Method for calibrating a frequency synthesiser using two-point FSK modulation

A method for calibrating a frequency synthesiser using two-point modulation in a transceiver, the frequency synthesiser comprising:

• a low-frequency or low-pass phase-locked loop, which comprises at least one phase comparator to compare a reference signal from a reference oscillator, and a modulation signal from a sigma-delta modulator, a charge pump connected to the output of the comparator, a loop filter connected to the output of the charge pump, a voltage-controlled oscillator VCO receiving an output signal from the loop filter, and generating a high frequency signal, and a programmable divider connected to the sigma-delta modulator and receiving the high frequency signal, and
• a high frequency access with a digital-to-analogue converter DAC, which is connected to the voltage-controlled oscillator VCO of the phase-locked loop directly or via an adder, which performs the addition of the output signal from the loop filter, and the output signal from the digital-to-analogue converter,
• wherein the calibration method includes:
• in a first phase of a calibration mode with the phase-locked loop closed, supplying an excitation signal generated by a calibration unit of the transceiver to the sigma-delta modulator by deactivating the digital-to-analogue converter and transmitting the output signal from the loop filter to the calibration unit, which digitally converts the incoming signal transmitted from the output of the loop filter, and offsetting the phase shift between the excitation signal and at least the output signal from the loop filter in the calibration unit, and
• in a second phase of the calibration mode with the phase-locked loop closed, supplying the excitation signal to the sigma-delta modulator and to the activated digital-to-analogue converter, and calibrating the gain of the digital-to-analogue converter, by checking in the calibration unit, the sign or the polarity of the output signal from the loop filter with respect to the excitation signal, and using a dichotomy algorithm to adjust the gain.