In the course of dealing with various synthesizer configurations, we will learn that the output noise of a synthesizer is an important design consideration. The main sources of noise are leakage of the reference frequency in phase-locked loops and incomplete suppression of the unwanted component of mixer output (spurious). Another source of noise is the noise inherent in the oscillator.

If the spectral power density is measured at the output of an oscillator, a curve such as that of Figure 2-1 is observed. Rather than all of the power being concentrated at the oscillator frequency, some is distributed in frequency bands on both sides of the oscillator frequency.

As noise is a form of stability, it is useful to characterize frequency stability in the time domain in several areas. Short-term stability extends between a very small fraction of a second to 1 s, maybe under some considerations up to 1 min, and the value for the stability between 1 s and 1 min will be about the same. For longer time periods, we talk about long-term stability or aging. The aging is typically expressed in forms of how many parts in 10−10 or 10−11 per day the frequency changes. This information is in the time domain; in the frequency domain, we find terms like “random walk,” “flicker,” and “wide phase noise,” which describe the slope of spectral density. The Fourier frequency, at times labeled m is at times called sideband frequency, ...

Get Microwave and Wireless Synthesizers: Theory and Design now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.