Suppression of phase-noise interference due to closely spaced data and calibration signals

At high frequencies, phase noise, which occurs at frequencies surrounding a data frequency, may seriously interfere with the measurement of nearby signals. Our motivation for removing phase noise was based on our use of simultaneous data and calibration signals, which were closely spaced in frequency. We found that by measuring the source phase noise, we are able to effectively remove phase-noise interference from the measured data and calibration signals. In order to accomplish this phase-noise suppression, a normalization procedure has been developed so that signals on differing measurement channels can be compared. Using this phase-noise suppression procedure and a prototype measurement system, we were able to improve magnitude mesurements by 36 dB. We were able to improve phase measurements by a factor of 70. We propose that this procedure can significantly improve measurement accuracy in many situations where two closely spaced signals, which have a common source, must be measured with high accuracy. This procedure can also be used to monitor, and thereby remove, other types of interference besides just phase noise.