Browse Prior Art Database

Method to reduce the amount of temperature/Vbat measurement

IP.com Disclosure Number: IPCOM000018501D
Original Publication Date: 2002-Sep-01
Included in the Prior Art Database: 2003-Jul-23
Document File: 1 page(s) / 157K

Publishing Venue

Siemens

Related People

Thomas Stichelbout: AUTHOR

Abstract

Temperature measurement is necessary in a mobile phone to control the frequency stability and other parameters. Until now the temperature measurement is done with a specific Analog-Digital-Converter (ADC) or with a receiver chain (RX chain). Using the RX chain implies that an extra ADC is not needed but the RX chain is not so often available for measurements. The temperature measurement can only be done when the RX chain is not receiving and the same RX chain is also used for other measure- ments such as voltage measurement on the battery (Vbat).

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Industrie

Method to reduce the amount oftemperature/Vbat measurement

Idee: Thomas Stichelbout, DK–Aalborg

Temperature measurement is necessary in a mobilephone  to  control  the  frequency  stability  and otherparameters. Until now the temperature measurementis  done  with  a specific Analog-Digital-Converter(ADC) or with a receiver chain (RX chain). Usingthe  RX  chain  implies  that  an  extra  ADC is notneeded but the RX chain is not so often available formeasurements.  The  temperature measurement canonly be done when the RX chain is not receiving andthe same RX chain is also used for other measure-ments such as voltage measurement on the battery(Vbat).

The proposed method uses an extrapolated the tem-perature/battery voltage curve from previous meas-urements (figure 1).  As  the  temperature  variationversus time depends a lot on the user scenario, sev-eral models of extrapolation are available to best suitthis evolution at a given time. A very simple modelto extrapolate the temperature curve is a linear ap-proximation based on the last two previous tempera-ture measurements. For each model an error is cal-culated and is used as an input to a neuronal networkwhich will decide which model fits the evolution bestor if extra temperature measurement should be usedinstead. The amount of measurements can be reducedwhen the temperature variation fits the models avail-able.

Fig. 1: Measurement method

Siemens  Technik Report      Jahrgang 5  Nr.18      August 2002