Hybrid ADC Biosignal Detection for Low-Noise Neural Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing brain-computer interface systems face challenges in detecting low signal levels amidst noise, requiring miniaturization, ultra-low power consumption, and precise neuronal signal collection, especially in fully-inserted semiconductor devices.
Innovation Solution
A bio-signal detection apparatus with an analog front end unit, comprising low-noise amplifiers and bandpass filters, and a hybrid analog-to-digital converter using successive-approximation and single-slope ADCs, sharing a comparator and capacitor array, to convert bio-signals efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Volume of moving object
If a fully-inserted semiconductor device is used for brain-computer interface, then miniaturization and integration are achieved, but detecting low signal levels amidst noise becomes more difficult
Solution Approach 1:
The device is segmented into multiple functional blocks including analog front end circuits, hybrid ADC, and digital signal processing units. Each block performs a specific function to progressively enhance signal quality while minimizing noise, enabling effective detection of low-level bio-signals despite the compact fully-inserted form factor
Solution Approach 2:
An analog front end unit is introduced as an intermediary between the bio-signal source and the digital processing unit. This intermediary performs critical functions including low-noise amplification, bandpass filtering, and hybrid analog-to-digital conversion, thereby mediating the transition from weak analog bio-signals to processed digital signals while maintaining signal integrity in a miniaturized device
2Measurement precision
If channel expansion is implemented to collect neuronal signals more precisely, then measurement precision is improved, but device complexity increases
Solution Approach 1:
Multiple ADC functions (successive-approximation ADC and single-slope ADC) are merged into a single hybrid ADC unit that shares common components such as the comparator and capacitor array. This merging enables precise conversion of multi-channel bio-signals while reducing the overall device complexity and power consumption compared to using separate ADCs for each channel
Solution Approach 2:
The hybrid ADC is designed with universal functionality to handle multiple channels of bio-signals through a single unit. The shared comparator and capacitor array serve multiple conversion purposes, allowing the device to collect and process neuronal signals from multiple channels with high precision without proportionally increasing device complexity
3Use of energy by stationary object
If ultra-low power consumption is achieved through optimization, then energy efficiency is improved, but signal processing capability may be compromised
Solution Approach 1:
The device employs periodic sampling and conversion operations rather than continuous processing. The hybrid ADC converts bio-signals at optimized intervals, and digital signal processing is performed periodically on accumulated data. This periodic action significantly reduces average power consumption while maintaining adequate signal processing capability for effective brain-computer interface operation
Data Source
AI summary
A bio-signal detection apparatus includes an analog front end unit includes a plurality of analog front end circuits that process a detected signal; an analog-to-digital conversion unit that converts an output signal of the analog front end unit to a digital code and a multiplexer (MUX) that outputs a processed signal by the analog front end circuits to the analog-to-digital conversion unit; wherein the analog digital conversion unit comprises: a comparator, a successive-approximate analog-to-digital converter (SAR ADC) that converts output signal of the analog front end circuit and form MSB side j-bit of the digital code, and a single slope ADC that converts the output signal of the analog front end circuit to k bit of the digital code.