How to estimate the talktime of a GSM or CDMA mobile?

[ratrace]

gsm-cdma talktime

Hello
Does anybody know how to estimate the talktime of a GSM or CDMA mobile?
I know the TX and RX current consumption, the PA efficiency, battery capacity and voltage, and the Pout vs Time.
Any formula, datasheets or links are welcome
thanks

 

[pisoiu]

Re: gsm-cdma talktime

I think it will be very hard to calculate this with exactity. The consumption is also dependant on network conditions (how far are you from the BTS, buildings around you, etc.), not only by the parameters of your mobile phone. You can perform practical measurements and compute an average.

/pisoiu

 

[ratrace]

Re: gsm-cdma talktime

As I said, I do have the transmitter distribution vs time.
This includes everything: path loss, propagation, multipath, etc.
I know that CMU200 from R&S and Racal 6103 has this option to measure talktime, using an internal algorithm. For something like this I am looking for.

 

[bird123]

Re: gsm-cdma talktime

How to measure Talk time and standby time but here is my 2 cents

1. make a call with call box to MSUT(mobile Unit)
2. change power level from low power to max power
3. record currents at all power levels
4. then, you can average them out

 

[toonafishy]

Re: gsm-cdma talktime

For CDMA, the transmitter is under power control from the base station so this must be taken into account.

 

[ratrace]

Re: gsm-cdma talktime

Things are more complicated that you think. You can not do a simple average of the powers for the reason that you have to follow the TX distribution vs time. As I said this distribution includes all the network variables. I am attaching for your knowledge an example of this distribution (measured) for a CDMA mobile in the network.
More than this, in the mobile are more variables that affect the talktime: internal resistance of the battery at different loads (i.e. at full charged battery, internal resistance is smaller compared to the resistance of a discharged battery). Another variable that affect the talktime is the transmitter efficiency (PA efficiency + post PA loss).
I remember that I saw long time ago a generic formula that includes all these variables and distributions, but I forgot where. For this I am looking for.

 

[zampanampa]

Re: gsm-cdma talktime

I know that the Pout vs. time plot for GSM mobile is very different from CDMA ones. GSM networks have a very slow power control. The average duration of a call is smaller than one would expect. I think, I have read something, which is little above of one minute.
Furthermore the settings are very conservative, because operators do not want to have interrupted calls.
The plot also depends, if you are located in urban, sub-urban or rural cell.

 

[jon2000]

Re: gsm-cdma talktime

For CDMA talk time, you can find the formula in CDG-4 document.

I just copy some infor here for calculating current, then you can find the talk time based on the battery capacity.

(1) IS-95
Mobile spends most of its time at an average power of ~-8 dBm for urban and ~-2 dBm for suburban environments. Total talk time is a combination of current drain at lower output powers and gated current where the transmitter is turned OFF during periods of no voice activity.

Weighted CDG current is calculated as:
I[CDG] = 0.6* I[gated] + 0.4*(5%* I[max power] + 15%* I[0 dBm] + 80%* I[-10 dBm] ) ---- Urban distribution
I[CDG] = 0.6* I[gated] + 0.4*(12%* I[max power] + 27%* I[0 dBm] + 61%* I[-10dBm] ) --- Suburban distribution

where 0.4 is the voice activity factor and the weighted percentages are the probability distribution of the different output powers at the antenna.


(2) CDMA2000
Mobile spends most of its time at an average power of ~-11 dBm for urban and ~-5 dBm for suburban environments.
In case of CDMA 2000, during periods of low or no speech activity, there is no gating allowed. Only the power spectral density of the data is reduced over the 16 power control groups.

Weighted CDG current can be calculated as
I[CDG] = (4%* I[max power] + 11%* I[0 dBm] + 85%* I[-10 dBm] ) ---- Urban distribution
I[CDG] = (9%* I[max power] + 19%* I[0 dBm] + 72%* I[-10 dBm] ) ---- Suburban distribution

where the weighted percentages are the probability distribution of the different output powers at the antenna.

 

[as_518]

Re: gsm-cdma talktime

in fact this is more complicated than that but you can consult the standards of cdm2000 it is available free of charge from tia.