MHz.
> 
>> Would a current computer be able to handle that amount of data?
> 
> I have no clue.  There are experiments being done with GPGPU methodologies to try to process things faster.  It also depends on what you want to do with the data once you get it there.
> 
>> What is the current maximum amount of data sent to a computer from Mercury?
> 
> The protocol allows for 8 firmware receivers each at 192kHz sampling rate.
> 
>> I’m wondering what the limitations are for the amount of data Mercury could send to a computer and how much of that data the computer could handle.
> 
> I think you'd probably run into processing power hurdles before you saturated the datarate.
> 
>> Thanks
>> Timothy - KD4IKY
> 
> --
> Jeremy McDermond (NH6Z)
> Xenotropic Systems
> mcdermj at xenotropic.com
> 
> 
> 

 1314294983.0