AirMarshall RSSI values

[RaphaelL]

Hi ,

I'm trying to understand the values returned by AirMarshal. 

I'm seeing SSIDs near my APs with some anormal values : 

I'm no wireless expert but a RSSI of 137db seems impossible. 

I can find approx 100ish other SSIDs on multiple sites with those values. Has anyone ever seen that ?

[alemabrahao]

That’s not just unusual, it’s physically impossible in the context of Wi-Fi signal strength.

RSSI is typically measured in dBm, a logarithmic unit that expresses power relative to 1 milliwatt.
In Wi-Fi, RSSI values are negative, ranging from around -30 dBm (excellent) to -90 dBm (very poor).
A value like 137 dBm would imply a signal strength far beyond any realistic or safe transmission level, it’s likely a misinterpretation or bug.

[RaphaelL]

That's what I'm thinking but I don't think Meraki is displaying RSSI in negative. I have never seen a negative value with Airmarshal. 

https://documentation.meraki.com/MR/Monitoring_and_Reporting/Location_Analytics

RSSI - 95 = signal strength in dBm

Still 137 doesn't make sense at all. 

I can also find frequently values of 255. Something is off

[alemabrahao]

I'll be honest, I don't remember why in Meraki the value is displayed in dB. 😅

[alemabrahao]

I'm thinking here, could 137 dB represent total gain (for example, sum of antenna + amplifier gain), but not the received signal?

[alemabrahao]

I think I found the explanation.

For example, if a client device's radio receives a signal at -75 dBm, and the noise floor is -90 dBm, then the effective SNR is 15 dB. This would then reflect as a signal strength of 15 dB for this wireless connection. 

https://documentation.meraki.com/MR/Wi-Fi_Basics_and_Best_Practices/Signal-to-Noise_Ratio_(SNR)_and_...

But 137db is still unrealistic.

[RWelch]

 

Seems odd (really really high) - curious what MR firmware are you running in your network?  

And do you have RRM enabled (perhaps the AI-RRM factor is contributing in this equation - somehow)?

Asking because when I look at Air Marshall using my APs (MR 31.1.8) I too see elevated #s.  The two at the bottom are both my APs in blue.  And I do have AI-RRM enabled.

[rhbirkelund]

Without argueing whether or not something is off in AirMarshall, often people tend to confuse dB witth dBm (or dBW or any other measurement). 

Decibel, or dB are generally just a measurement of ratio, or gain as @alemabrahao also notes. When we talk about a change in 10 dB, this refers to a change in order of magnitude, while 3 dB is a change in 2's +3dB is times 2, while -3 dB is times 0.5). 

When we talk about dBm, this is a measurement of change in decibels in reference to milliwatts. If the reference point it watts, it'll be dBW. 

Honestly, showing neighbors or rogues seen by own Access Points with a measurement of dB, doens't really make sense to me. A measurement of 137 dB could be a lot - but it could also be insignificant. It depends on the frame of reference.

137 dBf, and no one bats an eye.

137 dBW, and you're cooked. Well done, even.

[RaphaelL]

Did more digging this morning. 

{"ssid":"DIRECT-E6-HP OfficeJet 3830","bssids":[{"bssid":"E7","contained":false,"detectedBy":[{"device":"KK","rssi":138},{"device":"AS","rssi":3},{"device":"J7","rssi":8},{"device":"TK","rssi":11}]}]

AirMarshal flags this BSSID at 130++db on a single AP. Going to RF Spectrum the RSSI ( in dbm ) is -70dbm which is low and makes more sense since the other APs are already reading this BSSID with a low value.

Tested with MR 30.7 and MR31.1.8 same results. It's always 1 AP on the lot that is showing odd values.

Also why are the values inconsistent?

AirMarshal RSSI = db  RF Spectrum RSSI + Client page = dbm

[pdeleuw]

My dashboard shows the nearest "Other SSIDs" with the lowest dB value. I think, it is not the RSSI. As other mentioned, this should be expressed as dBm and should be negative. Just take a look at Network-wide > Monitor > Clients. Choose any wireless client and click on it. You see a metrik "Signal", expressed in dB. This is, according to the documentation, the SNR. I think, the value given by Air Marshal is the SNR, too.

[RaphaelL]

Hi Raphael 

This is a known issue that is resolved in firmware version MR 31.1.6 (Stable release candidate). Please upgrade the devices and let me know if the issue still occurs.

I just wished ( real bad ) that MR firmware would include more fixes instead of "General fixes and improvement". This would prevent me from opening a ton of cases. 

Will try that firmware and report