Related Experiment Video
Updated: Apr 15, 2026

Studying Cavitation Enhanced Therapy
Published on: April 9, 2021
Low frequency noise elimination technique for 24-bit Σ-Δ data acquisition systems
Shao-Bo Qu1, Olivier Robert1, Philippe Lognonné1
1Institut de Physique du Globe de Paris, Sorbonne Paris Cité, Univ Paris Diderot, CNRS, F-75013 Paris, France.
Abstract:
Low frequency 1/f noise is one of the key limiting factors of high precision measurement instruments. In this paper, digital correlated double sampling is implemented to reduce the offset and low frequency 1/f noise of a data acquisition system with 24-bit sigma delta (Σ-Δ) analog to digital converter (ADC). The input voltage is modulated by cross-coupled switches, which are synchronized to the sampling clock, and converted into digital signal by ADC. By using a proper switch frequency, the unwanted parasitic signal frequencies generated by the switches are avoided. The noise elimination processing is made through the principle of digital correlated double sampling, which is equivalent to a time shifted subtraction for the sampled voltage. The low frequency 1/f noise spectrum density of the data acquisition system is reduced to be flat down to the measurement frequency lower limit, which is about 0.0001 Hz in this paper. The noise spectrum density is eliminated by more than 60 dB at 0.0001 Hz, with a residual noise floor of (9 ± 2) nV/Hz(1/2) which is limited by the intrinsic white noise floor of the ADC above its corner frequency.
Related Concept Videos
Upsampling
Aliasing
If the sampling frequency is below the Nyquist rate, these replicas overlap, preventing the original...
Downsampling
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...
Sampling Continuous Time Signal
In the...
Reconstruction of Signal using Interpolation
Sampling Theorem

