Phase Noise Measurement Solutions

Technical Overviews

Selection Guide

Introduction

Finding the Best Fit for Your Test Requirement

Built upon 40 years of RF design and measurement experience, Keysight Technologies, Inc. continues to offer a wide range of the phase noise measurement solutions based on a variety of test and measurement techniques. However, finding the right solution can be a challenge. 

This document will help guide you in selecting the solution that best fits your specific measurement requirements. More information about phase noise can be found in the Additional Resources section in the back of this document. 

Phase Noise Overview

Phase noise is one of the most important figures of merit of a signal generating device and can well be a limiting factor in a mission-critical application in aerospace and defense, as well as in communications. 

The basic concept of phase noise involves frequency stability, which is defined as the degree to which an oscillating source produces the same frequency throughout a specific period of time. Frequency stability consists of two components: long term and short term. 

Long-term stability describes the frequency variations that occur over hours, days, months, or even years. By contrast, short-term frequency stability is about changes in the nominal carrier frequency of less than a few seconds duration. The focus of this document is on short-term frequency stability. 

While there are many technical terms to quantify phase noise, one of the most commonly adopted measures is the "single side-band (SSB) phase noise", ((f). Mathematically, the US National Institute of Standard and Technology (NIST) defines ((f) as the ratio of the power density at an offset frequency from the carrier to the total power of the carrier signal. 

Summary Comparison of Measurement Techniques

A variety of measurement techniques have been developed to meet various requirements for phase noise measurements. The three most widely-adopted techniques are: direct spectrum, phase detector and two-channel cross-correlation. While the direct spectrum technique measures phase noise with the existence of the carrier signal, the other two remove the carrier (demodulation) before phase noise is measured. 

Primary Phase Noise Measurement Techniques

Direct spectrum technique

This is the simplest and perhaps oldest technique for making phase noise measurements. The signal from the device under test (DUT) is input into a spectrum/signal analyzer tuned to the DUT frequency, directly measuring the power spectral density of the oscillator in terms of (f).

As the spectral density is measured with existence of the carrier, this method can be significantly limited by the spectrum/ signal analyzer’s dynamic range. 

Though this method may not be useful for measuring very close-in phase noise to a drifting carrier, it is convenient for qualitative quick evaluation on sources with relatively high noise. The measurement is valid if the following conditions are met: 

-The spectrum/signal analyzer’s inherent SSB phase noise at the offset of interest must be lower than the noise of the DUT. 

- If the phase noise measurement implementation does not differentiate amplitude (AM) noise from the phase noise (PN), like in most of the legacy spectrum/signal analyzers, then the AM noise of the DUT must be significantly below its PN (typically 10 dB will suffice).

The innovative Keysight X-Series phase noise measurement application (N/W9068A), differentiates the AM noise from PN and implements the AM rejection (at offset frequency below 1 MHz) effectively removing the impact of the AM noise to the PN measurement results. 

Phase detector techniques

To separate phase noise from amplitude noise, a phase detector is required. Figure 2 depicts the basic concept for the phase detector technique. The phase detector converts the phase difference of the two input signals into a voltage at the output of the detector. When the phase difference is set to 90° (quadrature), the voltage output will be zero volts. Any phase fluctuation from quadrature will result in a voltage fluctuation at the output. 

Several methods have been developed based upon the phase detector concept. Among them, the reference source/PLL (phase-locked-loop) is one of the most widely used methods. Additionally, the phase detector technique also enables residual/additive noise for two-port devices. 

Absolute phase noise measurements

Reference source/PLL method

As shown in Figure 3, the basis of this method is the double balanced mixer used as a phase detector. Two sources, one from the DUT and the other from the reference source, provide inputs to the mixer.

The reference source is controlled such that it follows the DUT at the same carrier frequency (fc) and in phase quadrature (90° out of phase) nominally. The mixer sum frequency (2fc) is filtered out by the low pass filter (LPF), and the mixer difference frequency is 0 Hz (dc) with an average voltage output of 0 V. 

Riding on this dc signal are ac voltage fluctuations proportional to the combined (rms-sum) noise contributions of the two input signals. For accurate phase noise measurements on signals from the DUT, the phase noise of the reference source should be either negligible or well characterized. The baseband signal is often amplified and input to a baseband spectrum analyzer. 

The reference source/PLL method yields the overall best sensitivity and widest measurement coverage (e.g. the frequency offset range is 0.01 Hz to 100 MHz). Additionally, this method is insensitive to AM noise and capable of tacking drifting sources.

Disadvantages of this method include requiring a clean, electronically tunable reference source, and that measuring high drift rate sources requires reference with a wide tuning range.