Ranging Explosion Events Using Smartphones Srinivas Chakravarthi Thandu * , Sriram Chellappan † , Zhaozheng Yin * * Department of Computer Science, Missouri University of Science & Technology, Rolla, MO 65401 USA. {stgk4, yinz}@mst.edu † Department of Computer Science and Engineering, University of South Florida, Tampa, FL 33620 USA. shri@cse.usf.edu Abstract—In this paper, we address the problem of ranging explosion events from sensing corresponding accelerometer read- ings from stationary smartphones. First, we statically emplaced a number of smartphones with built-in accelerometers at various locations in the vicinity of real explosions (conducted at a university training facility). An app was installed in 4 off-the- shelf smartphones to collect accelerometer readings continuously, and effectively retaining only those readings that correspond to an explosion event (while filtering out the rest). As a result, a total of 52 data-sets from 4 individual explosion blast-experiments (with Dynamite acting as the explosive charge) were collected. Using these data-sets, we developed a non linear regression model to estimate the distance of the source of an explosion event, and the intensity of the explosion (measured in terms of charge weight of the explosive material) based on extracting a number of statistical features from the accelerometer sensor readings in three dimensions (lateral (x), longitudinal (y), and vertical (z) directions) from smartphones. We are able to range the explosion event, with an average case error of 12.86% in our experiments. We were also able to estimate the intensity of the explosion event with a high accuracy, with an average case error of 11.26%. To the best of our knowledge, this is the first work that attempts to range explosion events leveraging sensor readings from smartphones. I. I NTRODUCTION Smartphones today are becoming both ubiquitous, as well as powerful with significant processing, networking and storage capabilities. In parallel, a critical development in modern smartphones come from the ability to embed multiple sensors in them for fine grained sensing of several phenomena. For instance, a modern smartphone like Samsung Galaxy S4 has built-in sensors that can measure acceleration, ambient tem- perature, pressure, humidity, light intensity, magnetic intensity, sound intensity, and much more, with high sampling rates. The LIS344ALH accelerometer sensor in the Samsung Galaxy S4 phone [1] can sample up to 200 samples per second, and the sampling rate is programmable. Furthermore, numerous studies have been conducted to optimize the performance of smartphone sensors today from the perspective of accuracy, energy efficiency and processing speed [2]. There is a clear and tangible reason for the continued inno- vation in sensing capabilities of smartphones today, and that lies in numerous innovative and societally useful applications leveraging smartphone sensors. The most significant one is emerging in the domain of health-care and well-being. In [7], smartphone accelerometer is leveraged to detect the gait of a subject with applications for fall detection in elder care. The acoustic sensors in smartphones have been leveraged for self-localization of smartphones in [5]. More recently, the pressure sensors in smartphones have been leveraged for context detection in the domain of urban transportation [10], and the magnetometers also available in most smartphones today have been used to detect the presence and shapes of metal pipes or bars embedded behind walls with applications related to building maintenance [12]. While all of the above works focus on applications leverag- ing a single smartphone, there is another trend of leveraging sensory data from multiple smartphones for societal scale applications. Of these the most significant one so far has been detecting earthquakes from accelerometer readings from multiple smartphones. Community Sense and Response (CSR) system proposed by Faulkner, et. al. [4] leverages accelerom- eter sensors in smartphones for monitoring earthquakes. The iShake project designed by Jack, et. al. [9] at the University of California, Berkeley resulted in the design of a mobile client back-end server architecture that uses sensor-equipped mobile devices to sense earthquakes. In [8], accelerometers of smartphones were used to record the acceleration in real time in order to detect earthquakes. Contributions of this paper: In this paper, we primarily focus on ranging the source of an explosion event using accelerometer readings obtained from statically placed smart- phones in the vicinity of the explosion event, which sense the associated seismic vibrations. A secondary focus is on estimating the intensity of the explosion (as a notion of the charge-weight of the explosive material). Unfortunately, these problems come with significant challenges. The first (and most significant) challenge is the access to a facility where real explosions take place, while being controlled suitably to place smartphone sensors and obtain corresponding ground truth data (in terms of explosives type, intensity, distance from the smartphones etc). Fortunately, the Explosives Research Lab at Missouri University of Science and Technology is a facility where regular blasts in a controlled facility are carried out to train students. We participated in multiple blasting experiments in May 2014, with a number of smartphones to collect corresponding sensory data (after following appropriate safety procedures) to demonstrate the feasibility of ranging explosion events. The second major challenge stems from continuously stor- ing and processing accelerometer readings from smartphones. Basically, storing all accelerometer readings and processing