Qualcomm Incorporated

Piero Zappi

Ashwin Deepak Swaminathan

Santiago Mazuelas

Christopher G Lott

Roy Franklin Quick, Jr

Shankar Sadasivam

Jin Won Lee

Aditya Narain Srivastava

Rexford Alan Hill

Abhijeet Bisain

Suhas H Sheshadri

Muralidhar Reddy Akula

Suhas Hariharapura Sheshadri

A method performed by a computing device includes determining a partition for depth-first processing by a multi-layer artificial neural network (ANN) of the computing device. The computing device comprising a processor, on-chip memory, and off-chip memory. The first partition determined based on an amount of on-chip memory used by the first partition, an available amount of on-chip memory, and a size of a write back to the off-chip memory. The method also includes processing, at the device via the multi-layer ANN, an input, using the depth-first processing in accordance with the partition.

Depth-first deep convolutional neural network inference

A method of auto-calibrating light sensor data of a mobile device includes, obtaining, by the mobile device, one or more reference parameters representative of light sensor data collected by a reference device. The method also includes collecting, by the mobile device, light sensor data from a light sensor included in the mobile device, itself. One or more sample parameters of the light sensor data obtained from the light sensor included in the mobile device are then calculated. A calibration model is then determined for auto-calibrating the light sensor data of the light sensor included in the mobile device based on the one or more reference parameters and the one or more sample parameters.

Auto-calibrating light sensor data of a mobile device

A detector in a mobile device receives input from a modem, determines whether the mobile device is indoor or outdoor based on the modem-supplied input, and stores in memory a binary value to indicate an indoor-outdoor state. In some embodiments, the detector extracts a feature from the modem-supplied input, and uses the extracted feature with a statistical classifier, to output an indoor-outdoor state and an associated probability of correctness of the indoor-outdoor state. In other embodiments, the detector determines whether the mobile device is indoor or outdoor based at least on the feature extracted to characterize temporal distribution of the wireless signal, by using a prior value of the state being indoor or outdoor.

Power delay profile based indoor outdoor detection

Growing community of inventors

San Diego, CA, United States of America

Piero Zappi