From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ali Abedi Date: Sat, 27 Oct 2012 19:50:26 -0400 Subject: [ath9k-devel] Channel busy cycles In-Reply-To: References: <5086EA9A.6030400@mailservices.uwaterloo.ca> <5086FFE3.4010108@mailservices.uwaterloo.ca> <508996FF.8060003@mailservices.uwaterloo.ca> <508C34D8.9060607@mailservices.uwaterloo.ca> Message-ID: <508C7342.7020906@mailservices.uwaterloo.ca> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ath9k-devel@lists.ath9k.org Thanks for the answer. In this experiment we got 0% frame loss, so is your reasoning is still valid for this case? And why it is almost constant? I thought maybe RX is not counted for OFDM signal extension or OFDM preamble? Does that make any sense at all? Thanks, Ali On 10/27/2012 7:42 PM, Adrian Chadd wrote: > Hi, > > Sometimes your TX isn't decoded, and is counted as noise. Hence not > showing up as RX. > > RX is only 'frames the PHY RXed'. "busy" is "time the PHY reported the > signal level was high enough to consider the air busy, and thus not > available for transmitting." > > > > Adrian > > On 27 October 2012 12:24, Ali Abedi wrote: >> Hello, >> >> I record the 4 registers showing tx, rx, busy, and total cycles. In a >> clean environment (checked with a spectrum analyzer) I expected to see >> busy = tx + rx. however, we have found that actually busy = tx + rx + >> constant. where this constant is rate dependent. For example for 54 >> Mbps, the constant is about 1000 cycles. I have to note that this is an >> 802.11g setup. Any explanation is appreciated. >> >> Best, >> Ali >> >> _______________________________________________ >> ath9k-devel mailing list >> ath9k-devel at lists.ath9k.org >> https://lists.ath9k.org/mailman/listinfo/ath9k-devel