PER on different protocols

[gianessa]

Hi!
I am new to the forum. Got to it while researching tools for Packet Error Rate measurements. Came accross a BER posting and it looked like a good place to ask my question. This is my first time testing above the physical layer and it is confusing me a bit. I very much appreciate any pointers.

From what I understand so far, most instruments will perform comparisons assuming that what goes in is formatted the same way as what goes out. But what if I have a receiver that decodes a signal and re-frames the received data into a different frame format (or protocol)? Any ideas on how to perform the data packet error rate measurement for this? My output interface would be an Ethernet LAN, while I could make the input interface be either ASI or a PC-like device on the Ethernet LAN.

 

[VEBill]

Convert it back and divide by two?

 

[sreid]

Just one thought that might be helpful. If there is a CRC error in the packet this is reported and normally results in a retry. For diagnostic purposes I know our software guys can count packets and errors. We use this as a diagnostic to check if this is the source of some problem. I wiould think you could do the same thing to determine Packet Error Rate.

 

[gianessa]

Thanks guys. I just realized that I did not do to well at explicitly stating the intent. I was hoping to find a real-time measurement method for system level test. My data is going into a transmitting sub-system, over RF, and back into the receiving sub-system. I am looking to characterize the whole system performance. The catch this time is that I do not have a probing point in between layers. I know I can capture the received traffic and perform post-analysis to calculate the PER. I was curious if anybody had something different.

The CRC check is indeed part of the sub-system diagnostics. Thanks for your post sreid.