Low-power adaptive spike detector based on a sigma-delta control loop.

Area of Science:

• Biomedical Engineering
• Signal Processing
• Neuroscience

Background:

• Action potential (AP) detection is crucial for neural signal analysis.
• Existing methods often require significant hardware resources, limiting their use in low-power applications.
• Efficient AP detection is needed for portable and implantable neural interfaces.

Purpose of the Study:

• To develop a resources-optimized digital action potential (AP) detector.
• To implement an adaptive threshold using a novel Sigma-delta control loop.
• To evaluate the detector's performance on low-power microcontrollers (MCUs).

Main Methods:

• Designed a digital AP detector with an adaptive threshold.
• Utilized a Sigma-delta modulator-based control loop for threshold estimation.
• Implemented and tested the detector on a low-power MCU using real neural signals.
• Compared performance against existing thresholding techniques.

Main Results:

• Achieved up to 100% true positive detection rate at high signal-to-noise ratios (SNR).
• Maintained up to 63% detection rate at 0 dB SNR.
• Required minimal execution time (11 μs) on an 8 MHz MCU.
• Demonstrated suitability for low-power MCUs due to low hardware resource utilization.

Conclusions:

• The proposed Sigma-delta based adaptive threshold AP detector is highly efficient.
• It offers a competitive alternative to resource-intensive spike detection methods.
• This technology enables effective neural signal processing on resource-constrained platforms.