FlowSense

Abstract

Proper indoor ventilation through buildings' heating, ventilation, and air conditioning (HVAC) systems has become an increasing public health concern that significantly impacts individuals' health and safety at home, work, and school. While much work has progressed in providing energy-efficient and user comfort for HVAC systems through IoT devices and mobile-sensing approaches, ventilation is an aspect that has received lesser attention despite its importance. With a motivation to monitor airflow from building ventilation systems through commodity sensing devices, we present FlowSense, a machine learning-based algorithm to predict airflow rate from sensed audio data in indoor spaces. Our ML technique can predict the state of an air vent---whether it is on or off---as well as the rate of air flowing through active vents. By exploiting a low-pass filter to obtain low-frequency audio signals, we put together a privacy-preserving pipeline that leverages a silence detection algorithm to only sense sounds of from HVAC air vents when no human speech is detected. We also propose the Minimum Persistent Sensing (MPS) as a post-processing algorithm to reduce interference from ambient noise, including ongoing human conversation, office machines, and traffic noises. Together, these techniques ensure user privacy and improve the robustness of FlowSense. We validate our approach yielding over 90% accuracy in predicting vent status and 0.96 MSE in predicting airflow rate when the device is placed within 2.25 meters away from an air vent. Our approach can be generalized to environments with similar vent dimensions and geometry outlets. Additionally, we demonstrate how our approach as a mobile audio-sensing platform is robust to smartphone models, distance, and orientation. Finally, we evaluate FlowSense privacy-preserving pipeline through a user study and a Google Speech Recognition service, confirming that the audio signals we used as input data are inaudible and inconstructible.

Ambient Intelligence for Next-Generation AR

“…One important point to make, having reviewed multiple different uses of IoT sensors and actuators, is that the devices physically deployed in a given environment will likely be used for multiple purposes concurrently (e.g., images from a single IoT camera may serve as input to spatial understanding, semantic understanding, photometric registration, and visual texture estimation algorithms), so consideration will need to be given to balancing the needs of each use in these cases. Similarly, our vision for ambient IoT for AR in general, is that it is combined with the use of ambient IoT devices for other purposes; for example, IoT-based systems incorporating devices also applicable to AR have been proposed for energy efficiency [171], occupant comfort and productivity [19], surveillance [214] and building ventilation [35], while in [196] the authors consider the role of ubiquitous displays in environment aesthetics. Moreover, as AR becomes more and more integrated into our lives in the coming years, we envision that built environments, products and materials will be designed with ambient intelligence for AR in mind, in order to support the use cases we have laid out in this book chapter.…”

A Survey of AI Techniques in IoT Applications with Use Case Investigations in the Smart Environmental Monitoring and Analytics in Real-Time IoT Platform

“…In this section, we review the literature studies that presented how algorithms in auditory perception are applied in various application scenarios, such as smart cities [59][60][61], smart homes [62], and smart environments [63].…”

“…This research demonstrates the effectiveness of the integrated approach in identifying various activities of daily living through auditory analysis. Chhaglani et al in [63] presented the application of auditory perception techniques in an intelligent environment. This study focused on predicting the airflow rate in a building by analyzing the sounds associated with air flowing through a duct in a building.…”

Quantifying the Suitability of Biosignals Acquired During Surgery for Multimodal Analysis

“…In contrast, airflow sensing is another technique used to monitor respiratory status which involves the use of sensors that detect the flow of air in and out of the lungs. There are several types of airflow sensors [35] and within SIESTA airflow is measured with a sensor at the nose. The inclusion of both measurements stems from the potential to present a robust method for evaluating two distinct signals associated with the same physiological process -respiration, in this instance.…”