Phase noise data is typically generated from a frequency spectrum plot which can represent a time domain signal in thefrequency domain. A frequency spectrum plot is generated via a Fourier transform of the signal, and the resulting values are plotted with power versus frequency. This is normally done using a spectrum analyzer. A frequency spectrum plot is used to define the spectral purity of a signal. The noise power in a band at a specific offset (FO) from the carrier frequency (FC) compared to the power of the carrier frequency is called the dBc Phase Noise.
A Phase Noise plot is generated using data from the frequency spectrum plot. Refer to Figure 1A. Using the dBc phase noise definition and calculating the dBc values for a continuous moving 1Hz band over the frequency offset range of interest generates a phase noise plot. A phase noise plot is simply these dBc values plotted versus the offset frequency (FO). Refer to Figure 1B. To summarize, phase noise is specified as the ratio of the noise power present in a 1Hz band at a specified offset from the carrier frequency to the power value of the carrier. Generally, a phase noise plot is single side, only one side of the frequency spectrum is plotted. This is due to the assumption that the frequency spectrum plot has a Gaussian profile and that both sides of the distribution are equal. For phase noise measurements which contain deterministic components, special consideration must be taken. Typically, a phase noise plot has the domain axis is in logarithmic format and the range in linear format. Refer to Figure 1B. Refer to application note AN-839 for more details. For other questions not addressed by the Knowledge Base, please submit a technical support request.