Three practical tips for optimizing EVM measurements of wideband signals
[Copy link]
Error vector magnitude (EVM) measurements can help engineers gain insight into the performance of digital communication transmitters and receivers. For digitally modulated signals, any signal imperfections that affect the signal's amplitude and phase traces will affect the EVM measurement and its display.
EVM measurements provide a simple, quantitative performance evaluation parameter for digitally modulated signals.
Figure 1 shows the demodulation analysis of common modulation formats. The IQ measurement waveform data is divided into two paths: one path enters the demodulator to recover the original data bits, and the data bits are modulated to obtain the IQ reference (ideal) waveform; the other path is processed by signal compensation and measurement filters to obtain the IQ measurement waveform data. The signal error is the difference between the reference waveform and the compensated measurement waveform.
Some wireless standards, such as Wi-Fi and LTE, use decibels (dB) as the unit for EVM results:
EVM (dB) = 20 log10 (EVM (%))
“If you want to do your work well, you must first sharpen your tools.” As an EVM test tool, the signal analyzer must have as little impact on the EVM measurement as possible. Therefore, optimizing the measurement settings of the signal analyzer is also the key to EVM measurement, which is even more important for the analysis of broadband signals such as 5G and Wi-Fi 6.
Wireless communication standards calibrate EVM measurements at maximum output power. The power level of the first mixer in the signal analyzer can usually be controlled to ensure that high-power input signals do not cause distortion in the signal analyzer. The optimal setting of the mixer level depends on the measurement hardware, the characteristics of the input signal, and the specification test requirements.
The nonlinear components of the signal analyzer (such as mixers) may generate distortion under certain conditions. When the signal power input to the signal analyzer is too large, the signal will cause distortion in the input mixer. Adjustable input attenuation can prevent the input mixer from being distorted by high-power signals.
[Adjust input attenuation]
The analyzer's input attenuator reduces the signal power entering the input mixer. However, the input mixer level setting is a trade-off between distortion performance and noise sensitivity. At higher input mixer levels, a better signal-to-noise ratio (SNR) can be achieved, while at lower input mixer levels, distortion is lower. A good signal analyzer can provide fine-step input attenuation, which provides better resolution for optimizing the input mixer level.
[Turn on the built-in preamplifier]
In over-the-air (OTA) testing and test system scenarios with large insertion loss, the input signal level may be lower than the optimal mixer level. The built-in preamplifier has a better noise figure. Turning on this setting when the input level is too low can ensure that the signal input to the mixer remains in the optimal range.
Optimizing the Signal-to-Noise Ratio of IF Digital Converters
The application of broadband millimeter waves is the development trend of wireless technology. The broadband noise in the millimeter wave band and the excessive path loss between the signal analyzer and the device under test (DUT) will cause the SNR of the digital converter to decrease. If the SNR is low, the measured EVM will deteriorate, thus failing to accurately display the performance of the device under test.
Impact of SNR on Transmitter Measurements
The signal analyzer's system intermediate frequency (IF) noise must be low enough to obtain the best EVM measurement results. On the other hand, the input signal level of the digitizer must be high enough without overloading the digitizer. Therefore, the RF attenuator, preamplifier, and IF gain values need to be set according to the peak level of the signal to be measured. Modern new signal analyzers can optimize these hardware settings with a single button to improve SNR and avoid overloading the digitizer.
Optimizing EVM measurements to improve 5G NR signal demodulation analysis
Phase noise describes the frequency stability of an oscillator and results in errors in the phase component of the error vector signal. The phase noise of a signal analyzer is one of the causes of errors in EVM measurements. To obtain the best phase noise performance of a signal analyzer for modulation analysis, not only the phase noise profile of the signal analyzer (close-in and wide offset) but also the operating frequency, bandwidth, and subcarrier spacing (OFDM signal) of the input signal must be considered.
Select a specific LO phase noise mode for different operating conditions
The phase noise of the signal analyzer's local oscillator signal is converted to the input of the signal analyzer's mixer. The direct impact of phase noise on the IQ constellation diagram is the radial tailing of the symbol. When using high-order modulation schemes, the constellation point distance is smaller and the requirements for EVM performance are higher. It should be ensured that the phase noise performance of the signal analyzer does not affect the EVM measurement results, and the phase noise of the signal analyzer does not become a bottleneck restricting the EVM measurement.
Want to know more? Scan the QR code below
Download relevant information and participate in the lucky draw:
| 
提升卡
变色卡
千斤顶
京公网安备 11010802033920号
