HUGELY erratic dBm readings in 2.4 GHz?

[jphughan]

UPDATE: This seems to have been fixed in the latest official firmware as of this update on 3/25/13.

Hello all,

I just got an RT-N66U thanks in large part to information on this forum. I spent a while today getting the firmware updated to 3.0.0.4.260 and dialing in all of the settings to what I thought would make sense (sure wish Asus had better documentation on the various settings....), and then settled down to enjoy some DLNA streaming from my laptop (using a wired connection) to my Oppo BDP-93 Blu-ray player using its USB Wifi-N 2.4 GHz adapter. I was streaming SD video at the time.

Every 15 minutes or so, the video would stutter severely for about 30 seconds before carrying on. This didn't happen with my previous Netgear WNR3500 router (2x2, 2.4 GHz only). So I opened up inSSIDer and saw ENORMOUS fluctuations on this router's 2.4 GHz band, from -15 dBm all the way down to -80 dBm several times per minute -- far more severe than even my DLNA experience would have suggested. I played with all sorts of settings -- specifying a channel manually, channel bandwidth options, b/g protection, tx bursting, and enhanced interference management -- and no changes had any effect one way or the other. This behavior persisted even after resetting the router to factory defaults.

I've attached a chart below showing 5 minutes' worth of readings. Measurements were taken from my laptop's Intel Wifi 5300 card positioned 5 feet from the N66U and 10 feet in the other direction from my previous Netgear router that I fired back up for comparison. Unfortunately the chart got heavily compressed when uploaded here, but the bold darker blue line around -30 dBm is my previous Netgear router, the aqua line around -45 dBm that's almost completely flat is the N66U's 5 GHz band, and and the crazy yellow line is the 2.4 GHz band; the others are random nearby networks I monitored for comparison. As you can see, every network on the chart remains within a fairly tight amplitude range except this router's 2.4 GHz band, which is all over the map!

Can anyone replicate this monitoring their RT-N66U with inSSIDer? Is this specific to certain firmware revisions, or is there some ideal settings combination that makes this disappear? I haven't yet switched to older firmware (hoping not to have to do that), but I've played with all sorts of settings combinations to no avail. At this point I'm wondering if I may have gotten a defective unit. If anyone else can do a quick check with inSSIDer on their N66U's 2.4 GHz band under similar test conditions I'd be most appreciative.

 

[TonyH]

Hi,
From your inSSIDer, are there many 2.4GHz WiFi users in your neighborhood? Did you select a channel to minimize the interference? If your microwave oven is nearby it will affect 2,4GHz signal when the oven is running. I only run 5GHz radio to avoid interference in my house.

 

[jphughan]

Hi,
From your inSSIDer, are there many 2.4GHz WiFi users in your neighborhood? Did you select a channel to minimize the interference? If your microwave oven is nearby it will affect 2,4GHz signal when the oven is running. I only run 5GHz radio to avoid interference in my house.

I can see 30 2.4 GHz networks from here (fairly dense apartment complex), but Channel 1 seems to be the least congested where I am, which is the one the router chose when set to Auto. I understand that interference can affect amplitude readings, but none of the other networks I'm monitoring show anywhere near this level of fluctuation; I included only a sampling in my chart for ease of reading. And it doesn't make sense to me that this 3x3 router would produce far more erratic readings than my previous 2x2 router (not to mention more erratic readings than even some single-antenna G routers around here), especially when my laptop is only a few feet away.

No cordless phones or active microwaves, and I can't imagine any significant Bluetooth activity nearby.

 

[nubyOz]

I can see 30 2.4 GHz networks from here (fairly dense apartment complex), but Channel 1 seems to be the least congested where I am, which is the one the router chose when set to Auto. I understand that interference can affect amplitude readings, but none of the other networks I'm monitoring show anywhere near this level of fluctuation; I included only a sampling in my chart for ease of reading. And it doesn't make sense to me that this 3x3 router would produce far more erratic readings than my previous 2x2 router (not to mention more erratic readings than even some single-antenna G routers around here), especially when my laptop is only a few feet away.

No cordless phones or active microwaves, and I can't imagine any significant Bluetooth activity nearby.

Just along shot, are you running single or dual channel?
If dual, try single channel. also trial dial down your power settings on the router, found often routers are poorly cooled, a slight manufacturing defect and they oeverheat quickly.

 

[jphughan]

I'm running single channel in 2.4 GHz and 20/40 MHz in 5 GHz, which goes immediately to 40 since there's only one other 5 GHz network here. I tried switching the 2.4 GHz band to 40 MHz and 20/40 MHz and nothing changed with respect to these readings.

I'll try dialing down my power settings tomorrow morning, but part of the reason I bought this router was for the incredible range since I'll be moving soon, so if that ends up being the fix I'll be rather disappointed -- especially since I'm running the stock setting now.

Looking more closely at the surrounding networks, I can see 3 other networks that occasionally drop as low as mine. They never get as high as mine does, presumably since they're farther away, but they also don't drop that low nearly as frequently -- and that's still only 3 of the 30 other networks near me exhibiting this behavior (one Cisco/Linksys, one Belkin, and one Netgear).

EDIT: That transmit power suggestion got me thinking. The admin page mentions that the selected power setting will still be adjusted to comply with regulations. Is it possible that this router is adjusting its transmit power so frequently across such a wide range in response to surrounding networks? Seems a bit excessive.

 

[dmillerzx]

Hi there, I am in an apartment with a crowed amount of APs in my apartment complex. Not a single channel is interference free. I am experiencing this same exact issue! I've been looking for days now on anyway I can troubleshoot this and perhaps resolve the signal dips in the 2.4ghz spectrum.

I'm trying to determine how likely this is a problem due to interference or if it's a problem with my specific router as it's a refurb.

I've done similar test as the OP and notice that many other SSIDs are showing similar dropping behavior; however not all of them. I have been playing with the settings extensively without any luck.

Desperately looking for help on this.

 

[jphughan]

Quick follow-up to this. I'm convinced that this is either a) an inherent behavior of the router and likely N routers in general that appears exaggerated on the InSSIDer graphs because I'm so close to my router and it's so much more powerful than most other routers on the market, or b) something that this router model simply does when in crowded 2.4 GHz areas. I don't have access to another RT-N66U to test Theory A, though finally hearing from someone else seeing the same behavior, I'm more confident that I do NOT have a defective unit. And since I'm not likely to find an area that DOESN'T have lots of 2.4 GHz interference, I don't know whether I can test Theory B.

But not being able to test Theory B doesn't really matter to me because first of all, I'd want this router to behave well even in congested environments (where I'm likely to be most of the time anyway), and second of all, it DOES seem to be behaving just fine even with the signal dips. Using LAN Speed Test (client on 2.4 GHz Wifi, server on Ethernet), I don't notice any occasions of substantial throughput drops, even when monitoring the signal strength from an alternate machine. Note that keeping InSSIDer open while running bandwidth tests from the same machine will cause horrendous Wifi throughput because InSSIDer has the Wifi adapter doing a ton of extra monitoring work, which leaves less available resources for actual data transmission. Therefore I'm thinking that these signal dips are part of the "back off" behavior specified in the 802.11n spec that WAPs are required to do in order to monitor the radio environment around them to make decisions like channel selection, whether to allow fat channels, etc. FWIW I still see those dips even when configuring the router on a fixed channel and setting it to 20 MHz only, but again it doesn't bother me since it doesn't seem to correlate with any ACTUAL performance degradation.

The only reason I was led to believe the InSSIDer dips might be problematic was because I was seeing poor performance streaming media to my Blu-ray player over Wifi -- but that has turned out to be a bug in the Wifi chipset driver for the router, specifically how it interfaces with certain Wifi chipsets in clients (including the Wifi chipset on the Xbox 360 S, but not the one in the Xbox 360 external Wifi adapter). Broadcom has apparently finally delivered a fix to ASUS, which they are testing and are expected to incorporate into a future firmware release.

 

[dmillerzx]

Thinking about giving this a try.

http://www.amazon.com/gp/product/B005UBNJ48/?tag=snbforums-20

Thanks for the feedback.

 

[jphughan]

Thinking about giving this a try.

http://www.amazon.com/gp/product/B005UBNJ48/?tag=snbforums-20

Thanks for the feedback.

No experience with it, but I've heard good things about the Mohu Bounce, which seems to be a similar concept.

 

[dmillerzx]

Not that this is effecting the end results or anything; however, are you using interference mitigation? I notice the same drops with or without... Currently I leave it on.

 

[jphughan]

Not that this is effecting the end results or anything; however, are you using interference mitigation? I notice the same drops with or without... Currently I leave it on.

I have Enhanced Interference Management enabled. I figured if there were ever a place to have it enabled, it would be in an apartment complex where each resident has their own Wifi router. But if memory serves, I saw the same dips with or without it enabled.

 

[dmillerzx]

Are you using the stock transmit power values of 80? I've done some experimenting without any difference.

 

[jphughan]

Are you using the stock transmit power values of 80? I've done some experimenting without any difference.

Yes, didn't think dropping below stock was necessary, and running above can generate excess heat and accelerate hardware failure -- plus signal strength isn't an issue for me.

 

[bluepoint]

I can see 30 2.4 GHz networks from here (fairly dense apartment complex), but Channel 1 seems to be the least congested where I am, which is the one the router chose when set to Auto. I understand that interference can affect amplitude readings, but none of the other networks I'm monitoring show anywhere near this level of fluctuation; I included only a sampling in my chart for ease of reading. And it doesn't make sense to me that this 3x3 router would produce far more erratic readings than my previous 2x2 router (not to mention more erratic readings than even some single-antenna G routers around here), especially when my laptop is only a few feet away.

No cordless phones or active microwaves, and I can't imagine any significant Bluetooth activity nearby.

I am on the same situation as yours. At peak time I see about 52 wireless stations around me and it's really bad with the data speed when I let the wireless in auto mode. Try all channels statically and find where the fastest speed is. By trial and error I found channel 8 the best for my area, I know there are overlapping channels where 8 is but it gives me the fastest speed.
My settings for 2.4 GHz

Wireless Mode: auto b/g protection checked
Control Channel: 8 (find the best for you)
Chan Bandwidth: 20 MHz
Authentication: WPA2-Personal
Encryption: AES

From Wireless 2.4 GHz/Professional

Disable "Enhanced Interference Management"

Note: My N66U is unstable when using the original Firmware 246,260 so I'd
recommend to use original 220 or 270 or Merlin's 220's or his latest
270.24(I like the best) firmwares.

 

[dmillerzx]

I am still noticing the dips in 2.4 dbm even with buying the new antenna I mentioned a few posts back. This doesn't appear to hinder performance; however, it is a bit worrying. I have however had some luck (so far) with resolving connection drops on some clients using the 2.4ghz network.

Using 20/mhz width vs 20/40 and disabling Interference Mitigation has resolved connection drops so far in my environment.

 

[gluon]

From the ddwrt forums, a user confirm that changing beacon interval variable to higher one will reduce this interferences, most of the AP will using 100 by default, try to increase it to 101 if such option is exist in merlin or vanilla asus.

 

[dmillerzx]

Any chance you can link to the forum where you saw this being discussed? I can't understand how a slightly higher beacon rate would help with the signal dbm drop. Willing to give anything a try though

Thanks!

 

[dmillerzx]

I would think it's the other way around. Shouldn't lowering the beacon rate help instead of increasing it??

 

[dmillerzx]

Maybe the new wireless drivers in the leaked beta will resolved this!? Fingers crossed

 

[Poseidon]

Thinking about giving this a try.

http://www.amazon.com/gp/product/B005UBNJ48/?tag=snbforums-20

Thanks for the feedback.

Anyone know where to purchase this from? Amazon is back ordered 1-3 weeks and I can't find it anywhere else on the net.