| Thread Links | Date Links | ||||
|---|---|---|---|---|---|
| Thread Prev | Thread Next | Thread Index | Date Prev | Date Next | Date Index |
Hi Gergely, please find below the reason why I think the number is incorrect. The test is considering 800 bytes frames. To that we want to add the part of the preamble that is used to synchronize the scrambler, that is 5 bytes (worst case). This contribution is actually negligible but I’m counting it anyway in the following computation. 805 bytes total means 805 * 8 = 6440 bits. Assuming that a single bit error causes a frame to be discarded (that is, no FCS collision in the age of the universe), the chance of discarding a packet due to a bit flip is 1 - (1 - BER)^NBIT. In other words is 1 minus the chanche of NOT having bit flips. For a BER of 10^-10 that gives ~6.4 10^-7. Thank you, Piergiorgio From: Huszak Gergely <Gergely.Huszak@xxxxxxxx> Hello Piergiorgio, We have received and are processing your comment numbered as shown in the subject with the content TO avoid spending unnecessary amount of time in Vienna, would you be so kind please to share the technical background (problem, rationale and actual computation ) on this matter? An email to the reflector with these would be fine. Upon your kind reply we would go ahead and process your comment swiftly. Thank you very much in advance, yours sincerely Gergely Comment: The computation of the frame error ratio versus the BER is not correct. Suggested Remedy: Change " 7.8 x 10^-7" to "6.4 x 10^-7" To unsubscribe from the STDS-802-3-10SPE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-10SPE&A=1 To unsubscribe from the STDS-802-3-10SPE list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-10SPE&A=1 |