Experimental Identification of Smartphones Using Fingerprints of Built-In Micro-Electro Mechanical Systems (MEMS)

Abstract

The correct identification of smartphones has various applications in the field of security or the fight against counterfeiting. As the level of sophistication in counterfeit electronics increases, detection procedures must become more accurate but also not destructive for the smartphone under testing. Some components of the smartphone are more likely to reveal their authenticity even without a physical inspection, since they are characterized by hardware fingerprints detectable by simply examining the data they provide. This is the case of MEMS (Micro Electro-Mechanical Systems) components like accelerometers and gyroscopes, where tiny differences and imprecisions in the manufacturing process determine unique patterns in the data output. In this paper, we present the experimental evaluation of the identification of smartphones through their built-in MEMS components. In our study, three different phones of the same model are subject to repeatable movements (composing a repeatable scenario) using an high precision robotic arm. The measurements from MEMS for each repeatable scenario are collected and analyzed. The identification algorithm is based on the extraction of the statistical features of the collected data for each scenario. The features are used in a support vector machine (SVM) classifier to identify the smartphone. The results of the evaluation are presented for different combinations of features and Inertial Measurement Unit (IMU) outputs, which show that detection accuracy of higher than 90% is achievable.

Keywords: MEMS; accelerometers; counterfeit; fingerprinting; gyroscopes; smartphone.

Figure 1

Workflow of the adopted methodology.

Figure 2

Robotic Arm used in the scenario with one of the Phones attached.

Figure 3

z-output from the Gyroscopes as received from the normal axis to the frontal plane of the phones.

Figure 4

Average scores of comparison of each phone against each other and against itself for different features: horizontal lines show the average values for all the features (66.6 and 55.7, respectively).

Figure 5

Threshold $P_{Thr}$ variation and accuracy results for Threshold Entropy, Sure Entropy and Norm Entropy features. The values that give the best average accuracies are, respectively, 1.1 (average 73.0), 1.1 (average 71.6) and 2.5 (average 73.3).

Figure 6

Accuracy and average accuracy for groups of two features (Radial Basis Function (RBF) kernel function, sigma 1.6).

Figure 6

Accuracy and average accuracy for groups of two features (Radial Basis Function (RBF) kernel function, sigma 1.6).

Figure 7

Features combination in groups of two for indirect comparisons. Best group: Variance, Sure Entropy, Average accuracy of 81%.

Figure 8

Accuracy for features one, four and five for all of the Microelectromechanical (MEMS) components.

Figure 9

Accuracy for top 10 groups of five features for Gyroscope x and y outputs.

Figure 10

Receiver Operating Characteristic (ROC) among different phones using the first scenario for Gyroscope X for a single day’s data in the first set of measurements.

Figure 11

ROC among different phones using the first scenario for Gyroscope Y for a single day’s data in the first set of measurements.

Figure 12

ROCs among different phones using the first scenario for Gyroscope X for a single day’s data in the second set of measurements.

Figure 13

ROCs among different phones using the first scenario for Gyroscope Y for a single day’s data in the second set of measurements.