Methods and systems for ascertaining factors contributing to the temperature of a device

An on-device method for estimating ambient temperature of a device, the method comprising:

• monitoring, by the device, a plurality of parameters comprising a battery temperature of a battery of the device, processor, thermal zones, voltage, current, instantaneous Central Processing Unit (CPU) load, average CPU load, Graphics Processing Unit (GPU) usage, and screen brightness;
• monitoring, by the device, a plurality of external parameters comprising lux, Wireless-Fidelity (Wi-Fi) signal strength, and network signal strength;
• determining, by a regression neural network in the device, a temperature representing a result of subtraction of the ambient temperature of the device from the battery temperature, based on the plurality of parameters and the plurality of external parameters;
• determining, by the device, the ambient temperature of the device based on the determined temperature and the battery temperature;
• determining, by a classifier neural network, whether a cause of overheating of the device is due to one of internal factors and external factors;
• classifying the cause of overheating of the device due to the internal factors when the battery temperature is greater than the ambient temperature of the device;
• and
• classifying the cause of overheating of the device due to the external factors, when the battery temperature is less than the ambient temperature of the device.

A device for determining a cause of overheating of the device, the device comprising:

• a neural network processor including a classifier neural network configured to:
• monitor a plurality of parameters that are contributing to a temperature of the device;
• estimate a degree of contribution of internal factors to the temperature of the device and a degree of contribution of external factors to the temperature of the device, based on the monitored plurality of parameters and a battery temperature of a battery of the device;
• and
• determine whether the degree of contribution of the internal factors to the temperature of the device is greater than the degree of contribution of the external factors to the temperature of the device, based on whether the battery temperature is greater than an ambient temperature of the device;
• wherein when the battery temperature is greater than the ambient temperature of the device, then the classifier neural network is configured to determine that the degree of contribution of the internal factors to the temperature of the device is greater than the degree of contribution of the external factors to the temperature of the device;
• and
• wherein when the battery temperature is less than the ambient temperature of the device, then the classifier neural network is configured to determine that the degree of contribution of the external factors to the temperature of the device is greater than the degree of contribution of the internal factors to the temperature of the device.

A device for estimating ambient temperature of the device, the device comprising:

• a neural network processor including a classifier neural network, wherein the neural network processor is configured to:
• monitor a plurality of parameters comprising a battery temperature of a battery of the device, processor, thermal zones, voltage, current, instantaneous Central Processing Unit (CPU) load, average CPU load, Graphics Processing Unit (GPU) usage, and screen brightness;
• monitor a plurality of external parameters comprising lux, Wireless-Fidelity (Wi-Fi) signal strength, and network signal strength;
• determining, by a regression neural network in the device, a temperature representing a result of subtraction of the ambient temperature of the device from the battery temperature, based on the plurality of parameters and the plurality of external parameters;
• determining, by the device, the ambient temperature of the device based on the determined temperature and the battery temperature;
• determine whether a cause of overheating of the device is due to one of internal factors and external factors;
• classify the cause of overheating of the device due to the internal factors when the battery temperature is greater than the ambient temperature of the device;
• and
• classify the cause of overheating of the device due to the external factors, when the battery temperature is less than the ambient temperature of the device.