I know that radio spectrum is limited and there may be environments where coexistence is impossible with many wireless networks. What I want is to optimize the configuration of WLCs to offer the best possible performance

in environments with other networks and interferences, and wireless connection is the fastest as possible.

I know that everything works in 802.11a, but I can't change the clients to operate under this standard.

I enclose a document which is an output of the command show ap auto-rf 802.11b of an access point with a normal operation without interference (although the profile of interference appears as "failed"), and three graphs that reflect the operation of various access points with interference problems.

I think this could be improved by varying some parameter because I think that is possible that clients receive higher levels of interference that the APs, and the APs do not adjust their radio interfaces for optimal performance when working with different values that exist in the client area.

All parameters are configured as you say (avoid noise and interference) Therefore I believe that varying the parameter TX power treshold of -70 to -60 could improve better performance. What do you think about?

With the information I have enclosed of environments with interference, do you think that at least some access points

should have changed the channel? The transmit power varies from time to time, but the channel of the APs have not

changed for many months, despite having configured with high

sensitivity.

Your APs are already too hot; making them hotter will just reduce your capacity.

Here's what you have now:

Nearby APs

AP :00 slot 0.................. -79 dBm on 6 (10.)

AP :b0 slot 0.................. -56 dBm on 1 (10.)

AP :b0 slot 0.................. -94 dBm on 6 (10.)

AP :c0 slot 0.................. -64 dBm on 6 (10.)

AP :60 slot 0.................. -65 dBm on 1 (10.)

AP :e0 slot 0.................. -63 dBm on 11 (10.)

AP :20 slot 0.................. -93 dBm on 11 (10.)

AP :90 slot 0.................. -85 dBm on 11 (10.)

AP :60 slot 0.................. -93 dBm on 11 (10.)

AP :50 slot 0.................. -59 dBm on 1 (10.)

AP :20 slot 0.................. -95 dBm on 11 (10.)

AP :00 slot 0.................. -55 dBm on 11 (10.)

AP :e0 slot 0.................. -71 dBm on 6 (10.)

So for example this AP is on channel 6, plus there's other APs at -64, -71 and -79 also on 6. All of the transmissions in each of these BSSIDs will interfere with the other BSSIDs.

Ideally, in each location in your coverage area, you would like to see exactly one AP on each of {1, 6, 11} at say -65 or hotter, and no other APs above say -85.

So if anything, I would reduce the TPC threshold, say to -74.

Currently I have configured the parameter TX power treshold in -70. How is it possible that the APs are heard each other with such force? Is not properly adjusting the algorithm TPC?

How will affect the changes you propose in an environment with high interference (which is what I really need), if the APs can hear each other more strongly that clients hear the APs?

This amendment would prevent interference between my own APs, but what would happen to the interference caused by rogue APs?

What can be causing that the channel of the APs never changes, even though there are such serious interferences that prevent the connection to the wireless network?

Thanks.

> Currently I have configured the parameter TX power treshold in -70.

> How is it possible that the APs are heard each other with such force?

> Is not properly adjusting the algorithm TPC?

You are right - with the threshold set for -70, and the APs hearing each other at this power level, they should have set their power down. Perhaps Coverage Hole Detection and Mitigation (CHDM) has kicked in and has boosted the APs' power to 1 - in your environment, CHDM should be turned off (set "Min Failed Client Count" to max.)

> How will affect the changes you propose in an environment with high

> interference (which is what I really need), if the APs can hear each other

> more strongly that clients hear the APs?

TPC does assume that the physical distribution of APs is similar to the physical distribution of clients. If the APs can hear each other hotter than the clients can hear the APs ... then that model may not apply. Regardless, your AP-to-AP co-channel interference will reduce your aggregate capacity.

> This amendment would prevent interference between my own APs, but

> what would happen to the interference caused by rogue APs?

Well, there is nothing you can really do about that, unless you have physical hegemony over your territory. That's life with unlicensed wireless.

> What can be causing that the channel of the APs never changes, even though

> there are such serious interferences that prevent the connection to the

> wireless network?

One would have to look at the details (not that I'm volunteering - at this point, this discussion is beyond the scope of a NetPro thread.) You can see a description of how Dynamic Channel Assignment (DCA) worked in 4.2 at http://www.cisco.com/en/US/tech/tk722/tk809/technologies_tech_note09186a008072c759.shtml#dcaa . DCA was extensively overhauled in 6.0, so if you're still running older WLC code, it is possible that an upgrade could yield better channel selection.

I changed the TX power treshold value -70 to -75 on all WLCs to avoid the interference between neighboring APs, and

although there have been several changes in power in some APs, the received power between APs is still very high.

I attach a document outlining the change.

Currently there is only 5 clients asociated around the compound,so is therefore not possible that these signal

levels as highare due to adjustments resulting from CHDM. Nor have I received alarms of coverage holes.

To continue the process of optimizing WLCs for offer the best performance possible in an environment with interferences, I think I need to know because the WLCs TPC works in a manner

that is not consistent or follows the guidelines of Cisco.

Can be affecting some other factor in this setting not occur properly? Could it be the same cause of channels to the APs never changes or changes in an unpredictable manner and illogical?

Thanks

The results attachments correspond to an AP belonging to the centralized system that has set up a TX power of -75 treshold.

I do not understand why there is interference on channel 11 and yet the AP chooses this channel to work. This also happens with other APs, which tend to choose the channel with more interference to work.

Alfonso,

I admit that, on the face of it, it's not clear why this AP chose channel 11 over the others. Factors should include interference from other APs in our RF group; interference from "rogue" APs; and non-802.11 noise.

I will note that DCA (Dynamic Channel Assigment) has been heavily reworked for 6.0. We haven't yet documented the 6.0 algorithm, but you can watch what's going on with:

debug airewave-director channel enable

Best,

Aaron

Using the debug command I see that the APs scan all channels in 802.11a, but only the DCA channels in 802.11b/g, so the algorithm will not consider rogue APs or interfering in other channels, although it is configured to scan country channels.

The debug command output shows that for single channel choice takes only the level of signal from other APs received in the DCA channels, and ignored other parameters such as interference or noise, although the WLCs have been configured for this. I don't see any output related to other parameters.

(Cisco Controller) >show advanced 802.11b channel

Automatic Channel Assignment

Channel Assignment Mode........................ AUTO

Channel Update Interval........................ 600 seconds

Anchor time (Hour of the day).................. 0

Channel Update Contribution.................... SNI.

Channel Assignment Leader...................... 00:1c:58:

Last Run....................................... 263 seconds ago

DCA Sensitivity Level: ...................... MEDIUM (15 dB)

Channel Energy Levels

Minimum...................................... -77 dBm

Average...................................... -65 dBm

Maximum...................................... -60 dBm

Channel Dwell Times

Minimum...................................... 7 days, 15 h 24 m 55 s

Average...................................... 53 days, 02 h 39 m 03 s

Maximum...................................... 64 days, 22 h 46 m 24 s

Auto-RF Allowed Channel List................... 1,6,11

Auto-RF Unused Channel List.................... 2,3,4,5,7,8,9,10,12,13

(Cisco Controller) >debug airewave director channel enable

Thu Nov 5 11:32:45 2009: Airewave Director: chan count = 13 from AP 00:1c:b1:(0)

Thu Nov 5 11:32:45 2009: 1 2 3 4

Thu Nov 5 11:32:45 2009: 5 6 7 8

Thu Nov 5 11:32:45 2009: 9 10 11 12

Thu Nov 5 11:32:45 2009: 13 0 0 0

Thu Nov 5 11:32:49 2009: Airewave Director: Computing channel assignment for 802.11bg

Thu Nov 5 11:32:49 2009: Airewave Director: Checking quality of current assignment for 802.11bg

Thu Nov 5 11:32:49 2009: Airewave Director: 802.11bg AP 00:1C:B1:(0) ch 11 (before -79.68, after -128.00)

Thu Nov 5 11:32:49 2009: Airewave Director: 00:1C:B1:(0)( 1, -65.35)( 6, -63.46)( 11, -79.68)

Thu Nov 5 11:32:52 2009: Airewave Director: Assignment is not significantly better on 802.11bg -- discarding assignment

Thu Nov 5 11:34:56 2009: Airewave Director: chan count = 19 from AP 00:1d:71:(1)

Thu Nov 5 11:34:56 2009: 36 40 44 48

Thu Nov 5 11:34:56 2009: 52 56 60 64

Thu Nov 5 11:34:56 2009: 100 104 108 112

Thu Nov 5 11:34:56 2009: 116 120 124 128

Thu Nov 5 11:34:56 2009: 132 136 140 0

Thu Nov 5 11:34:58 2009: Airewave Director: Computing channel assignment for 802.11a

Thu Nov 5 11:34:58 2009: Airewave Director: Checking quality of current assignment for 802.11a

Thu Nov 5 11:34:58 2009: Airewave Director: 802.11a AP 00:1C:B1:07:41:20(1) ch 116 (before -86.87, after -128.00)

Thu Nov 5 11:34:58 2009: Airewave Director: 00:1C:B1:(1)( 36, -66.91)( 40, -66.87)( 44, -66.87)( 48, -66.87)

Thu Nov 5 11:34:58 2009: Airewave Director: 00:1C:B1:(1)( 52, -66.87)( 56, -66.87)( 60, -66.87)( 64, -66.91)

Thu Nov 5 11:34:58 2009: Airewave Director: 00:1C:B1:(1)(100, -66.91)(104, -66.87)(108, -66.87)(112, -66.87)

Thu Nov 5 11:34:58 2009: Airewave Director: 00:1C:B1:(1)(116, -86.87)(120, -66.87)(124, -66.87)(128, -66.87)

Thu Nov 5 11:34:58 2009: Airewave Director: 00:1C:B1:(1)(132, -66.87)(136, -66.87)(140, -66.91)