Spectrometer for personal context
AssigneeQUALCOMM Incorporated
InventorsEdward Harrison Teague, Ashwin Swaminathan
Life of this patent
Estimating the term…
Abstract
A method of performing context inference is described. The method includes collecting ambient light at a spectrometer sensor integrated in a portable device, characterizing the collected light to obtain optical information, comparing the optical information to optical data predetermined to match one or more contexts, inferring at least one characteristic of a specific context based on the comparison, and determining a probability that the portable device is in the specific context.
Claims27
A method of performing context inference, comprising:
- collecting ambient light at a spectrometer sensor integrated in a portable device;
- characterizing the collected light to obtain optical information;
- comparing the optical information to optical data predetermined to match one or more contexts;
- inferring at least one characteristic of a specific context based on the comparison;
- and
- determining a probability that the portable device is in the specific context.
A non-transitory machine-readable medium comprising a plurality of machine-readable instructions which, when executed by one or more processors, are adapted to cause the one or more processors to perform a method comprising:
- collecting ambient light at a spectrometer sensor integrated in a portable device;
- characterizing collected light to obtain optical information;
- comparing the optical information to optical data predetermined to match one or more contexts;
- inferring at least one characteristic of a specific context based on the comparison;
- and
- determining a probability that the portable device is in the specific context.
A method of performing context inference, comprising:
- training a machine learning system, the training comprising:
- positioning a spectrometer sensor in a specific context;
- choosing one or more desired ranges of frequencies in the specific context;
- receiving optical data from the spectrometer sensor;
- and
- linking the optical data to the specific context;
- collecting ambient light at a spectrometer sensor integrated in a portable device;
- characterizing the collected light to obtain optical information;
- and
- submitting the optical information to the machine learning system.
A mobile device comprising:
- a spectrometer sensor;
- a memory device storing optical data predetermined to match one or more contexts;
- and
- one or more processors in communication with the memory device and operable to:
- choose a desired range of frequencies to collect optical information;
- collect ambient light at the spectrometer sensor;
- characterize the collected light to obtain the optical information;
- compare the optical information to the optical data;
- infer at least one characteristic of a specific context based on the comparison;
- and
- determine a probability that the mobile device is in the specific context.
A system for performing context inference, comprising:
- means for collecting ambient light at a spectrometer sensor integrated in a portable device;
- means for characterizing the collected light to obtain optical information;
- means for comparing the optical information to optical data predetermined to match one or more contexts;
- means for inferring at least one characteristic of a specific context based on the comparison;
- and
- means for determining a probability that the portable device is in the specific context.
Dependencies are inferred from the claim text.