Time Domain EVM Measurement for RF Device Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional EVM measurement techniques in the frequency domain suffer from high implementation losses and inefficiencies, particularly when testing RF devices, which can lead to increased testing time and reduced throughput in evaluating multiple devices under test.
Innovation Solution
A method and system that perform EVM measurement in the time domain using a processor to time-align and de-rotate OFDM signals, apply complex equalization filters, and calculate EVM by minimizing mean square error, allowing for efficient determination of error vector magnitude with reduced losses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional frequency domain EVM measurement techniques are used, then measurement accuracy can be achieved, but implementation losses increase and testing time extends
Solution Approach 1:
The patent changes the measurement domain parameter from frequency domain to time domain. By performing EVM measurement directly in the time domain using cross-correlation between input and output signals, the method eliminates the need for FFT transformation and demodulation steps, thereby reducing implementation losses and testing time while maintaining measurement accuracy
Solution Approach 2:
The patent extracts the essential EVM measurement function from the complex frequency domain processing chain. By directly computing the cross-correlation in the time domain and extracting the distortion signal, the method removes unnecessary transformation steps, reducing both implementation loss and testing duration
2Measurement precision
If conventional frequency domain EVM measurement techniques are used, then comprehensive signal quality assessment is achieved, but device complexity and implementation losses increase
Solution Approach 1:
The patent replaces the mechanical signal processing chain (demodulator, FFT transformer, frequency domain comparator) with a direct time domain cross-correlation computational approach. This substitution simplifies the implementation by using straightforward time domain operations instead of complex multi-stage processing
Solution Approach 2:
The time domain cross-correlation method serves multiple functions simultaneously: it performs signal alignment, distortion extraction, and EVM calculation in a single unified process, reducing the need for separate processing stages and simplifying the overall measurement system
3Measurement precision
If multiple DUTs are tested using conventional methods, then comprehensive evaluation is achieved, but productivity decreases due to extended testing time
Solution Approach 1:
The time domain measurement method enables continuous and faster processing of multiple DUTs by eliminating the time-consuming frequency domain transformation steps. The direct cross-correlation approach allows for quicker measurement cycles, increasing the throughput of devices that can be evaluated per unit time while maintaining comprehensive evaluation capability
Data Source
AI summary
an orthogonal frequency division multiplexed (OFDM) output signal produced by a device in response to an OFDM input signal is accessed. The OFDM input signal includes OFDM input symbols in the time domain and the OFDM output signal includes OFDM output symbols in the time domain. The OFDM output symbols are time-aligned to the OFDM input symbols and a phase of the OFDM output signal is de-rotated with respect to the OFDM input signal. A complex equalization filter is applied to the OFDM output symbols in the time domain to obtain an estimate of the OFDM input symbols A distortion signal of the OFDM output signal is determined by subtracting the estimate of the OFDM input symbols. An error vector magnitude (EVM) is determined by dividing a root mean square of the distortion, by a root mean square of the OFDM input signal.