PWM Power Encoder With Non-Uniform Thresholds for RF Linearity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional direct digital-RF transmitters face challenges in achieving high power coding efficiency due to non-linearity issues, particularly for high-frequency transmissions, where delta sigma modulation leads to power inefficiency and noise spreading across the frequency domain, and existing PWM techniques suffer from degraded linearity performance.
Innovation Solution
Implementing a power encoder with non-uniform distributions of voltage thresholds and current levels, optimized through pre-emphasis linearization and look-up table-based pre-distortion, to enhance power coding efficiency by more than 5%.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If delta sigma modulation (DSM) is used as the power encoder to achieve high in-band signal-to-noise ratio, then the in-band SNR can be greater than 60dB, but the power coding efficiency becomes low and quantization noise increases abruptly near the band
Solution Approach 1:
The patent changes the fundamental parameter of the modulation scheme from delta sigma modulation to pulse width modulation (PWM). This parameter change transforms the power encoder's operation mode, enabling it to achieve both high power coding efficiency (above 70%) and acceptable linearity performance, thereby resolving the contradiction between in-band SNR and power coding efficiency
Solution Approach 2:
The patent substitutes the DSM mechanism with a PWM mechanism. Instead of using feedback loops and noise shaping characteristic of DSM, the invention employs PWM with non-uniform voltage threshold distribution, replacing the mechanical feedback-based system with a direct pulse width control approach that inherently achieves better power efficiency
2Loss of energy
If conventional PWM techniques are used to improve power coding efficiency, then power efficiency increases, but linearity performance degrades due to inherent nonlinearity of PWM quantization
Solution Approach 1:
The patent applies asymmetry by using non-uniform distribution of voltage thresholds in the PWM encoder. Instead of equally spaced thresholds, the invention employs asymmetric threshold values that are optimized to compensate for the inherent nonlinearity of PWM quantization, thereby maintaining both high power efficiency and good linearity performance
Solution Approach 2:
The patent implements preliminary action through pre-emphasis linearization and look-up table-based pre-distortion. These techniques pre-compensate for the expected nonlinearity in the PWM encoding process, allowing the system to achieve both high power coding efficiency and acceptable linearity by correcting distortions before they occur
3Measurement precision
If DSM-based direct digital-RF transmitter is used to achieve high in-band SNR, then the in-band signal quality improves, but the overall power efficiency decreases
Solution Approach 1:
The patent changes the power encoder type from DSM to PWM, fundamentally altering the energy consumption characteristics of the transmitter. This parameter change enables the system to achieve high power coding efficiency (above 70%) while maintaining acceptable signal quality, directly resolving the contradiction between in-band SNR and overall power efficiency
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
A power encoder (600) includes a pulse width modulator (610) for modulating a signal (630) according to a set of thresholds (620) to produce a pulse width modulated (PWM) signal and a switch mode power amplifier for amplifying the PWM signal (615) by switching states of switching devices (650) according to amplitudes of the PWM signal. At least one or combination of a distribution of values of the voltage thresholds in the set and a distribution of values of a current generated by different switching devices are non-uniform. The set of voltage thresholds includes at least two positive voltage thresholds.