Generic selectors
Exact matches only
Search in title
Search in content
Post Type Selectors

# C/I for GSM

Viewing 15 posts - 1 through 15 (of 15 total)
• Author
Posts
• #38771
bpong
Guest

can anyone confirm if C/I= 12dB for GSM?
can I interpret that if I got a wanted signal at -90dBm, the interference cannot be higher than -102dBm?

#38772
Arindam
Guest

Hi,
In GSM the acceptable value of C/I is 12dB or higher. Allthough some vendors say that with hopping even 9dB or higher can be acceptable.

This means if your serving frequency (C) should be at least 12dB higher than the interferer(I).

So if you have a serving ARFCN at -90, then the interferer should be at -102 or lower.

#38773
bpong
Guest

Does the interferer (I) mean co-channel ARFCN or adjacent channel?

#38774
Arindam
Guest

When you’re talking about C/I, it means co-channel. For adjacent channel , the relevant parameter is C/A.

#38775
Arindam
Guest

The term “I” indeed is the co-channel interferer. When we are talking about adjacent channel interference, the relevant parameter is C/A.

#38776
bpong
Guest

What’s the C/A value for GSM, offset at 1 channel & 2 channels away?

#38777
Arindam
Guest

Usually the recommended value of C/A is 3dB. This is an Ericsson recommended value.

#38778
bpong
Guest

Is it so low, 3dB? I thought it should be around 50-60dBc.

#38779
Arindam
Guest

Hi,
I am afraid it is not 50-60. May be ,you could give me your mail id, I can send you a document dealing with C/A and C/I.

#38780
bpong
Guest

I just read the GSM spec.
C/A @ 200khz = 9db

#38781
AliShah
Guest

Can anyone tell me the range of C/A in GSM ???? please……… m much confused

#38782
pix
Guest

h,

C/A is just the C/I with I = the adjacent freq.

an ok value is C/A > -9dB, for speech purposes.

You can find all the thresholds you need in 45.003 i believe.

regards
pix

#38783
gm
Guest

When predicting C/Ic=9dB for the purpose of frequency planning, one must keep in mind that a prediction is a prediction. In reality, the measured value can be higher or smaller with a probability of 50%. Theoretical mathematical calculation can show more details. Therefore it is better to predict a value higher than 9dB, at least 12dB or 18dB or more.

#38784
gm
Guest

9dB is for speech.. EGPRS can require a C/I up to 30dB ! (mcs9)

#38785
festus
Guest

Am finding it difficualtto calculate the percentage of C/I while prepering report for ssv. could someone help me with the fomular?

Viewing 15 posts - 1 through 15 (of 15 total)
• The forum ‘Telecom Design’ is closed to new topics and replies.