Abstract: A method for monitoring air quality is described. The method includes measuring environmental components at multiple locations using multiple mobile sensor platforms to provide sensor data. The environmental components include particulate matter having a size range and ambient gases. The sensor data includes particulate matter data having the size range and ambient gas data captured at the plurality of locations. The method also includes determining a signature based on the particulate matter data including the size range, and at least one additional factor. The at least one additional factor includes the ambient gas data. The method also includes identifying a source based on the signature.

Abstract: Embodiments for implementing a softmax function in an analog circuit. The analog circuit may comprise a plurality of input nodes to accept voltage inputs; a plurality of diodes connected to each of the plurality of input nodes to perform a current adding function; a log amplifier coupled to the plurality of diodes; a plurality of analog adders coupled to the voltage inputs and an output of the log amplifier; and a plurality of exponential amplifiers, each of the plurality of exponential amplifiers coupled to one of the plurality of analog adders.

Abstract: The invention relates to a device for measuring, in near-real-time, the level of black carbon, brown carbon, organic carbon, total carbon and CO2 in air. The device also provides for a direct calculation of aerosol angstrom coefficient as well as estimation of emissions rates of black carbon or brown carbon from nearby combustion sources.

Abstract: A CMOS-based resistive processing unit (RPU) for a neural network. The RPU includes a capacitor device configured to store a charge representing a weight value associated with a neural network circuit operation. A current source Field Effect Transistor (FET) device is operatively connected to the capacitor device for increasing a charge on the capacitor. A current sink FET device operatively connected to the capacitor device is configured for decreasing the stored capacitor charge. An analog weight update circuit receives one or more update signals generated in conjunction with the neural network circuit operation, the analog weight update circuit controlling the current source FET device and the current sink FET device to provide either a determined amount of current to increase the stored charge on the capacitor device, or sink a determined amount of current to decrease the stored charge on the capacitor device.

Abstract: A CMOS-based resistive processing unit (RPU) and method for a neural network. The RPU includes a capacitor device configured to store a charge representing a weight value associated with a neural network circuit operation. A current source Field Effect Transistor (FET) device is operatively connected to the capacitor device for increasing a charge on the capacitor. A current sink FET device operatively connected to the capacitor device is configured for decreasing the stored capacitor charge. An analog weight update circuit receives one or more update signals generated in conjunction with the neural network circuit operation, the analog weight update circuit controlling the current source FET device and the current sink FET device to provide either a determined amount of current to increase the stored charge on the capacitor device, or sink a determined amount of current to decrease the stored charge on the capacitor device.

Abstract: The method of the invention provides a simple, reproducible reference, or set of reference standards, as a validation of a calibration standard for filter-based measurements of light absorbing aerosol particles (black carbon and brown carbon) and organic carbon and elemental carbon particles found in indoor and outdoor environments, and in emission streams. The set of reference standards comprise grayscale images deposited by a mist or print at a known density on a substrate material. The reference standards may be made by a printer using as a substrate either paper or another material such as glass fiber, quartz fiber, polypropylene, or cellulose.

Abstract: A field-effect transistor (FET) and method of manufacture thereof include patterning a mask above a source and drain of a FET to form holes in the mask, growing epitaxial structures from the holes in the mask, and growing a doped epitaxial shell to coat sidewalls of the epitaxial structures.

Abstract: Embodiments for implementing a softmax function in an analog circuit. The analog circuit may comprise a plurality of input nodes to accept voltage inputs; a plurality of diodes connected to each of the plurality of input nodes to perform a current adding function; a log amplifier coupled to the plurality of diodes; a plurality of analog adders coupled to the voltage inputs and an output of the log amplifier; and a plurality of exponential amplifiers, each of the plurality of exponential amplifiers coupled to one of the plurality of analog adders.

Abstract: A field-effect transistor (FET) and method of manufacture thereof include patterning a mask above a source and drain of a FET to form holes in the mask, growing epitaxial structures from the holes in the mask, and growing a doped epitaxial shell to coat sidewalls of the epitaxial structures.

Abstract: A field-effect transistor (FET) and method of manufacture thereof include a gate, a doped semiconductor structure formed on top of the planar source and drain regions, and a sheath of conducting materials flanking the formed doped semiconductor structure, where the sheath is perpendicular to a surface of the planar source and drain regions.

Abstract: Microfabricated PM sensors measure concentrations of particulate matter (PM) in air. Some sensors improve the accuracy of measurements by accounting for the effect of ambient conditions (e.g., temperature or humidity) on mass-sensitive elements employed to determine a mass of the PM in a stream of air. Some sensors improve the accuracy of measurements by controlling humidity in the stream of air measured by mass-sensitive elements. Some sensors employ a plurality of mass-sensitive elements to extend the useful life of the PM sensor. Some sensors employ one or more mass-sensitive elements and heating elements to cause deposition and allow measurement of different sizes of PM. Some sensors can measure mass concentration of coarse PM in addition to fine PM in a stream of air. Some sensors control the flow rate of a stream of air measured by mass-sensitive elements. Some sensors include features to mitigate electromagnetic interference or electromagnetic signal loss.

Abstract: The invention relates to a device for measuring, in near-real-time, the level of black carbon, brown carbon, organic carbon, total carbon and CO2 in air. The device also provides for a direct calculation of aerosol angstrom coefficient as well as estimation of emissions rates of black carbon or brown carbon from nearby combustion sources.

Abstract: A field-effect transistor (FET) and method of manufacture thereof include a gate, a pillar of grown on a top of the planar source and drain regions, and a conductive sheath flanking the pillar, the sheath is bent up, alongside, and over the gate.

Abstract: The method of the invention provides a simple, reproducible reference, or set of reference standards, as a validation of a calibration standard for filter-based measurements of light absorbing aerosol particles (black carbon and brown carbon) and organic carbon and elemental carbon particles found in indoor and outdoor environments, and in emission streams. The set of reference standards comprise grayscale images deposited by a mist or print at a known density on a substrate material. The reference standards may be made by a printer using as a substrate either paper or another material such as glass fiber, quartz fiber, polypropylene, or cellulose.

Abstract: A memory device including a plurality of memory unit cells arranged in a crossbar configuration for a neural network is provided. Each of the memory unit cells includes a readout transistor, a charging transistor, a discharging transistor, and a metal-insulator-metal (MIM) capacitor connected to one of source/drain regions of each of the charging transistor and the discharging transistor and a functional gate of the readout transistor for storing analog information.

Abstract: A CMOS-based resistive processing unit (RPU) for a neural network. The RPU includes a capacitor device configured to store a charge representing a weight value associated with a neural network circuit operation. A current source Field Effect Transistor (FET) device is operatively connected to the capacitor device for increasing a charge on the capacitor. A current sink FET device operatively connected to the capacitor device is configured for decreasing the stored capacitor charge. An analog weight update circuit receives one or more update signals generated in conjunction with the neural network circuit operation, the analog weight update circuit controlling the current source FET device and the current sink FET device to provide either a determined amount of current to increase the stored charge on the capacitor device, or sink a determined amount of current to decrease the stored charge on the capacitor device.

Abstract: A CMOS-based resistive processing unit (RPU) and method for a neural network. The RPU includes a capacitor device configured to store a charge representing a weight value associated with a neural network circuit operation. A current source Field Effect Transistor (FET) device is operatively connected to the capacitor device for increasing a charge on the capacitor. A current sink FET device operatively connected to the capacitor device is configured for decreasing the stored capacitor charge. An analog weight update circuit receives one or more update signals generated in conjunction with the neural network circuit operation, the analog weight update circuit controlling the current source FET device and the current sink FET device to provide either a determined amount of current to increase the stored charge on the capacitor device, or sink a determined amount of current to decrease the stored charge on the capacitor device.

Abstract: A memory device including a plurality of memory unit cells arranged in a crossbar configuration for a neural network is provided. Each of the memory unit cells includes a readout transistor, a charging transistor, a discharging transistor, and a metal-insulator-metal (MIM) capacitor connected to one of source/drain regions of each of the charging transistor and the discharging transistor and a functional gate of the readout transistor for storing analog information.

Abstract: A memory device including a plurality of memory unit cells arranged in a crossbar configuration for a neural network is provided. Each of the memory unit cells includes a readout transistor, a charging transistor, a discharging transistor, and a metal-insulator-metal (MIM) capacitor connected to one of source/drain regions of each of the charging transistor and the discharging transistor and a functional gate of the readout transistor for storing analog information.

Abstract: A multi-filter chemical speciation sampler and a virtual impaction particle separation inlet therefore are provided. The inlet includes a housing having a bottom, a collection tube that extends through the bottom, and collection apertures formed in the bottom, arranged around the collection tube; a first plate disposed on top of the housing, having acceleration nozzles disposed at the perimeter thereof; a second plate disposed in the housing below the first plate, having a central aperture and separation apertures disposed around the central aperture. The sampler includes: an inlet; a virtual impaction separator to further fractionate the PM into a course fraction and a fine fraction; a first separation assembly to divide the course fraction into coarse aliquots, comprising first filters to collect the coarse aliquots; a second separation assembly to divide the fine fraction into fine aliquots, comprising second filters to collect the fine aliquots.