Systems, apparatuses and methods for adaptive noise reduction
Inventors
CENCI, IVAN • TOGNETTI, SIMONE
Assignees
Publication Number
US-11922915-B2
Publication Date
2024-03-05
Expiration Date
2035-04-16
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Abstract
An apparatus includes a sensor module configured for receiving sensed information indicative of a sensed signal. The sensed signal includes a source signal component and a source noise component. The apparatus also includes a reference module configured for reference information indicative of a reference signal. The reference signal also includes a reference noise component. The apparatus also includes a filter module configured as a fixed lag Kalman smoother. The filter module is configured for adaptively filtering the reference signal to generate an estimate of the source noise component. The apparatus also includes a processing module configured for calculating an output signal based on the sensed signal and the estimate of the source noise component. The apparatus also includes an interface module configured for transmitting an indication of the output signal. The filter module is further configured for, based on the output signal, tuning the Kalman smoother.
Core Innovation
The invention provides an apparatus configured for adaptive noise reduction by using a fixed lag Kalman smoother to filter a reference signal in order to generate an estimate of the noise component present in a sensed signal. The sensed signal contains both a source signal component and a noise component, and the reference signal includes a noise component correlated with the source noise component. The apparatus includes modules for receiving the sensed and reference signals, adaptively filtering the reference signal with a fixed lag Kalman smoother, calculating an output signal based on the sensed signal and the noise estimate, and tuning the Kalman smoother based on the output signal.
The problem addressed relates to limitations in existing adaptive noise cancellation approaches, such as limited adaptivity and lag error when removing noise from signals corrupted by additive noise or interference. Conventional adaptive filtering techniques, including least mean squares and Kalman filtering variants, may not adequately handle non-stationary signals or produce lag-free noise cancellation. The invention aims to provide systems, apparatuses, and methods for adaptive noise reduction with enhanced adaptivity and reduced or eliminated lag error by employing a fixed lag Kalman smoother configured to adaptively filter correlated reference signals and tune its parameters based on the output signal.
Claims Coverage
The claims disclose multiple inventive features directed to an apparatus and method for adaptive noise reduction using a fixed lag Kalman smoother. The main innovative features focus on the adaptive filtering technique, parameter tuning of the Kalman smoother, and signal processing specifics.
Adaptive filtering of reference signal using fixed lag Kalman smoother
The apparatus receives sensed information of a first photoplethysmographic signal including source signal and noise components, and reference information of a second photoplethysmographic signal with correlated noise. It adaptively filters the reference signal using a fixed lag Kalman smoother to estimate the source noise component.
Tuning the order of the fixed lag Kalman smoother
The Kalman smoother is configured with an even number order N, with the ability to tune this order to modify the delay introduced between the output and the reference signal, where the delay is N/2 samples.
Modifying state noise covariance based on sensed or reference signal criteria
The state noise covariance of the fixed lag Kalman smoother is modified based on sensed signal threshold or reference signal threshold, including stepwise incrementing or decrementing the covariance according to whether the signals exceed or fall below thresholds.
Modifying oblivion coefficient to adjust responsiveness
The oblivion coefficient of the fixed lag Kalman smoother is tuned based on the output signal to control responsiveness, constrained between zero and one.
Setting Kalman smoother parameters for adaptive filtering
The adaptive filtering is performed by setting the observation data to the source signal component, the observation matrix to the reference signal, and the oblivion coefficient to about one, followed by calculating state vector or tap weight estimates.
Generating noise estimate by convolution with state or tap weight estimates
The estimate of the source noise component is generated by convolving the reference signal with either a state vector estimate or tap weight estimate of the fixed lag Kalman smoother.
Calculating output signal by subtracting noise estimate
The output signal associated with the source signal component is calculated by subtracting the estimated source noise component from the sensed signal.
Including photoplethysmographic sensor and accelerometer inputs
The apparatus may include a photoplethysmographic sensor to measure the sensed information and an accelerometer to measure reference signals.
The claims collectively cover an apparatus and method implementing adaptive noise cancellation using a fixed lag Kalman smoother with configurable parameters such as order, state noise covariance, and oblivion coefficient, applied specifically to photoplethysmographic signals with correlated reference noise components, and enabling generation of a noise-reduced output signal.
Stated Advantages
Improved adaptivity in noise reduction from signals corrupted by additive noise or interference.
Reduced or eliminated lag error in the noise canceling output signal due to use of fixed lag Kalman smoother with tunable order.
Ability to tune key parameters of the Kalman smoother, including state noise covariance and oblivion coefficient, to enhance filtering performance and responsiveness.
Capability to handle non-stationary signals with sharp spikes or discontinuities more effectively than prior approaches.
Documented Applications
Removal of motion artifacts from photoplethysmographic signals.
Heart rate detection, including separating fetal ECG from maternal ECG in combined ECG signals.
Electrocardiography noise removal, such as 60 Hz noise or donor ECG interference in heart transplant patients.
Speech noise cancellation to remove ambient noise from speech signals.
Antenna design improvements, including elimination of antenna sidelobe interference.
General removal of periodic interference, such as background hum in music recordings.
Interested in licensing this patent?