Occupancy Estimation from Blurred Video: A Multifaceted Approach with Privacy Consideration.

Building occupancy information is significant for a variety of reasons, from allocation of resources in smart buildings to responding during emergency situations. As most people spend more than 90% of their time indoors, a comfortable indoor environment is crucial. To ensure comfort, traditional HVAC systems condition rooms assuming maximum occupancy, accounting for more than 50% of buildings' energy budgets in the US. Occupancy level is a key factor in ensuring energy efficiency, as occupancy-controlled HVAC systems can reduce energy waste by conditioning rooms based on actual usage. Numerous studies have focused on developing occupancy estimation models leveraging existing sensors, with camera-based methods gaining popularity due to their high precision and widespread availability. However, the main concern with using cameras for occupancy estimation is the potential violation of occupants' privacy. Unlike previous video-/image-based occupancy estimation methods, we addressed the issue of occupants' privacy in this work by proposing and investigating both motion-based and motion-independent occupancy counting methods on intentionally blurred video frames. Our proposed approach included the development of a motion-based technique that inherently preserves privacy, as well as motion-independent techniques such as detection-based and density-estimation-based methods. To improve the accuracy of the motion-independent approaches, we utilized deblurring methods: an iterative statistical technique and a deep-learning-based method. Furthermore, we conducted an analysis of the privacy implications of our motion-independent occupancy counting system by comparing the original, blurred, and deblurred frames using different image quality assessment metrics. This analysis provided insights into the trade-off between occupancy estimation accuracy and the preservation of occupants' visual privacy. The combination of iterative statistical deblurring and density estimation achieved a 16.29% counting error, outperforming our other proposed approaches while preserving occupants' visual privacy to a certain extent. Our multifaceted approach aims to contribute to the field of occupancy estimation by proposing a solution that seeks to balance the trade-off between accuracy and privacy. While further research is needed to fully address this complex issue, our work provides insights and a step towards a more privacy-aware occupancy estimation system.

Dynamic Arctangent Center-Tracking Method for Respiratory Displacement Monitoring of Subjects in Arbitrary Positions.

Accurate continuous measurement of respiratory displacement using continuous wave Doppler radar requires rigorous management of dc offset which changes when a subject changes distance from the radar measurement system. Effective measurement, therefore, requires robust dynamic calibration which can recognize and compensate for changes in the nominal position of a subject. In this paper, a respiratory displacement measurement algorithm is proposed which can differentiate between sedentary and non-sedentary conditions and continuously adapt to provide long-term monitoring of a subject's sedentary respiration. Arctangent demodulation is an effective means of quantifying continuous displacement using a quadrature Doppler radar, yet it depends on accurate identification of dc offset and dc information contributions in the radar I-Q arc with the subject in a particular position. The dynamic calibration method proposed here is demonstrated to differentiate between sedentary and non-sedentary conditions for six subjects to produce accurate sedentary respiration measurements even when the subject arbitrarily changes position, once the appropriate thresholds are established for the measurement environment.

Doppler radar remote sensing of respiratory function.

Doppler radar remote sensing of torso kinematics can provide an indirect measure of cardiopulmonary function. Motion at the human body surface due to heart and lung activity has been successfully used to characterize such measures as respiratory rate and depth, obstructive sleep apnea, and even the identity of an individual subject. For a sedentary subject, Doppler radar can track the periodic motion of the portion of the body moving as a result of the respiratory cycle as distinct from other extraneous motions that may occur, to provide a spatial temporal displacement pattern that can be combined with a mathematical model to indirectly assess quantities such as tidal volume, and paradoxical breathing. Furthermore, it has been demonstrated that even healthy respiratory function results in distinct motion patterns between individuals that vary as a function of relative time and depth measures over the body surface during the inhalation/exhalation cycle. Potentially, the biomechanics that results in different measurements between individuals can be further exploited to recognize pathology related to lung ventilation heterogeneity and other respiratory diagnostics.

Unobtrusive occupancy and vital signs sensing for human building interactive systems.

Cognitive buildings use data on how occupants respond to the built environment to proactively make occupant-centric adjustments to lighting, temperature, ventilation, and other environmental parameters. However, sensors that unobtrusively and ubiquitously measure occupant responses are lacking. Here we show that Doppler-radar based sensors, which can sense small physiological motions, provide accurate occupancy detection and estimation of vital signs in challenging, realistic circumstances. Occupancy was differentiated from an empty room over 93% of the time in a 3.4 m × 8.5 m conference room with a single sensor in both wall and ceiling-mounted configurations. Occupancy was successfully detected while an occupant was under the table, visibly blocked from the sensor, a scenario where infrared, ultrasound, and video-based occupancy sensors would fail. Heart and respiratory rates were detected in all seats in the conference room with a single ceiling-mounted sensor. The occupancy sensor can be used to control HVAC and lighting with a short, 1-2 min delay and to provide information for space utilization optimization. Heart and respiratory rate sensing could provide additional feedback to future human-building interactive systems that use vital signs to determine how occupant comfort and wellness is changing with time.

Identification of COVID-19 Type Respiratory Disorders Using Channel State Analysis of Wireless Communications Links.

One deadly aspect of COVID-19 is that those infected can often be contagious before exhibiting overt symptoms. While methods such as temperature checks and sinus swabs have aided with early detection, the former does not always provide a reliable indicator of COVID-19, and the latter is invasive and requires significant human and material resources to administer. This paper presents a non-invasive COVID-19 early screening system implementable with commercial off-the-shelf wireless communications devices. The system leverages the Doppler radar principle to monitor respiratory-related chest motion and identifies breathing rates that indicate COVID-19 infection. A prototype was developed from software-defined radios (SDRs) designed for 5G NR wireless communications and system performance was evaluated using a robotic mover simulating human breathing, and using actual breathing, resulting in a consistent respiratory rate accuracy better than one breath per minute, exceeding that used in common medical practice.Clinical Relevance-This establishes the potential efficacy of wireless communications based radar for recognizing respiratory disorders such as COVID-19.

Separation of Respiratory Signatures for Multiple Subjects Using Independent Component Analysis with the JADE Algorithm.

Respiration monitoring using microwave Doppler radar has attracted significant interest over the last four decades due to its non-invasive and non-contact form of measurement. However, this technology is still not at the level of practical implementations in healthcare due to motion artifacts and interference from multiple subjects within the range of the Doppler radar sensor. Most reported results in literature focus only on single subject measurements because when multiple subjects are present there are interfering respiration signals which are difficult to separate as individual respiration signals. This paper investigates the feasibility of separating respiratory signatures from the multiple subjects. We employed a new approach using Independent Component Analysis (ICA) with the Joint Approximate Diagonalization of Eignematrices (JADE) algorithm to achieve this for closely spaced subjects, and the system is also capable of estimating Direction of Arrival (DOA) for well-spaced subjects. Experimental results demonstrated that the ICA-JADE method can separate respiratory signatures from two subjects one meter apart from each other at a distance from the radar of 2.89 meters. The separated respiratory pattern closely correlates with reference chest belt respiration patterns, and the mean square error is approximately 11.58%. Concisely, this paper clearly demonstrates that by integrating ICA with the JADE algorithm in a Doppler radar physiological monitoring system, multiple subjects can be monitored simultaneously.

Feasibility assessment of Doppler radar long-term physiological measurements.

In this paper we examine the feasibility of applying doppler radar technique for a long-term health monitoring. Doppler radar was used to detect and eliminate periods of significant motion. This technique was verified using a human study on 17 subjects, and it was determined that for 15 out of 17 subjects there was no significant motion for over 85% of the measurement interval in supine positions. Majority of subjects exhibited significantly less motion in supine position, which is promising for sleep monitoring, and monitoring of hospitalized patients.

Non-contact respiratory rate measurement validation for hospitalized patients.

This paper presents the first clinical results for validating the accuracy of respiratory rate obtained for hospitalized patients using a non-contact, low power 2.4 GHz Doppler radar system. Twenty-four patients were measured in this study. The respiratory rate accuracy was benchmarked against the respiratory rate obtained using Welch Allyn Propaq Encore model 242, the Embla Embletta system with Universal XactTrace respiratory effort sensor and Somnologica for Embletta software, and by counting chest excursions. The 95% limits of agreement between the Doppler radar and reference measurements fall within +/-5 breaths per minute.

Radar cross section of human cardiopulmonary activity for recumbent subject.

The radar cross section (RCS) corresponding to human cardio-respiratory motion is measured for a subject in two different recumbent positions. Lying face-up (supine), the subject showed an RCS of 0.326 m(2). But when lying face-down (prone), the RCS increased to 2.9 m(2). This is the first reported RCS measurement corresponding to human cardio-respiratory motion. The results obtained in this experiment suggest modeling the upper part of the human body as a half-cylinder where the front body corresponds to the cylindrical surface and the back corresponds to the rectangular one.

Microwave non-invasive sensing of respiratory tidal volume.

This paper describes the use of Doppler radar to measure respiration rate and air volume. The respiratory volume is measured indirectly via chest wall position. Calibration of displacement to airflow prior to subject measurements and accurate chest wall position information enable mean differences of less than 10 ml; with standard deviation of the difference of 20 ml between radar and reference measurements.

DC information preservation for cardiopulmonary monitor utilizing CW Doppler radar.

Direct conversion RF receivers introduce large DC offsets, reducing the dynamic range of the baseband signal. Coupled with the relatively small time varying signals in human vital sign monitoring using CW Doppler radar, extraction of cardio-pulmonary information becomes difficult. Previous DC offset compensation techniques utilizing AC coupling have proven detrimental to the performance of the system and the integrity of the low-frequency cardiopulmonary signals. A proposed system utilizing digitally controlled voltage feedback and center finding preserves the important DC information for optimal extraction of phase information in the quadrature system.

Feasibility of heart rate variability measurement from quadrature Doppler radar using arctangent demodulation with DC offset compensation.

This paper describes the experimental results of the beat-to-beat interval measurement from a quadrature Doppler radar system utilizing arctangent demodulation with DC offset compensation techniques. The comparison in SDNN and in RMSDD of both signals demonstrates the potential of using quadrature Doppler radar for HRV analysis.

Data acquisition system for Doppler radar vital-sign monitor.

Automatic gain control (AGC) units increase the dynamic range of a system to compensate for the limited dynamic range of analog to digital converters. This problem is compounded in wireless systems in which large changes in signal strength are effects of a changing environment. These issues are evident in the direct-conversion Doppler radar vital-sign monitor. Utilizing microwave radar signals reflecting off a human subject, a two-channel quadrature receiver can detect periodic movement resulting from cardio-pulmonary activity. The quadrature signal is analyzed using an arctangent demodulation that extracts vital phase information. A data acquisition (DAQ) system is proposed to deal with issues inherent in arctangent demodulation of a quadrature radar signal.

Non-contact cardiopulmonary sensing with a baby monitor.

Cardiopulmonary signals can be detected at a distance using simple Doppler radars operating in CW mode. Tests with and without audio modulation show the feasibility of measurements with this hardware, providing a maximum measured difference to the reference of just 1.6bpm for heart rate. Tests show good correspondence of heart/respiration rate with the reference data.