A new method for estimating knee joint flexion/extension angles from segment acceleration and angular velocity data is described. The approach uses a combination of Kalman filters and biomechanical constraints based on anatomical knowledge. In contrast to many recently published methods, the proposed approach does not make use of the earth's magnetic field and hence is insensitive to the complex field distortions commonly found in modern buildings. The method was validated experimentally by calculating knee angle from measurements taken from two IMUs placed on adjacent body segments. In contrast to many previous studies which have validated their approach during relatively slow activities or over short durations, the performance of the algorithm was evaluated during both walking and running over 5 minute periods. Seven healthy subjects were tested at various speeds from 1 to 5 miles/hour. Errors were estimated by comparing the results against data obtained simultaneously from a 10 camera motion tracking system (Qualysis). The average measurement error ranged from 0.7 degrees for slow walking (1 mph) to 3.4 degrees for running (5mph). The joint constraint used in the IMU analysis was derived from the Qualysis data.Limitations of the method, its clinical application and its possible extension are discussed.
Sensors are an essential component of many computer systems today. Mobile devices are a good example, containing a vast array of sensors from accelerometers and GPS units, to cameras and microphones. Data from these sensors are accessible to application programmers who can use this data to build context-aware applications. Good sensor accuracy is often crucial, and therefore manufacturers often use perdevice factory calibration to compensate for systematic errors introduced during manufacture. In this paper we explore a new type of fingerprinting attack on sensor data: calibration fingerprinting. A calibration fingerprinting attack infers the perdevice factory calibration data from a device by careful analysis of the sensor output alone. Such an attack does not require direct access to any calibration parameters since these are often embedded inside the firmware of the device and are not directly accessible by application developers. We demonstrate the potential of this new class of attack by performing calibration fingerprinting attacks on the inertial measurement unit sensors found in iOS and Android devices. These sensors are good candidates because access to these sensors does not require any special permissions, and the data can be accessed via both a native app installed on a device and also by JavaScript when visiting a website on an iOS and Android device. We find we are able to perform a very effective calibration fingerprinting attack: our approach requires fewer than 100 samples of sensor data and takes less than one second to collect and process into a device fingerprint that does not change over time or after factory reset. We demonstrate that our approach is very likely to produce globally unique fingerprints for iOS devices, with an estimated 67 bits of entropy in the fingerprint for iPhone 6S devices. In addition, we find that the accelerometer of Google Pixel 2 and Pixel 3 devices can also be fingerprinted by our approach.
Device fingerprinting aims to generate a distinctive signature, or fingerprint, that uniquely identifies individual computing devices. Fingerprints may be a privacy concern since apps and websites can use them to track user activity online. To protect user privacy, both Android and iOS have included a variety of measures to prevent such tracking. In this paper we present a new type of fingerprinting, factory calibration fingerprinting, that bypasses existing tracking protection. Our attack recovers embedded per-device factory calibration data from the accelerometer, gyroscope, and magnetometer sensors that are pervasive in modern smartphones by careful analysis of the sensor output alone. We discuss the factory calibration behaviour of each sensor and show that the calibration fingerprint is fast to generate, does not change over time or after a factory reset, and can be used to track users across apps and websites without any special permission from the user. We find the calibration fingerprint is very likely to be globally unique for iOS devices, with an estimated 67 bits of entropy for the iPhone 6S. In addition, we have analysed 146 Android device models from 11 vendors and found the attack also works on recent Google Pixel devices. For Pixel 4/4 XL, we estimate the calibration fingerprint provides about 57 bits of entropy. Following our disclosures, Apple deployed a mitigation in iOS 12.2 and Google in Android 11. We analyse Apple's fix and show that the mitigation is imperfect although it is likely to be sufficient in most threat models.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.