Multichannel Predistortion with Sigma-Delta Noise Decorrelation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing signal transmission devices with power amplifiers and digital predistortion face challenges due to the high number of bits required for the DAC, leading to increased complexity, surface area, and consumption, especially in multichannel systems.

Innovation Solution

The implementation of a sigma-delta encoder between the predistortion block and the DAC, coupled with a method to cancel out quantization noise in the feedback loop, allows for a moderate sampling frequency and reduced complexity in multichannel systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a high number of bits is used for the DAC to maintain signal-to-noise ratio, then the signal quality is improved, but the complexity, surface area, and energy consumption of the device increase

Engineering Contradiction:

Improvesignal-to-noise ratioVSAvoidDAC complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the signal processing into separate frequency bands using a polyphase filter bank. The sigma-delta encoder operates at a lower sampling rate for each band, segmenting the overall high-resolution requirement into multiple lower-resolution parallel channels. This segmentation allows the DAC to use fewer bits while maintaining overall signal quality through the combined output of multiple bands.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent changes the sampling frequency parameter by using a lower sampling rate in the sigma-delta encoder compared to traditional high-rate DACs. By operating at a moderate sampling frequency and using noise shaping to push quantization noise to higher frequencies, the system achieves acceptable signal-to-noise ratio with fewer DAC bits, thereby reducing complexity and power consumption.

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If a high sampling frequency is used to maintain signal quality, then the signal fidelity is improved, but the processing complexity and power consumption increase

Engineering Contradiction:

Improvesignal fidelityVSAvoidpower consumption

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent employs periodic modulation and noise shaping techniques where quantization noise is deliberately shaped to occur at specific frequency periods outside the signal band. This periodic action allows the system to use lower sampling frequencies while maintaining signal fidelity through constructive and destructive interference patterns in the frequency domain, reducing the overall processing burden and power consumption.

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The patent converts the harmful quantization noise introduced by low-bit DACs into a beneficial tool by using noise shaping to push the noise spectrum outside the signal bandwidth. The quantization noise, instead of degrading signal quality, is redirected to frequency regions that can be easily filtered out, allowing the system to achieve high signal fidelity with lower sampling frequencies and reduced power consumption.

Inventive Principle:

Principle #22Blessing in disguise (Convert harm into benefit)

3Measurement precision

If noise filtering is applied to suppress quantization noise, then the signal quality is improved, but the bandwidth and sampling frequency requirements increase

Engineering Contradiction:

Improvesignal qualityVSAvoidsampling frequency

Core Design Contradiction:

Measurement precisionVSSpeed

Solution Approach 1:

The patent applies preliminary noise shaping and spectral redirection before the signal reaches the DAC and subsequent filtering stages. By pre-shaping the quantization noise spectrum to concentrate noise energy outside the signal band, the system avoids the need for high-frequency filtering that would require higher sampling rates. The noise is already positioned in the desired location before conversion, eliminating the need for aggressive post-conversion filtering.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS12289124B2Transmitter device and methods for predistortion and decorrelation of noise in such a transmitter device

Publication Date: 2025.04.29 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES

AI summary

A multichannel transmitter device includes, on each channel: a transmission processing channel designed to process an input signal and comprising a predistortion block applying a predistortion to the input signal on the basis of predistortion coefficients, a DAC, a first analogue filter, a power amplifier and a sigma-delta encoder between the predistortion block and the digital-to-analogue converter and designed to carry out notably a quantization of the predistortion block; a return channel associated with the transmission processing channel and comprising a block for estimating predistortion coefficients so as to estimate predistortion coefficients on the basis of the input signal and of a feedback signal; the transmitter device wherein the block for estimating predistortion coefficients is designed to estimate the predistortion coefficients on the basis of at least the input signal and of a signal resulting from the subtraction, from the feedback signal, of a signal representing the quantization noise resulting from the quantization carried out by the sigma-delta encoder.