We have an SM which gets disconnected from AP for around 2/3 minutes .THis happens 4 to 5 times a day.This SM is at a distance of
6 kms.THe link efficiency in down does not go > 80 %, where as up is almost 100 % (most of the time & > 95 % ).
Another SM at 12 kms is also on the same AP.Which is stable.The link efficiency for this SM is > 95 in bith up & down case.
The spectrum Analysis at SM end for the SM which gets disconnected has noise at around -50 to -55 for all freq .
The sessions page has proper RSSI/Power & jitte values.Are as follows when te 6 km SM got disconnected.
Can some one let me know what is the reason for lesser down efficiency, is the spectrum at SM not in order can result in reduced down efficiency…?
Also when the SM loses the AP , it does not find anything in Aiming more for some time (2.3 minutes)…
LUID: 002 : MAC: State: IN SESSION (Encrypt Active)
Site Name :
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 979 (approximately 9.09 miles (47971 feet))
Session Count: 1, Reg Count 1, Re-Reg Count 0
RSSI (Avg/Last): 1101/1096 Jitter (Avg/Last): 6/6 Power Level (Avg/Last): -64/-64
DnRate(SM): 3500 DnLimit(SM): 500000 UpRate(SM): 3500 UpLimit(SM): 500000 (kbit)
LoUpCIR(SM): 20000 LoDnCIR(SM): 20000 (kbps)
Rate : VC 18 Rate 1X/1X
LUID: 003 : MAC: State: IDLE
Session Timeout: 0, AirDelay 334 (approximately 3.10 miles (16366 feet))
Session Count: 17, Reg Count 16, Re-Reg Count 5
RSSI (Avg/Last): 1134/0 Jitter (Avg/Last): 4/0 Power Level (Avg/Last): -63/-96
DnRate(AP): 20000 DnLimit(AP): 500000 UpRate(AP): 20000 UpLimit(AP): 500000 (kbit)
LoUpCIR(D): 0 LoDnCIR(D): 0 (kbps)
Rate : VC 19 Rate 1X/1X
To clarify, the noise level at the SM with the problem is -50 to -55dB?
If that’s true, you are near a very high power RF source in the 900 band. I am surprised it will register at all. -55dB is an extremely high noise level. Try moving the SM to another location on the property and see if you can find a spot where the noise floor is lower.
The good news is that the source of interference is localized to that customer so I would suspect a baby monitor, 900 cordless phone, etc.
"To clarify, the noise level at the SM with the problem is -50 to -55dB? "
Yes.
Forgot to mention.It is a 2400 gear and not the 900 ones.900 in India is use by GSM Cellphone Providers and is not available.
Same issue at 900 or 2400. You have a very strong interference source nearby.
Ok. Considering this we have good stability !!.Since in 24 hours this link with reflector breaks 3/4 times a day for 2/3 minutes. 
Can changing following in configuration.html help & what r this , does it also change type of modulation ?
2.5ms Frame Time / 10.0Mbps Bit Rate
5.0ms Frame Time / 5.00Mbps Bit Rate
5.0ms Frame Time / 3.33Mbps Bit Rate
I have not played with those settings.
Is the link set at 2X rate or 1X rate? Setting to 1X will improve stability quite a bit.
Link is 1X/1X
Is there another AP this SM can see at a different frequency?
Yes . But Same thing happens , infect the Power level on Another AP is far better , but this problem remains same when this SM is shifted to another AP
We have 2 AP.And r using 2415 & 2437.5 .Have not tried 2457.5 on another AP , as I have to change all freq. for other SMs to .Since all other SMs on the 2nd AP r very stable.So was just tryig to find if this 2/3 minute disconnect can be sorted out…We tried moving to some of the location on the same property , but results r same.
This Terrace has got some GSM equipements installed by Cell company.But they do not use 2400 frequencies.They r into 1.8g/1.9ghz band
The GSM is the problem. The power levels of Cellular transmitters are so high, that the harmonics on either side of the center frequency are strong enough to interfere with the very low power SM.
The first odd-order harmonic from 1.8GHz is 2.39GHz, and the first odd-order harmonic from 1.9GHz is 2.53GHz.
I assume you have a reflector on this? If not, adding a reflector and lowering the Tx power on the SM might be the solution.
Ok. We have a reflector.With the reflector On the SMs the ,AP session for both SM
is as follows.Where the LUID:002 is the SM which is having Problem.
LUID: 002 : MAC: State: IN SESSION (Encrypt Active)
Site Name :
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 334 (approximately 3.10 miles (16366 feet))
Session Count: 14, Reg Count 14, Re-Reg Count 9
RSSI (Avg/Last): 1139/1175 Jitter (Avg/Last): 5/3 Power Level (Avg/Last): -63/-62
DnRate(SM): 1000 DnLimit(SM): 10000 UpRate(SM): 3500 UpLimit(SM): 500000 (kbit)
LoUpCIR(SM): 20000 LoDnCIR(SM): 20000 (kbps)
Rate : VC 18 Rate 1X/1X
LUID: 003 : MAC: State: IN SESSION (Encrypt Active)
Site Name :
Software Version : CANOPY 7.3.6 Oct 24 2005 12:06:56
Software Boot Version : CANOPYBOOT 3.0
FPGA Version : 070605 (DES Sched) P9
Session Timeout: 0, AirDelay 978 (approximately 9.08 miles (47922 feet))
Session Count: 1, Reg Count 1, Re-Reg Count 0
RSSI (Avg/Last): 1027/983 Jitter (Avg/Last): 8/12 Power Level (Avg/Last): -65/-67
DnRate(SM): 3500 DnLimit(SM): 500000 UpRate(SM): 3500 UpLimit(SM): 500000 (kbit)
LoUpCIR(SM): 20000 LoDnCIR(SM): 20000 (kbps)
Rate : VC 19 Rate 1X/1X
You are not having a problem getting enough signal to the AP, the problem is that the SM is having difficulty getting enough signal from the AP sue to the abnormally high noise floor.
"You are not having a problem getting enough signal to the AP, the problem is that the SM is having difficulty getting enough signal from the AP sue to the abnormally high noise floor."
at SM am I correct ?
I have reduced the Power level at SM . Not sure if this would help.? Reducing Power level at Sm will also result in SM pickup less signal & less noise…Am i correct ?
This reduction has resulted in less uplink efficiency.
Will check with reduced Power level tmrow in working hours , by shifting this SM to Another AP where it see more power
Reducing the power level on the SM only affects transmit power from the SM to the AP. It has no bearing on the receive sensitivity of the SM. Turn the power back up on the SM.
The answer to your question is that the noise floor is too high for that SM. If you have a reflector and you have moved the SM to different locations, there is nothing more you can do short of trying to isolate the SM/Dish from the GSM by creating an RF shield.
ok.Thanks. Any idea on how could be an RF shield like…?Thanks in Advance.
It is Jus that overall availability for this SM still is still very good & this interruptions are critical for some of the customers getting services on this SM , that this needs best possible effort , before moving SM t alternate location …
Ok. Thanks once again.Will give this a try & put a feedback soon
Ok. We r changing the ethernet Cable from UTP to STP & after that will construct the above type cage .
The shielded cable may be enough.
If you do need to add the shielding, the wavelength at 1900MHz is 6.22in. If you can find mesh that is 1/8 of this wavelength or about 3/4" you should get some good shielding.
Yes. I also hope so.
In case the Shield is used , what should be the metal of the mesh ? , copper/alluminium/steel ?