Spectrum Analysis Basics, Part 1 – What is a spectrum analyzer?

The concept of spectrum analysis is a little daunting. From simple amplitude and frequency measurements to complex, application-specific measurements, there’s a lot to cover and understand. In this blog, I’ll go over the basics of spectrum analyzers to lay a foundation of understanding.

To begin, a spectrum analyzer (at the most basic level) functions as a frequency-selective, peak-responding voltmeter calibrated to display the RMS (root-mean-square) value of a sine wave. Not to be confused with a power meter, a modern spectrum analyzer performs digital signal processing with added capabilities to dive deeper into signals.

Types of spectrum analyzers

Spectrum analyzers originally measured only amplitude. These swept-tuned, super heterodyne analyzers evolved over time with the communications industry. With the need for phase measurements, signal analyzers took the place of their more basic counterparts. Spectrum analyzers measure the magnitude of an input versus signal frequency. Vector signal analyzers measure the magnitude and phase of an input signal at a single frequency. Today’s signal analyzers combine functionality of the earlier evolutions of spectrum analyzers, such as analog, vector, and FFT (fast Fourier transform) measurements. Along with these new high-functioning signal analyzers, more compact options have come about, such as the Keysight FieldFox and modular spectrum analyzers. While the terms signal analyzer and spectrum analyzer are not interchangeable, they often refer to the same measurements and concepts. Many people use “spectrum analyzer” to refer to signal analyzers.

Frequency domain vs. time domain

If a signal analyzer measures complex signals and helps you understand the functionality of your DUT (device under test), what’s the difference between that and an oscilloscope? An oscilloscope performs measurements in the time domain, with units like Volts/sec. It allows us to view the instantaneous value of an electrical event as a function of time.

Using a Fourier transform, we can move from the time domain to the frequency domain. Fourier theory tells us any time-domain phenomenon consists of one or more sine waves of appropriate frequency, amplitude, and phase. The frequency domain unveils new information about your signals, such as the amount of energy present at each frequency. Using the frequency domain helps you view individual sinusoidal waves or spectral components and see how they contribute to your overall response.

Why measure in the frequency domain?

A complex signal in the time domain looks vastly different than in the frequency domain. As seen in Figure 1[G(3] , the time domain measurement shows an impure sine wave. Without measuring in the frequency domain, the source and frequency of the second harmonic remains unknown. Spectrum analysis uncovers sources of interference by displaying the spectral components independently. The time domain still provides useful information, such as the pulse rise and fall times of a signal, but the frequency domain allows us to determine the harmonic content of a signal, such as out-of-band emissions and distortion.

Figure 1: Different frequency components contributing to a complex time domain signal.

Applications of signal analyzers

Now that we know why we use signal analyzers, let’s look at where they’re used.

Spectrum analysis applies across many factions of engineering. For example, engineers must check cellular radio systems for harmonics of the carrier signal in order to prevent interference between systems at the same frequencies. Distortion of the message modulated onto a carrier also requires spectrum analysis, shown in Figure 2. An especially troublesome issue, third-order intermodulation, occurs when two tones of a complex signal modulate each other. The distortion components from this phenomenon can fall within the band of interest, meaning a simple filter cannot easily remove them.

Figure 2: harmonic distortion measurement on Keysight’s X-series measurement application.

Government agencies also utilize spectrum analysis to monitor usage of allocated spectrum. Different frequency bands are reserved for certain activities such as wireless internet, mobile phones, emergency communications, and more. If transmitters do not stay within their allocated frequency bands, signal energy enters adjacent channels and causes interference, shown in Figure 3.

Figure 3: Keysight’s GSM radio test showing unwanted emissions.

Another type of interference, electromagnetic interference (EMI), impairs the operation of other systems. EMI refers to unwanted radiated or conducted emissions, such as emissions from an electronic device or power lines. Electronic products must undergo EMI compliance testing to ensure the emission levels adhere to government or industry-standard regulations. Some signal analyzers feature EMI pre-compliance testing to help ensure DUTs pass compliance tests, shown in Figure 4.

Figure 4: EMI pre-compliance test on an X-Series spectrum analyzer.

Another common signal to analyze is noise. Noise refers to random, unwanted variations in or disturbances to your signal amplitude. Noise can distort your original signal, sometimes even masking the signal entirely. Measurements such as noise figure (seen in Figure 5), signal-to-noise (SNR) ratio, and phase noise can help engineers understand the performance of a device or the overall system.

Figure 5: A noise figure measurement on a Keysight X-Series signal analyzer.

The applications described in this blog represent just a small percentage of what spectrum analysis enables engineers to do today. Tune in to Part 2 for a look at what’s inside a spectrum analyzer and an explanation of how it works.

Tag: Measurement Fundamentals

Related Tags

Spectrum Analysis Basics, Part 1 - What is a Spectrum Analyzer?