The nerdiest of our in-house nerds recently took the plunge on researching smartphone sensor data and its potential for use as forensic evidence. Since we know ya’ll can’t go more than a couple of hours without accessing your smartphones, we’ve summarized the “meat & potatoes” of her findings for you. Enjoy…
Technological advances are progressing at a faster rate of change than all of social change; business change; and legislative change put together. The distribution of such change has created an environment where society must adapt to face the influence that technology has on the future of crime and criminal investigations. According to the Department of Justice, one of the three main influences that technology has on the future of crime is the development of new technologies to either stop or deter criminal activities. These technologies will include new forms of digital forensic evidence.
Change in consumer needs and preferences is a critical variable in the evolution of digital forensic evidence. As consumers, we drive the popularity of devices that will ultimately provide digital evidence. The wireless communications market continues to experience exponential growth in Canada and abroad. A 2015 study by Catalyst shows an increase in smartphone ownership by 24% year over year, declaring smartphones virtually ubiquitous in Canada. As smartphone penetration grows, the demand for newer replacement models follows with consumers increasingly using smartphones for activities that previously would have been accomplished by use of a landline or laptop. This wide-spread use of smartphones and their evolving capabilities introduces new opportunities in the field of digital forensics not to mention new concerns for user privacy.
Today, criminal investigations often involve telephone record analysis which identifies patterns in contact between subjects. This process involves a review of the number of calls made, time of the calls, parties called, etc., for the purpose of identifying devices that may be used in a conspiracy. Just as traditional telephone record analysis provides valuable corroborating evidence of an overt act, it has been proposed that Law Enforcement Agency’s (LEA’s) may obtain authority to intercept smartphone sensor data to serve as digital evidence in the near future. The sensor data that are stored in most smartphones include: cameras, microphones, global positioning systems, motion sensors called accelerometer and gyroscope, and, environment sensors that capture proximity, light and temperature. These types of sensor data may provide context that constitutes evidence beyond the current scope of lawful interception systems by forming evidence chains when they are combined or when corroborated with other forms of evidence.
The context that is referred to herein has been demonstrated through research at Fordham University in New York where developers at the Wireless Sensor Data Mining (WISDM) Lab have created a system using smartphone sensor data that recognizes such activities as sitting, standing, walking, climbing stairs, and jogging. The WISDM states it is also “able to predict a user’s gender with 71 percent accuracy, and can distinguish between “tall” and “short” people and “heavy” and “light” people, each with about 80 percent accuracy.” Perhaps more relevant is the establishment that one’s gait as measured by a smartphone accelerometer, may be distinctive enough to be used for identification purposes. The progress made at WISDM Lab is mirrored by Shaun Gallager who, in his thesis, found that identifying an individual’s gait is as unique as identifying an individual’s fingerprint.
In addition to the identification of a user’s physical activities and stature, it has been suggested that smartphone sensor data may in fact provide a near-total data profile on the life of an individual. One study, found that by analyzing application usage and communication history, they were able to statistically infer a user’s daily mood average. Initially with a rate of 66% accuracy, which gradually improved to 93% accuracy after a two-month personalized period. Similar discoveries were made by Lathia, N., et al (2013) who claim that smartphone sensor data “can unobtrusively sense human behavior and deliver feedback and behavioral therapy.” In their article, the authors discuss applications for behavioral monitoring and change and present the first holistic platform for large-scale digital behavior change intervention.
Given the vast array of future uses for smartphone sensor data, it is important to note that LEA’s will be required to standardize appropriate techniques for obtaining and processing this valuable sensor data. Pioneers in the field will need to delve into this discipline in order to develop sound techniques for transferring and presenting the smartphone sensor data, perhaps looking to industry leaders for tools to bring efficiency to the process. And, they may be well on their way because both Apple and Samsung have introduced software for collecting and sharing smartphone sensor data signaling a growing interest in the use of sensor data across sectors.
So, what does this mean to the average citizen? Simply put: your smartphone is collecting more than just your selfies, it is literally storing a near-total data profile on you that one day (soon), may be able to serve as digital forensic evidence before the courts. We’ll be keeping tabs on the forces that are supporting and / or hindering the increased use of smartphone sensor data as digital forensic evidence such as emerging case law and relevant sensor data research.