[orbit-user] Follow up on: RSSI per received packet using Atheros

[Haris Kremo]

Hi,

As Kishore pointed out, the card must be in monitor mode.

If this is easier and faster for you:

Luis made image baseline-8.3-dev.ndz with radiotap enabled for both
ipw2200 and madwifi.

other than

tcpdump -i <interface>

tshark is useful as well

tshark -V -i eth2

decodes complete headers bit by bit

RSSI is given in dBm, this is correct assuming the conversion from
RSSI to dBm is known and accurate for a particular card

Hope it helps,

H.

On 6/19/08, Rodney Berriman <rodney.berriman at nicta.com.au> wrote:
>
>  oops, found the prism2 header, in net80211/ieee80211_monitor.h
>
>  Still don't know why it doesn't appear in tcpdump.
>
>
>
> Rodney
> thanks for your replies Kishore and Gautam.
>
>
>  We have tried to catch the RSSI in Prism2 header, having first ensured:
>  /proc/sys/net/ath0/dev_type = 802     as per
> http://madwifi.org/wiki/UserDocs/MonitorModeInterface
>
>   then run command:
>      tcpdump -i ath0 -vvv -s 0 -y PRISM_HEADER -e
>
>   Typical result for a ping request packet is:
>
>  23:10:44.580785 314us DA:00:0b:6b:0a:83:78 SA:00:0b:6b:0a:83:21
> BSSID:02:0b:6b:0a:83:7e LLC, dsap SNAP (0xaa), ssap SNAP (0xaa), cmd 0x03,
> IP (tos 0x0, ttl  64, id 0, offset 0, flags [DF], length: 84)
> 192.168.200.109 > 192.168.200.126: icmp 64: echo request seq 5379
>
>  tcpdump version info:
>     tcpdump version 3.8.3
>     libpcap version 0.9.4
>
>  These are quite up to date, but I'm wondering if libpcap has the necessary
> decoder for all the fields of prism2?
>
>  There was no warning about libpcap being unable to decode, such as does
> appear if I change the format to  804 "Atheros descriptors".
>
>
>  Now looking through the madwifi source code to see if RSSI is hiding in the
> binary format.
>  Not seeing any reference to Prism2 header in the code, but maybe IEEE802_11
> RX stats is relevant.
>  Is prism2 an extension of IEEE802_11 header?
>
>  /* not all hardware provides this information for frames received */  (that
> might be significant!)
>  struct ieee80211_rx_stats {
>    u32 mac_time;
>    s8 rssi;   /* what we want */
>    ...
>
>  Any pointers most welcome,
>
>  Rodney
>
>
>
>
>
> Networks Systems Group NICTA, Sydney
>
>
>
> _______________________________________________
>  orbit-user mailing list
>  orbit-user at orbit-lab.org
>  http://orbit-lab.org/cgi-bin/mailman/listinfo/orbit-user
>
>