International Business Machines Corporation

The University of Melbourne

Technische Universiteit Eindhoven

Hgst Netherlands, B.v.

Andrea Fasoli

Luisa Dominica Bozano

Linda Karin Sundberg

Krystelle Lionti

Mohammed Abdi

Alberto Mannari

Aminat Adebiyi

Robert Dennis Miller

Ronald Robert Labby

Amy N Bowers

Tom Van Der Pol

Elizabeth M Lofano

Thomas Geoffrey Pattison

Kyu-hyoun Kim

Nimrod Megiddo

Chia-Yu Chen

Pawan Raghunath Chowdhary

Ankur Agrawal

Qing Zhu

Simone Bianco

Hitesh Arora

Siddhartha Das

Vito Paolo Pastore

Abubeker Abdullahi

An apparatus includes multiple analog to digital converters. Individual analog to digital converters are configured to produce a digital output from an analog input and configured to compute a least significant bit of the digital output by comparing an internal residual voltage for determination of the least significant bit and a residual voltage from another analog to digital converter.

Stochastic rounding switched capacitor computation cores useful for efficient deep learning inference

A small-sized, portable enclosure protects a gas sensor against degradation due to environmental exposure and changes in atmospheric conditions. The protective enclosure includes an inlet for introduction of a gas into the enclosure, an outlet for release of the gas upon completion of a sensing run, and at least one retractable filter that removes from the inflowing gas deleterious compounds that can compromise the integrity of the sensor or cause the sensor to degrade over time. The enclosure does not include any filters during the measurement phase of the sensing run in order to allow the gas sensor to accurately measure an unmodified gas mixture and/or analyte.

Protective enclosure for gas sensors

Volatile organic compounds classification by receiving test data associated with detecting volatile organic compounds (VOCs), analyzing the test data according to a set of data features associated with known VOCs, determining a match between each feature of the test data and a corresponding feature of the set of data features, yielding a set of matches, defining a first degree of anomaly for the test data according to the set of matches, and classifying the test data according to the first degree of anomaly.

Volatile organic compound detection and classification

A method of identifying a gas is provided. The method includes providing a gas sensor device comprising at least two stacked metal oxide layers, wherein a change in conductance of the gas sensor device in a presence of a gas varies with a temperature of the stacked metal oxide layers. The method includes bringing the gas into proximity with the stacked metal oxide layers. The method also includes measuring the conductance of the gas sensor device when the gas is in proximity with the stacked layers at multiple temperatures to generate a temperature-conductance profile. The method also includes identifying a gas of interest based on the temperature-conductance profile.

Multifunctional heterojunction metal oxide gas sensor

A small-sized, portable enclosure protects a gas sensor against degradation due to environmental exposure and changes in atmospheric conditions. The protective enclosure includes an inlet for introduction of a gas into the enclosure, an outlet for release of the gas upon completion of a sensing run, and a number of in-line filters that remove from the inflowing gas sample analytes, contaminants, and other materials that can compromise the integrity of the sensor or cause the sensor to degrade over time. The enclosure does not include any filters during the measurement phase of the sensing run in order to allow the gas sensor to accurately measure an unmodified gas mixture and/or analyte.

Provided is a system and method for tuning an array of sensors to enable selection of the most suitable sensors for a target application. After extracting features from sensor raw data, the extracted features are ranked with gradient boosting decision trees to assign an importance value to each extracted feature. A threshold value for the entire set of extracted features is calculated and an importance score is calculated for the individual sensors of the array. Individual sensors with an importance score on or above the threshold value are selected for the target application.

Sensor tuning&#x2014;sensor specific selection for IoT&#x2014;electronic nose application using gradient boosting decision trees

Provided is a gas sensor system with a gas sensor, and a microprocessor programmed to control the gas sensor with at least two operational modes. The first operational mode controls the gas sensor from a baseline level through analyte detection. Upon initiation of the recovery phase after analyte withdrawal, the gas sensor system switches to the second operational mode, which changes conditions of the gas sensor to (i) accelerate removal of the analyte from the gas sensor and (ii) accelerate recovery of the gas sensor output towards the baseline level. When no further analyte is detected, the gas sensor switches back to the first operational mode or to an additional operational mode to complete recovery.

Adaptive sensor temperature control for fast recovery

Provided is a method and system for extracting features from raw data for machine learning processing. Using an array of gas sensors, raw data for at least one compound of interest are extracted based upon the type of output signals. Where the output signals are amplitude-variant, mean features are extracted by chunking the raw data into slices and calculating the mean area under the curve. Where the output signals are amplitude-and-time-variant, mean-plus-slope features are extracted by taking logarithmic values of the raw data and calculating the mean area under the curve.

Feature tuning&#x2014;application dependent feature type selection for improved classification accuracy

Provided is a system for detection and discrimination of gases, such as volatile organic compounds (VOCs), in a sample comprising two chambers, a first chamber equipped with a crystalline microporous material (CMM) and a second chamber equipped with a gas detector. A gas sample is introduced first into the second chamber for detection by the gas detector and is then re-routed to the first chamber for adsorption/desorption on the CMM. The gas detector in the second chamber produces electronic signals that correspond to the adsorption/desorption profile for the gas, which allows for discrimination of the gas in the sample from other possible gas samples.

Adsorption/desorption-based sensor for volatile organic compounds (VOCs)

Initial film layers prepared from tin(II) chloride spontaneously generate open cavities when the initial film layers are thermally cured to about 400&#xb0; C. using a temperature ramp of 1&#xb0; C./minute to 10&#xb0; C./minute while exposed to air. The openings of the bowl-shaped cavities have characteristic dimensions whose lengths are in a range of 30 nm to 300 nm in the plane of the top surfaces of the cured film layers. The cured film layers comprise tin oxide and have utility in gas sensors, electrodes, photocells, and solar cells.

Porous tin oxide films

A perpendicular magnetic recording (PMR) disk has a patterned template layer for the growth of the magnetic grains and the nonmagnetic material surrounding the grains. The template layer is a substantially planar platinum (Pt) or palladium (Pd) layer that is patterned to have Pt or Pd regions arranged in a hexagonal-close-packed (hcp) pattern with the Pt or Pd regions surrounded by Pt-oxide or Pd-oxide regions. The two separate regions of the template layer have different surface chemistries and energies, which provide a 'chemical contrast' to impinging atoms during deposition of the metallic magnetic material and nonmagnetic (typically oxide) material, effectively guiding the deposition. The metallic magnetic material is preferentially deposited on the pristine, epitaxial Pt or Pd regions to form the magnetic grains, while the oxide migrates to the oxidized Pt or Pd regions due to the matching of lower surface energy.

Growing community of inventors

San Jose, CA, United States of America

Andrea Fasoli