The Rohde & Schwarz signal and spectrum analyzer portfolio offers options ranging from low-cost, yet powerful 1 GHz analyzers to handheld and mid-range models to full-featured 85 GHz spectrum analyzers. Designed by the RF experts at Rohde & Schwarz, all spectrum analyzers feature exceptional signal integrity, high value and excellent reliability.
A spectrum analyzer does what the name suggests: it detects the signals present in a selected range of spectrum. The basic function is to represent the signals in a graphical display as amplitude—or power level—on the y-axis, against frequency on the x-axis; the amplitudes of detected signals are represented in the frequency domain. An RF spectrum analyzer covers radio and microwave frequencies. The maximum frequency range with preselection currently available is from 2 hertz to 85 GHz; higher frequencies are possible with external mixers. Usually, linear scale is used for frequency on the x-axis, and a logarithmic or decibel scale (also logarithmic), for amplitude on the y-axis, so that signals of widely varying amplitude can be seen at the same time. Spectrum analyzers are widely used in RF test to display not only properties of wanted signals, such as whether a signal is occupying the designated bandwidth, but also to search for unwanted signals.
For RF test, a pure spectrum analyzer for detecting the level of wanted and unwanted signals by displaying the spectral components in a frequency range hardly exists any more. The nature of many modern pulsed signals, plus the necessity to detect and investigate transient signals, means that the classical spectrum analyzer using the same super heterodyne principle as radio receivers cannot reliably detect all signals intermittently present as transients, or measure the phase of a signal. As the frequency range of interest (the frequency span) exceeds the capability of the spectrum analyzer to process data simultaneously, the frequency span is scanned (swept) from low to high. If a transient signal is not present while the frequency is being swept, it is not detected.
Digital processing using Fast Fourier Transformation (FTT) from the time domain to the frequency domain has greatly extended both the signal detection and analysis capabilities of the super heterodyne spectrum analyzer. FFT provides much faster capture and analysis of the frequency span: using FFTs in parallel results in a wider instantaneous bandwidth so that, with suitable filters, pulsed and transient signals are also detected. Many spectrum analyzers will also offer zero span mode to analyze phase as well as amplitude of a signal, and demodulate the signal at the selected frequency. Apart from the simple representation of the detected signals on a screen, measurements of noise, gain, phase, occupied signal bandwidth, and adjacent channel power are all possible. The digital signal can be exported for post-processing by software tools providing additional analysis.
A signal analyzer, correctly a vector signal analyzer (VSA), is used to demodulate and analyze signals with complex, digital modulation. A VSA captures signals at a fixed center frequency, using filters to set the bandwidth—or span—of the spectrum display; a spectrum analyzer sweeps a wider range of frequency. Compared to a dedicated spectrum analyzer, a VSA includes phase information, plus additional advanced measurements of signal properties not obtainable using spectrum analysis, and uses digital processing to demodulate signals based on digital in-phase (I) and quadrature (Q) modulation components. A VSA analyzes signal characteristics such as signal-to-noise ratio (or carrier-to-noise ratio), error vector magnitude (EVM) and code domain power. All the characteristics of pulsed or transient signals can be measured, including all level, frequency, phase, noise, gain, occupied signal bandwidth and adjacent channel power values.
Noise level and bandwidth for measurement instruments is invariably a trade-off; as a VSA is centered on a fixed frequency, a narrower analysis bandwidth is sufficient so that a well designed VSA has a low noise floor and excellent sensitivity for detecting low level signals.
Most VSA will also include a spectrum analysis mode better suited for (unwanted) signal detection, increasing the span of the captured signal, but reducing the demodulation possibilities to AM, FM, or /фM.
The frequency range needed for a spectrum analyzer will depend on the application, meaning the frequencies to be investigated for both wanted and unwanted signals, and the purpose of the signal detection. For spectrum monitoring, for example, the frequency range only needs to include the frequencies to be monitored. For device development and EMI investigations, many standards require measurements of spurious emissions at the third harmonic of the fundamental frequency; for a device operating in the 2.4 GHz ISM band, such as a Wi-Fi or Bluetooth® device, a frequency range of at least 7.2 GHz is required. For compliance to standards, in some cases spurious emissions to the 5th harmonic are required; for the 2.4 GHz device a frequency range of 12 GHz is required. For 5G devices operating in the n258 band from 24.25 to 27.50 GHz, there are very few spectrum analyzers available with the necessary maximum frequency of 82.5 GHz. Many standards from organizations such as ETSI, ANSI or 3GPP specify limits for out-of-bands emissions, much closer to the fundamental frequency. In every case, check the standards that apply to the device to test and, as a rule of thumb, aim for a maximum frequency that exceeds the maximum anticipated by 20%.
In general, dynamic range describes the maximum and minimum values an instrument can measure; for a spectrum analyzer designed to detect several signals simultaneously, the definition is the analyzer’s ability to detect a weak signal in the presence of strong signal. The dynamic range of a spectrum analyzer is defined as the ratio, in dB, of a larger to a smaller signal at which the spectrum analyzer can measure the smaller to a given accuracy in the presence of the larger.
As a common use of a spectrum analyzer is to search for spurious emissions in the presence of the wanted signal, the ability of the analyzer to detect a weak signal in the presence of a strong signal is a fundamental performance criterion. The maximum signal level range, noise floor, phase noise, and spurious response of the instrument all play important roles in determining dynamic range.
The dynamic range is limited for the weaker signal by the analyzer’s inherent noise and for the stronger signal by nonlinearities.
The inherent noise is specified by the displayed average noise level (DANL) given in dBm and normalized to a 1 Hz resolution bandwidth. A preamplifier reduces the DANL, which helps to detect weak signals but actually reduces the overall dynamic range.
The nonlinearities are shown by the 1 dB compression point, the second harmonic distortion, and the TOI (third-order intercept).
The phase noise of a waveform means brief, rapid, fluctuations in the frequency, seen on a spectrum analyzer screen as blurring or judder of the waveform on the display. Phase noise spreads the power of a signal to adjacent frequencies, resulting in noise sidebands, weakening the useable signal power, and reducing signal quality. A weak signal can disappear into the phase noise of a strong adjacent signal.
Phase noise in the frequency domain corresponds to jitter in the time domain; a fluctuation in frequency is also a deviation of the edge of a signal in time.
The cause of phase noise (and jitter) is irregularities in the performance of the oscillator clocking the waveform.
An ideal oscillator would generate a pure sine wave; all the power of the signal is at a single frequency. However, all real oscillators have instabilities causing phase-modulated noise components. The phase noise components spread the power of a signal to adjacent frequencies. Oscillator phase noise often includes low frequency flicker noise and may include white noise. Phase noise describes the stability of an oscillator in the Frequency Domain, while jitter describes stability in the Time Domain.
Phase noise can be measured using a spectrum analyzer, so long as the phase noise of the device under test is large compared to the phase noise of the local oscillator in the spectrum analyzer's local oscillator.
The inherent phase noise of the spectrum analyzer will limit the ability to perform phase noise measurements and impacts error vector magnitude (EVM) measurements on digitally modulated signals, especially narrowband signals.
Some spectrum analyzers will offer optional higher accuracy oscillators at an additional cost to improve the sensitivity of phase noise measurements.
There is no “correct” answer to this question, the best spectrum analyzer will depend on the individual circumstances. The key deciders will be the frequency of the signals to measure, the characteristics of signals that shall be measured, exactly which measurements are required, where the signals shall be measured, and the available budget. The spectrum analyzer frequency range will determine the minimum and maximum frequency that can be measured. The required speed of measurement, analysis bandwidth, phase noise, dynamic range, and sensitivity all depend on the signals to measure, and the accuracy required. The analysis capabilities must match the individual requirements. Additional criteria include issues such as portability and weight, option concepts for adding additional performance after the original purchase, service and calibration support, and suitability for the existing test infrastructure; can the new analyzer act as a plug-in replacement for previous equipment?
The Rohde & Schwarz spectrum analyzer portfolio includes solutions for both general purpose measurements and specific industry standards. It covers:
