Proximity Detection During Epidemics: Direct UWB TOA Versus Machine Learning Based RSSI.

In this paper, we compare the direct TOA-based UWB technology with the RSSI-based BLE technology using machine learning algorithms for proximity detection during epidemics in terms of complexity of implementation, availability in existing smart phones, and precision of the results. We establish the theoretical limits on the precision and confidence of proximity estimation for both technologies using the Cramer Rao Lower Bound (CRLB) and validate the theoretical foundations using empirical data gathered in diverse practical operating scenarios. We perform our empirical experiments at eight distances in three flat environments and one non-flat environment encompassing both Line of Sight (LOS) and Obstructed-LOS (OLOS) situations. We also analyze the effects of various postures (eight angles) of the person carrying the sensor, and four on-body locations of the sensor. To estimate the range with BLE RSSI, we use 14 features for training the Gradient Boosted Machines (GBM) learning algorithm and we compare the precision of results with those obtained from memoryless UWB TOA ranging algorithm. We show that the memoryless UWB TOA algorithm achieves 93.60% confidence, slightly outperforming the 92.85% confidence of the BLE RSSI with more complex GBM machine learning (ML) algorithm and the need for substantial training. The training process for the RSSI-based BLE social distance measurements involved 3000 measurements to create a training dataset for each scenario and post-processing of data to extract 14 features of RSSI, and the ML classification algorithm consumed 200 s of computational time. The memoryless UWB ranging algorithm achieves more robust results without any need for training in less than 0.5 s of computation time.

Understanding of RF Cloud Interference Measurement and Modeling.

Importance of spectrum regulation and management was first revealed on May of 1985 after the release of unlicensed ISM bands resulting in emergence of Wi-Fi, Bluetooth and many other wireless technologies that has affected our daily lives by enabling the emergence of the smart world and IoT era. Today, the idea of a liberated spectrum is circulating around, which can potentially direct wireless networking industry into another revolution by enabling a new paradigm in intelligent spectrum regulation and management. The RF signal radiated from IoT devices as well as other wireless technologies create an RF cloud causing co- and cross-interference to each other. Lack of a science and technology for understanding, measurement, and modeling of the RF cloud interference in near real-time results in inefficient utilization of the precious spectrum, a unique natural resource shared among all wireless devices of the universe in frequency, time, and space. Near real time forecasting of the RF cloud interference is essential to pursue the path to the optimal utilization of spectrum and a liberated spectrum management. This paper presents a historical perspective on the evolution of spectrum regulation and management, explains the diversified meanings of interference for different sectors of the wireless industry, and presents a path for implementing a theoretical foundation for interference monitoring and forecasting to enable the emergence of a liberated spectrum industry and a new paradigm in spectrum management and regulations.

Machine Learning Estimation of COVID-19 Social Distance using Smartphone Sensor Data.

Airborne infectious diseases such as COVID-19 spread when healthy people are in close proximity to infected people. Technology-assisted methods to detect proximity in order to alert people are needed. In this work we systematically investigating Machine Learning (ML) methods to detect proximity by analyzing data gathered from smartphones' built-in Bluetooth, accelerometer and gyroscope sensors. We extracted 20 statistical features from raw sensor data, which were then classified (< 6ft or not) and regressed (distance estimate) using ML algorithms. We found that elliptical filtering of accelerometer and gyroscope sensors signal improved the performance of ML regression. The most predictive features were z-axis mean and fourth momentum for the accelerometer sensors, z-axis mean y-axis mean for the gyroscope sensor, and advertiser time and mean RSSI for Bluetooth radio. After rigorous evaluation of the performance of 19 ML classification and regression methods, we found that ensemble (boosted and bagged tree) methods and regression trees ML algorithms performed best when using data from a combination of Bluetooth radio, accelerometer and the gyroscope. We were able to classify proximity (< 6ft or not) with 100% accuracy using the accelerometer sensor and with 62%-97% accuracy with the Bluetooth radio.

Performance Evaluation of COVID-19 Proximity Detection Using Bluetooth LE Signal.

The risk of COVID-19 transmission increases when an uninfected person is less than 6 ft from an infected person for longer than 15 minutes. Infectious disease experts working on the COVID-19 pandemic call this high-risk situation being Too Close for Too Long (TCTL). Consequently, the problem of detecting the TCTL situation in order to maintain appropriate social distance has attracted considerable attention recently. One of the most prominent TCTL detection ideas being explored involves utilizing the Bluetooth Low-Energy (BLE) Received Signal Strength Indicator (RSSI) to determine whether the owners of two smartphones are observing the acceptable social distance of 6 ft. However, using RSSI measurements to detect the TCTL situation is extremely challenging due to the significant signal variance caused by multipath fading in indoor radio channel, carrying the smartphone in different pockets or positions, and differences in smartphone manufacturer and type of the device. In this study we utilize the Mitre Range Angle Structured (MRAS) Private Automated Contact Tracing (PACT) dataset to extensively evaluate the effectiveness of Machine Learning (ML) algorithms in comparison to classical estimation theory techniques to solve the TCTL problem. We provide a comparative performance evaluation of proximity classification accuracy and the corresponding confidence levels using classical estimation theory and a variety of ML algorithms. As the classical estimation method utilizes RSSI characteristics models, it is faster to compute, is more explainable, and drives an analytical solution for the precision bounds proximity estimation. The ML algorithms, Support Vector Machines (SVM), Random Forest, and Gradient Boosted Machines (GBM) utilized thirteen spatial, time-domain, frequency-domain, and statistical features extracted from the BLE RSSI data to generate the same results as classical estimation algorithms. We show that ML algorithms can achieve 3.60%~19.98% better precision, getting closer to achievable bounds for estimation.

Machine learning-based motor assessment of Parkinson's disease using postural sway, gait and lifestyle features on crowdsourced smartphone data.

OBJECTIVES: Remote assessment of gait in patients' homes has become a valuable tool for monitoring the progression of Parkinson's disease (PD). However, these measurements are often not as accurate or reliable as clinical evaluations because it is challenging to objectively distinguish the unique gait characteristics of PD. We explore the inference of patients' stage of PD from their gait using machine learning analyses of data gathered from their smartphone sensors. Specifically, we investigate supervised machine learning (ML) models to classify the severity of the motor part of the UPDRS (MDS-UPDRS 2.10-2.13). Our goals are to facilitate remote monitoring of PD patients and to answer the following questions: (1) What is the patient PD stage based on their gait? (2) Which features are best for understanding and classifying PD gait severities? (3) Which ML classifier types best discriminate PD patients from healthy controls (HC)? and (4) Which ML classifier types can discriminate the severity of PD gait anomalies? METHODOLOGY: Our work uses smartphone sensor data gathered from 9520 patients in the mPower study, of whom 3101 participants uploaded gait recordings and 344 subjects and 471 controls uploaded at least 3 walking activities. We selected 152 PD patients who performed at least 3 recordings before and 3 recordings after taking medications and 304 HC who performed at least 3 walking recordings. From the accelerometer and gyroscope sensor data, we extracted statistical, time, wavelet and frequency domain features, and other lifestyle features were derived directly from participants' survey data. We conducted supervised classification experiments using 10-fold cross-validation and measured the model precision, accuracy, and area under the curve (AUC). RESULTS: The best classification model, best feature, highest classification accuracy, and AUC were (1) random forest and entropy rate, 93% and 0.97, respectively, for walking balance (MDS-UPDRS-2.12); (2) bagged trees and MinMaxDiff, 95% and 0.92, respectively, for shaking/tremor (MDS-UPDRS-2.10); (3) bagged trees and entropy rate, 98% and 0.98, respectively, for freeze of gait; and (4) random forest and MinMaxDiff, 95% and 0.99, respectively, for distinguishing PD patients from HC. CONCLUSION: Machine learning classification was challenging due to the use of data that were subjectively labeled based on patients' answers to the MDS-UPDRS survey questions. However, with use of a significantly larger number of subjects than in prior work and clinically validated gait features, we were able to demonstrate that automatic patient classification based on smartphone sensor data can be used to objectively infer the severity of PD and the extent of specific gait anomalies.

Review of Computational Techniques for Performance Evaluation of RF Localization Inside the Human Body.

Location estimation within the human body by means of wireless signals is becoming popular for a variety of purposes, including wireless endoscopy using camera pills. The precision of wireless ranging in any medium is contingent upon the methodology employed. Two of the most popular wireless tracking methods are received signal strength (RSS) and time of arrival (TOA). The scope of this study is an assessment of the precision of TOA- and RSS-based ranging in the proximity of anthropomorphic tissue by means of simulation software designed to mimic signal transmission in the human body environment. Software simulations of wireless signals traveling within a human body are exceptionally challenging and require extensive computational resources. We created a rudimentary, MATLAB script using the finite-difference time-domain (FDTD) method to simulate the signal transmission inside and outside a human body and correlated the simulation outcomes of this script with the high-end commercial finite-element method (FEM) tool, ANSYS HFSS. First, we demonstrated that the FDTD modeling produces similar outcomes. Next, we employed the script to emulate the RSS and TOA of the wide bandwidth radio transmission within the human body for wireless ranging and estimated the accuracy of each technology. The precision of both methods was also evaluated with the Cramer-Rao lower bound (CRLB), which is frequently used to estimate the ranging methodologies and the effect of human tissue and its motion.

Indoor Motion Detection Using Wi-Fi Channel State Information in Flat Floor Environments Versus in Staircase Environments.

Recently, Wi-Fi channel state information (CSI) motion detection systems have been widely researched for applications in human health care and security in flat floor environments. However, these systems disregard the indoor context, which is often complex and consists of unique features, such as staircases. Motion detection on a staircase is also meaningful and important for various applications, such as fall detection and intruder detection. In this paper, we present the difference in CSI motion detection in flat floor and staircase environments through analysing the radio propagation model and experiments in real settings. For comparison in the two environments, an indoor CSI motion detection system is proposed with several novel methods including correlation-based fusion, moving variance segmentation (MVS), Doppler spread spectrum to improve the system performance, and a correlation check to reduce the implementation cost. Compared with existing systems, our system is validated to have a better performance in both flat floor and staircase environments, and further utilized to verify the superior CSI motion detection performance in staircase environments versus flat floor environments.

Precision of EM Simulation Based Wireless Location Estimation in Multi-Sensor Capsule Endoscopy.

In this paper, we compute and examine two-way localization limits for an RF endoscopy pill as it passes through an individuals gastrointestinal (GI) tract. We obtain finite-difference time-domain and finite element method-based simulation results position assessment employing time of arrival (TOA). By means of a 3-D human body representation from a full-wave simulation software and lognormal models for TOA propagation from implant organs to body surface, we calculate bounds on location estimators in three digestive organs: stomach, small intestine, and large intestine. We present an investigation of the causes influencing localization precision, consisting of a range of organ properties; peripheral sensor array arrangements, number of pills in cooperation, and the random variations in transmit power of sensor nodes. We also perform a localization precision investigation for the situation where the transmission signal of the antenna is arbitrary with a known probability distribution. The computational solver outcome shows that the number of receiver antennas on the exterior of the body has higher impact on the precision of the location than the amount of capsules in collaboration within the GI region. The large intestine is influenced the most by the transmitter power probability distribution.

Measurement of motion detection of wireless capsule endoscope inside large intestine.

Wireless Capsule Endoscope (WCE) provides a noninvasive way to inspect the entire Gastrointestinal (GI) tract, including large intestine, where intestinal diseases most likely occur. As a critical component of capsule endoscopic examination, physicians need to know the precise position of the endoscopic capsule in order to identify the position of detected intestinal diseases. Knowing how the capsule moves inside the large intestine would greatly complement the existing wireless localization systems by providing the motion information. Since the most recently released WCE can take up to 6 frames per second, it's possible to estimate the movement of the capsule by processing the successive image sequence. In this paper, a computer vision based approach without utilizing any external device is proposed to estimate the motion of WCE inside the large intestine. The proposed approach estimate the displacement and rotation of the capsule by calculating entropy and mutual information between frames using Fibonacci method. The obtained results of this approach show its stability and better performance over other existing approaches of motion measurements. Meanwhile, findings of this paper lay a foundation for motion pattern of WCEs inside the large intestine, which will benefit other medical applications.

A video-based speed estimation technique for localizing the wireless capsule endoscope inside gastrointestinal tract.

Wireless Capsule Endoscopy (WCE) is progressively emerging as one of the most popular non-invasive imaging tools for gastrointestinal (GI) tract inspection. As a critical component of capsule endoscopic examination, physicians need to know the precise position of the endoscopic capsule in order to identify the position of intestinal disease. For the WCE, the position of the capsule is defined as the linear distance it is away from certain fixed anatomical landmarks. In order to measure the distance the capsule has traveled, a precise knowledge of how fast the capsule moves is urgently needed. In this paper, we present a novel computer vision based speed estimation technique that is able to extract the speed of the endoscopic capsule by analyzing the displacements between consecutive frames. The proposed approach is validated using a virtual testbed as well as the real endoscopic images. Results show that the proposed method is able to precisely estimate the speed of the endoscopic capsule with 93% accuracy on average, which enhances the localization accuracy of the WCE to less than 2.49 cm.

Comparison of TOA and RSS based techniques for RF localization inside human tissue.

Localization inside the human body using radio frequency (RF) transmission is gaining importance in a number of applications such as Capsule Endoscopy. The accuracy of RF localization depends on the technology adopted for this purpose. The two most common RF localization technologies use received signal strength (RSS) and time-of-arrival (TOA). This paper presents a comparison of the accuracy of TOA and RSS based localization inside human tissue. Analysis of the propagation of radio waves inside the human body is extremely challenging and computationally intensive. We use our proprietary finite difference time domain (FDTD) technique algorithm reported in [1] to simulate waveform transmissions inside the human body, which is almost 60 times faster than commercially available solvers used for similar purposes. The RSS and TOA of the waveforms are extracted for localization and the accuracies of the two methods are compared. The accuracy of each technique is compared with traditional CRLB commonly used for calculation of bounds for the performance of localization techniques.