Chun-Yeow,
Did you mean that you fix the tx rate in rc?
I did not fix the rate in rc but rather achieved MCS7 by pysically
connecting the antenna ports between two STA's.
STA A (port1) ----- 50dB atten ------ STA B (port1)
The rate control then settled on MCS7 (150Mbps) on each STA.
The stream that I
mentioned is iperf parallel stream (-P).
Iperf parallel streams:
# iperf -c 192.168.11.2 -p 7777 -u -b 150m -t 10 -i 1 -P10
------------------------------------------------------------
Client connecting to 192.168.11.2, UDP port 7777
Sending 1470 byte datagrams
UDP buffer size: 160 KByte (default)
------------------------------------------------------------
[ 4] local 192.168.11.1 port 33764 connected with 192.168.11.2 port 7777
[ 5] local 192.168.11.1 port 60989 connected with 192.168.11.2 port 7777
[ 6] local 192.168.11.1 port 43993 connected with 192.168.11.2 port 7777
[ 7] local 192.168.11.1 port 33732 connected with 192.168.11.2 port 7777
[ 8] local 192.168.11.1 port 60437 connected with 192.168.11.2 port 7777
[ 9] local 192.168.11.1 port 42379 connected with 192.168.11.2 port 7777
[ 10] local 192.168.11.1 port 47369 connected with 192.168.11.2 port 7777
[ 11] local 192.168.11.1 port 50438 connected with 192.168.11.2 port 7777
[ 3] local 192.168.11.1 port 59780 connected with 192.168.11.2 port 7777
[ 12] local 192.168.11.1 port 57701 connected with 192.168.11.2 port 7777
[ ID] Interval Transfer Bandwidth
[ 11] 0.0- 1.0 sec 8.16 MBytes 68.4 Mbits/sec
[ 11] 1.0- 2.0 sec 15.9 MBytes 133 Mbits/sec
[ 11] 2.0- 3.0 sec 15.8 MBytes 133 Mbits/sec
[ 11] 3.0- 4.0 sec 16.0 MBytes 134 Mbits/sec
[ 11] 4.0- 5.0 sec 16.0 MBytes 134 Mbits/sec
[ 11] 5.0- 6.0 sec 15.8 MBytes 133 Mbits/sec
[ 11] 6.0- 7.0 sec 16.1 MBytes 135 Mbits/sec
[ 11] 7.0- 8.0 sec 16.2 MBytes 136 Mbits/sec
[ 11] 8.0- 9.0 sec 16.2 MBytes 136 Mbits/sec
[ 11] 0.0-10.0 sec 152 MBytes 128 Mbits/sec
[ 11] Sent 108554 datagrams
[ 11] Server Report:
[ 11] 0.0-10.2 sec 82.4 MBytes 67.5 Mbits/sec 14.405 ms 49757/108552
(46%)
[ 11] 0.0-10.2 sec 1 datagrams received out-of-order
As you can see the overall throughput and datagram loss is similar to
the iperf single parallel case shown in previous posts.
Anyway, based on the rc statistic, I think that your transmission has
almost no loss using HT40/SGI MCS7.
Could you help me understand how you arrived at that conclusion from the
two rc_stats dumps? What about the difference between the sender station
(A) and the receiver station (B) rc_stats dumps? For example the
throughput on A seems to be 2x B. I am not sure what this means.
Please bear in mind that the
achievable throughput via iperf is not related to PHY rate. You
probably can get 130Mbps since iperf is L3 throughput measurement.
How do I estimate the PHY rate? Can you suggest some tools/techniques
for measurements on L2/L3?
Thanks for all of your help,
-Stephen
_______________________________________________
Devel mailing list
[email protected]
http://lists.open80211s.org/cgi-bin/mailman/listinfo/devel
