Delta-Sigma ADC Peak AGC Under Out-of-Band Interference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Delta Sigma (ΔΣ) analog to digital converters face challenges in signal strength estimation due to the removal of both quantization noise and interferer signals by filtering, leading to inaccurate gain control and potential overload, especially when interferer signals dominate, and the process is slow due to the need for filtering.
Innovation Solution
An improved ΔΣ analog to digital converter system with automatic gain control that provides an unfiltered digital output including in-band signals and out-of-band interferers, utilizing a signal peak estimator circuit to generate a gain control signal based on peak values rather than average values, and includes a multibit ΔΣ converter with adjustable full-scale, allowing for all-digital implementation and accounting for the signal transfer function.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If filtering is used to eliminate out-of-band noise, then quantization noise is removed, but interferer signals are also removed resulting in inaccurate signal-strength estimation
Solution Approach 1:
The patent segments the signal processing into distinct functional blocks: a wideband filter that passes both in-band signals and out-of-band interferers, and a separate narrowband filter that processes only the in-band signal for decimation. This segmentation allows the signal-strength estimator to access the wideband signal containing interferer information while the narrowband filter handles quantization noise reduction independently.
Solution Approach 2:
The patent introduces an intermediary wideband filter between the ΔΣ converter and the signal-strength estimator. This intermediary component passes both quantization noise and interferer signals to the estimator, enabling accurate signal-strength measurement without losing interferer information, while a separate narrowband filter handles the decimation function.
2Reliability
If filtering is used to remove quantization noise, then noise is reduced, but the process takes time limiting the speed of the automatic gain control loop
Solution Approach 1:
The patent divides the filtering function into two parallel paths: a wideband filter path that provides rapid signal-strength estimation without extensive filtering, and a narrowband filter path that performs thorough quantization noise removal for high-quality decimation. This segmentation enables the AGC loop to operate at high speed using the wideband path while maintaining signal processing quality through the narrowband path.
Solution Approach 2:
The patent applies partial filtering through the wideband filter for signal-strength estimation purposes, rather than applying full narrowband filtering to all signal paths. This partial action provides sufficient noise reduction for accurate estimation while maintaining high processing speed, reserving the more time-consuming narrowband filtering only for the decimation path where quality is paramount.
3Measurement precision
If gain is increased to compensate for low signal-strength estimate, then estimation accuracy is improved, but out-of-band signals may overload the analog to digital converter
Solution Approach 1:
The patent introduces a wideband filter as an intermediary component between the ΔΣ converter and the signal-strength estimator. This filter passes both in-band signals and out-of-band interferers to the estimator, enabling accurate signal-strength measurement that reflects the true total signal power. Consequently, the AGC loop sets appropriate gain levels that prevent converter overload while maintaining accurate estimation.
Solution Approach 2:
The patent implements a feedback mechanism where the signal-strength estimator continuously monitors the wideband signal containing both in-band and out-of-band components, and feeds this information back to the AGC loop. This feedback enables dynamic gain adjustment that responds to the actual total signal strength, preventing overload conditions while maintaining optimal signal processing.
Data Source
AI summary
An improved ΔΣ analog to digital converter system with automatic gain control response to out-of-band interferers including a ΔΣ multibit analog to digital converter responsive to an analog input for providing a digital output including the in-band signal and out-of-band interferers and quantization noise and a signal peak estimator circuit responsive to the out-of-band interferers for generating a gain control signal for adjusting the gain of a variable gain element which may be an independent element such as a variable gain amplifier or it may be the analog to digital converter itself by virtue of its having an adjustable full-scale.