Real Android Devices Research Articles

PIN Inference Attack: A Threat to Mobile Security and Smartphone-Controlled Robots

The proliferation of smartphones equipped with various sensors created a large user base, which includes robots (smartphone controlled) as well. The high processor enabled smartphones are used as software robots to track hour-to-hour activities of users using on-board motion sensors. The magnetometer, accelerometer, and gyroscope are non-permission based motion sensors. The logs of these sensors create discriminative patterns to identify users’ movements. This leverages an adversary to cause privacy threats using these non-permission based sensors without user consent. Recent works also highlighted the privacy threats due to the motion sensors in smartphones. Therefore, in this paper, we present a potential misuse of technology where smartphones can be hacked by PIN Inference attack endangering the mobile security. Once the smartphone is compromised, it may lead to various repercussions. We discuss one such case where after PIN is inferred the adversary incorrectly uses our robot made for corridor detection. With this example, we show the severity of the PIN-Inference attack as nowadays many user activities are governed by robots. We perform experiments to collect motion sensors’ data and infer 4-digit PIN of a smartphone. Our strategy is to detect every key pressed one by one constituting a 4-digit PIN. Our inference results are promising and prove that an adversary can make use of sensors’ data to infiltrate user’s sensitive information from smartphones. Our proposed model infers 84% of the PINs correctly within 40 attempts when tested on 50 PINs. Given real Android devices with different users, we can identify the PIN of smartphones by training our machine learning based model. We also compare our approach with existing state-of-the-art approaches to show the efficacy of our attack model.

Maintaining Application Context of Smartphones by Selectively Supporting Swap and Kill

With the rapid diffusion of smartphones as well as the technical advances in mobile platform technologies, application domains supported by smartphones have dramatically increased. Unlike traditional computer systems, however, the context of applications may be lost in a smartphone as mobile OS usually kills applications without user's approval when free memory space is exhausted. This was not a serious issue when a smartphone was a personal entertainment device, but it is now significant as a lot of official works are also performed by a smartphone. Instead of killing a process, the context of an application can be backed up to swap storage, but smartphones do not accept it as swap incurs excessively heavy I/O traffic. However, our findings show that the overhead of swap is not serious in recent smartphone devices and it can also be eliminated by judicious software management. This article proposes a selective swap scheme that classifies applications based on their context-saving characteristics, and controls the number of processes involved in swap by monitoring system situations and application characteristics. That is, we maintain the context of applications selectively by swap or application's own state-saving. We implement and measure the effectiveness of our scheme on a real Android device, showing that it provides the context-savings of applications without additional overhead.

Rethinking anti-emulation techniques for large-scale software deployment

From the security perspective, emulation is often utilized to analyze unknown malware owing to its capability of tracing fine-grained runtime behavior (i.e., execution path exploration). To this end, attackers equip their malware with powerful anti-emulation techniques that fingerprint the emulated system environment, thereby avoiding dynamic analysis. However, this is not the only use case of anti-emulation. Recently, legitimate software vendors are also putting significant efforts to prevent their products running on top of the emulated execution environment. There are mainly two reasons for this which are: (i) securing the intellectual property from emulation-assisted reverse-engineering, and (ii) disallowing the customers using the application without purchasing the actual hardware. From the previous literature, various anti-emulation techniques were explored. Unfortunately, existing techniques are mostly discussed and developed with malware’s perspective. In this paper, we flip this conventional paradigm and discuss anti-emulation techniques in terms of protecting Commercial-Off-the-Shelf (COTS) software. Due to the higher requirements for usability, existing anti-emulation techniques are inapt for large-scale application vendors. To overcome such problem, we introduce three new techniques in vendors perspective for deploying their product. We evaluate the efficacy of our techniques in five aspects: (i) fast detection speed, (ii) high accuracy, (iii) low power consumption, (iv) a broad range of compatibility, and (v) high cost of bypassing. Based on our experiments, we demonstrate that misaligning the vectorization (e.g., Intel SIMD, ARM NEON) can be utilized as a promising anti-emulation technique among the proposed ones. To confirm the effectiveness, we applied our technology against 176 real Android devices and various emulators as a test bed.

Android malware detection with unbiased confidence guarantees

Abstract The impressive growth of smartphone devices in combination with the rising ubiquity of using mobile platforms for sensitive applications such as Internet banking, have triggered a rapid increase in mobile malware. In recent literature, many studies examine Machine Learning techniques, as the most promising approach for mobile malware detection, without however quantifying the uncertainty involved in their detections. In this paper, we address this problem by proposing a machine learning dynamic analysis approach that provides provably valid confidence guarantees in each malware detection. Moreover the particular guarantees hold for both the malicious and benign classes independently and are unaffected by any bias in the data. The proposed approach is based on a novel machine learning framework, called Conformal Prediction, combined with a random forests classifier. We examine its performance on a large-scale dataset collected by installing 1866 malicious and 4816 benign applications on a real android device. We make this collection of dynamic analysis data available to the research community. The obtained experimental results demonstrate the empirical validity, usefulness and unbiased nature of the outputs produced by the proposed approach.

WARM: Workload-Aware Reliability Management in Linux/Android

With CMOS scaling beyond 14 nm, reliability is a major concern for IC manufacturers. Reliability-aware design has a non-negligible overhead and cannot account for user experience in mobile devices. An alternative is dynamic reliability management (DRM), which counteracts degradation by adapting the operating conditions at runtime. In this paper, for the first time we formulate DRM as an optimization problem that accounts for reliability, temperature and performance. We develop an optimal policy for multicores using convex optimization, and show that it is not feasible to implement on real systems. For this reason, we propose workload-aware reliability management (WARM), a fast DRM technique adapting to diverse workload requirements to trade reliability and user experience. WARM is implemented and tested on a real Android device. WARM approximates the solution of the convex solver within 5% on average, while executing more than $400 {\times }$ faster. WARM integrates a thermal controller that allocates tasks to meet thermal constraints. This is required since degradation strongly depends on temperature. We show that WARM meets temperature constraints within 5% in 87.5% more cases than the state-of-the-art. We show that WARM task allocation achieves up to one year lifetime improvement for a multicore platform. It can achieve up to 100% of performance improvement on cluster architectures, such as big.LITTLE, while still guaranteeing the reliability target. Finally, we show that it achieves performance in the 4% of the maximum for a broad range of a applications, while meeting the reliability constraints.

Enersave API: Android-based power-saving framework for mobile devices

Power consumption is a major factor to be taken into consideration when using mobile devices in the IoT field. Good Power management requires proper understanding of the way in which it is being consumed by the end-devices. This paper is a continuation of the work in Ref. [1] and proposes an energy saving API for the Android Operating System in order to help developers turn their applications into energy-aware ones. The main features heavily used for building smart applications, greatly impact battery life of Android devices and which have been taken into consideration are: Screen brightness, Colour scheme, CPU frequency, 2G/3G network, Maps, Low power localisation, Bluetooth and Wi-Fi. The assessment of the power-saving API has been performed on real Android devices and also compared to the most powerful power-saving applications – DU Battery Saver and Battery Saver 2016 – currently available on the Android market. Comparisons demonstrate that the Enersave API has a significant impact on power saving when incorporated in android applications. While DU Battery Saver and Battery Saver 2016 help saving 22.2% and 40.5% of the battery power respectively, the incorporation of the Enersave API in android applications can help save 84.6% of battery power.