Phone-based ambient temperature sensing using opportunistic crowdsensing and machine learning

Trivedi, Amee; Bovornkeeratiroj, Phuthipong; Breda, Joseph; Shenoy, Prashant; Taneja, Jay; Irwin, David

doi:10.1016/j.suscom.2020.100479

Cited by 15 publications

(16 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Their bias dropped to 1.9°C when manual user input was allowed as well. Trivedi et al (2021) show that this bias could be reduced by using machine learning and multiple phone records (they report 0.5°F).…”

Section: Discussionmentioning

confidence: 99%

“…Hamdi et al (2020) mention the need to identify the signal-to-noise ratio in crowdsourced observations. Machine learning is often applied as a way to reduce this noise; e.g., Trivedi et al (2021) successfully use machine learning to use smartphone records for estimating indoor temperatures, and Li et al (2021) devised a bias-correction method for smartphone pressure data based on a machine learning approach. Napoly et al (2018) and Meier et al (2017) have developed a quality-control procedure for personal weather stations measuring temperature, and similar procedures have been developed for rainfall (de Vos et al, 2019) and wind observations (Droste et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Potential of a Smartphone as an Urban Weather Station—An Exploratory Analysis

Cabrera¹,

Droste²,

Heusinkveld³

et al. 2021

Front. Environ. Sci.

View full text Add to dashboard Cite

The ongoing urbanization requires enhanced understanding of the local meteorological and climatological conditions within the urban environment for multiple applications, concerning energy demand, human health, and spatial planning. Identifying areas with harmful meteorological conditions enables citizens and local authorities to take actions to optimize quality of life for urban dwellers. At the moment cities have (in general) limited networks of meteorological monitoring stations. To overcome this lack of observations, the use of non-traditional data sources is rapidly increasing. However, the use of such data sources without enough prior verification has become a controversial topic in the scientific community. This study aims to verify and assess one of the main non-traditional data sources, i.e. smartphones. The goal is to research the potential of smartphones (using the Samsung Galaxy S4 as an example phone model) to correctly sense air temperature, relative humidity, and solar radiation, and to determine to what extent environmental conditions negatively affect their performance. The smartphone readings were evaluated against observations from reference instrumentation at a weather station and a mobile measurement platform. We test the response time of the smartphone thermometer and hygrometer, and the light sensor’s cosine response. In a lab setting, we find that a smartphone can provide reliable temperature information when it is not exposed to direct solar radiation. The smartphone’s hygrometer performs better at low relative humidity levels while it can over-saturate at higher levels. The light sensor records show substantial correlation with global radiation observations, and short response times. Measurements along an urban transect of 10 km show the smartphone’s ability to react to fast changes of temperature in the field, both in time and space. However, a bias correction (dependent on wind speed and radiation) is required to represent the reference temperature. Finally we show that after such a bias correction, a smartphone record can successfully capture spatial variability over a transect as well.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The Potential of a Smartphone as an Urban Weather Station—An Exploratory Analysis

Cabrera¹,

Droste²,

Heusinkveld³

et al. 2021

Front. Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…According to the literature [12], the three financial indicator pairs of shareholders' equity/assets and liabilities, working capital/total capital, and current assets/current liabilities have the best ability to predict financial crises. e results of a study [13] comparing and analyzing four financial ratio indicators show that the current ratio and debt ratio have the best early warning effect.…”

Section: Related Workmentioning

confidence: 99%

Financial Data Center Configuration Management System Based on Random Forest Algorithm and Few-Shot Learning

Wang

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

To form a unified configuration and information management platform, FCCMS (financial center configuration management system) will integrate and sort information based on various configuration data and relationships as well as integrate processes and permissions. However, the most serious issue that data centers are currently facing is how to effectively manage these infrastructures. For various infrastructures, the data center currently uses a decentralized operation and maintenance management model. When an infrastructure fails due to inexperienced configuration management, this mode is not conducive to quickly locating and resolving the problem. A detection method of RFCO (random forest algorithm based on clustering optimization) is proposed, and an appropriate tree is selected from RF to integrate, so as to achieve the best effect. In this paper, the target matching algorithm based on FSL (few-shot learning) is deeply studied, and the target detection model is applied to the target matching and positioning task by using the ML method. The performance of the algorithm is tested by experiments on relevant datasets to verify the effectiveness of the algorithm in various scenarios.

show abstract

“…While these thermistors are designed to monitor the temperature of the device itself, they can sense changes in the temperature of the ambient environment. For example, these thermistors -primarily the thermistor present in the phone battery -have been used in the past to physically model both outdoor air temperature [19] and ambient indoor air temperature [5,11,23].…”

Section: Introductionmentioning

confidence: 99%

“…In our paper we build on this existing research to leverage these thermistors for sensing core body temperature of a user during an interaction where the device is pressed against the user's body. This new application provides 2 new constraints to the problem: (1) the process of making temperature estimates now becomes an active interaction executed by the user rather than an ambient sensing application, adding user-dependent confounds, and (2) the estimates of temperature are now made over a brief period of time during the active interaction as opposed to longitudinally as in [5,11,19,23]. On top of this, we also leverage all available thermistors on the smart device through root access as opposed to the single battery thermistors available to developers through the Android API.…”

Section: Introductionmentioning

confidence: 99%

Intuitive and Ubiquitous Fever Monitoring Using Smartphones and Smartwatches

Breda,

Patel

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Inside all smart devices, such as smartphones or smartwatches, there are thermally sensitive resistors known as thermistors which are used to monitor the temperature of the device. These thermistors are sensitive to temperature changes near their location on-device. While they are designed to measure the temperature of the device components such as the battery, they can also sense changes in the temperature of the ambient environment or thermal entities in contact with the device. We have developed a model to estimate core body temperature from signals sensed by these thermistors during a user interaction in which the user places the capacitive touchscreen of a smart device against a thermal site on their body such as their forehead. During the interaction, the device logs the temperature sensed by the thermistors as well as the raw capacitance seen by the touch screen to capture features describing the rate of heat transfer from the body to the device and device-to-skin contact respectively. These temperature and contact features are then used to model the rate of heat transferred from the user's body to the device and thus core-body temperature of the user for ubiquitous and accessible fever monitoring using only a smart device. We validate this system in a lab environment on a simulated skin-like heat source with a temperature estimate mean absolute error of 0.743 • F (roughly 0.4 • C) and limit of agreement of ±2.374 • F (roughly 1.3 • C) which is comparable to some off-the-shelf peripheral and tympanic thermometers. We found a Pearson's correlation 𝑅 2 of 0.837 between ground truth temperature and temperature estimated by our system. We also deploy this system in an ongoing clinical study on a population of 7 participants in a clinical environment to show the similarity between simulated and clinical trials.

show abstract

Phone-based ambient temperature sensing using opportunistic crowdsensing and machine learning

Cited by 15 publications

References 17 publications

The Potential of a Smartphone as an Urban Weather Station—An Exploratory Analysis

The Potential of a Smartphone as an Urban Weather Station—An Exploratory Analysis

Financial Data Center Configuration Management System Based on Random Forest Algorithm and Few-Shot Learning

Intuitive and Ubiquitous Fever Monitoring Using Smartphones and Smartwatches

Contact Info

Product

Resources

About