Vector Signal Alignment Using Transform-Domain Phase Rotation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital predistortion techniques for wireless communication systems face challenges in accurately time-aligning input and feedback signals due to computational intensity and power consumption, which affects the efficacy of nonlinear power amplifier compensation.
Innovation Solution
The method involves transforming feedback vector signals from the time domain to a transform domain, rotating them based on a measured time delay, and then transforming them back to align them with input vector signals, using techniques like Fourier or wavelet transforms, and applying scaling factors to minimize residue.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If polynomial interpolation (e.g., Farrow structure) is used for time alignment, then time alignment precision is improved, but computational complexity and power consumption increase
Solution Approach 1:
The patent replaces the conventional time-domain polynomial interpolation mechanism with a frequency-domain phase rotation mechanism. By transforming the alignment problem from the time domain to the frequency domain using Fourier transforms, the complex polynomial operations are substituted with simpler complex multiplications (phase rotations), significantly reducing computational complexity while maintaining alignment precision.
Solution Approach 2:
The patent changes the domain parameter from time domain to frequency domain, and changes the operation parameter from polynomial coefficients to phase rotation angles. This parameter transformation allows the same time alignment function to be achieved with computationally more efficient operations in the frequency domain.
2Measurement precision
If polynomial interpolation (e.g., Farrow structure) is used for time alignment, then time alignment precision is improved, but power consumption increases
Solution Approach 1:
The patent substitutes the high-power polynomial interpolation operations with lower-power frequency-domain phase rotation operations. The transformation to frequency domain converts computationally intensive time-domain filtering into simpler complex exponential multiplications, directly reducing power consumption while preserving alignment accuracy.
3Measurement precision
If higher degree or more taps are implemented in interpolator, then time alignment precision is improved, but computation time increases
Solution Approach 1:
The patent replaces the time-consuming polynomial interpolation computations with efficient frequency-domain phase rotations. By using the properties of Fourier transforms, the alignment operation becomes a simple phase shift in the frequency domain, dramatically reducing computation time while achieving the same precision goals.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A processor receives a first vector signal and a second vector signal from a circuit in response to the circuit receiving the first vector signal. The processor transforms the second vector signal from a time domain to a transform domain. The processor rotates the transformed second vector signal by a phase that is proportional to a time delay between the first and second vector signals to time-align the second vector signal to the first vector signal.