With the development of 5G and Internet Vehicles technology, possibility remote wireless attack on an in-vehicle network has been proven by security researchers. Anomaly detection technology can effectively alleviate threat, as first line defense. Based this, this paper proposes a distributed anomaly system using hierarchical temporal memory (HTM) to enhance vehicular controller area bus. The HTM model predict flow data in real time, which depends state previous learning. In addition, we...

App markets, being crucial and critical for today's mobile ecosystem, have also become a natural malware delivery channel since they actually "lend credibility" to malicious apps. In the past decade, machine learning (ML) techniques been explored automated, robust detection. Unfortunately, date, we yet see an ML-based detection solution deployed at market scales. To better understand real-world challenges, conduct collaborative study with major Android app (T-Market) offering us large-scale...

Indoor localization has become an essential demand driven by indoor location-based services (ILBSs) for mobile users. With the rising of Internet Things (IoT), heterogeneous smartphones and wearables have ubiquitous. However, ILBSs IoT devices confront significant challenges, such as received signal strength (RSS) variances caused hardware heterogeneity, multipath reflections from complex environments, time restricted computation resources. This article proposes EdgeLoc, a robust real-time...

With the popularity of Mobility-on-Demand (MOD) vehicles, a new market called MOD-Vehicular-Crowdsensing (MOVE-CS) was introduced for drivers to earn more by collecting road data. Unfortunately, MOVE-CS failed after two years operation. To identify root cause, we survey 581 and reveal its simple operation model based on blindly competitive rewards. This brings most few yields, resulting in their withdrawals. In contrast, similar termed MOD-Human-Crowdsensing (MOMAN-CS) remains successful...

With the rapid development of Android applications in recent years, applications' security has more and attention paid to it. The malware detection can be divided into two types: behavior-based code-based detection. In this paper, we present a quick accurate malicious scheme based on sensitive API calls. training process, calls various are extracted as large eigenvector through reverse analysis. Then employ mutual information measure correlation between specific malware, generate set...

Doors as densely-deployed natural landmarks play an important role in improving indoor positioning systems. However, the state-of-the-art door event detection works are based on either vision or infrastructure, thus incurring non-trivial device management cost. To address these problems, we present a Light-weight Magnetic-based Door Event Detection method, called LMDD. It leverages built-in magnetic sensors of common smartphones to achieve infrastructure-free detection. After analyzing...

With recent advances on cellular technologies (such as 5G) that push the boundary of performance, reliability has become a key concern technology adoption and deployment. However, this fundamental never been addressed due to challenges measuring mobile devices cost conducting large-scale measurements. This paper closes knowledge gap by presenting first large-scale, in-depth study with more than 70 million Android phones across 34 different hardware models. Our identifies critical factors...

In recent years, the development of Internet Things technology has been very rapid. However, with explosive growth IoT devices, challenges facing environment are becoming more and serious. How to ensure security becomes a key issue. The primary purpose is protect private data users, infrastructure, devices. This paper reviews threat models at each level in years discusses some past future solutions. At first, threats systematically introduced from perspectives physics, network data....

The following topics are dealt with: learning (artificial intelligence); data mining; social networking (online); pattern classification; text analysis; security of data; feature extraction; Internet; graph theory; computer network security.

With the rapid development of smart cities, buildings are generating a massive amount building sensing data by equipped sensors. Indeed, provides promising way to enrich series data-demanding and cost-expensive urban mobile applications. In this paper, we study how reuse predict traffic volume on nearby roads. Nevertheless, it is non-trivial achieve accurate prediction such cross-domain with two major challenges. First, relationships between not unknown as prior, spatio-temporal complexities...

WiFi infrastructures are widely deployed in both public and private buildings. They make the connection to internet more convenient. Recently, researchers find that signals have ability sense changes environment can detect human motion even identify activities his identity a device-free manner, has many potential security applications smart home. Previous detection systems only of regular moving patterns. However, they may significant performance degradation when used intrusion detection. In...

Virtual devices based on device emulation have been widely used in lab research of mobile app testing for their efficiency and low cost. However, it remains controversial to use virtual industry, given the inherent difficulties high-fidelity across diverse systems devices. Hence, companies still rely physical farms or services like AWS Device Farm.

Android system has been widely deployed in energy-constrained IoT devices for many practical applications, such as smart phone, home, healthcare, fitness, and beacons. However, users oftentimes suffer from app crashes, which directly disrupt user experience could lead to data loss. Till now, the community have limited understanding of their prevalence, characteristics, root causes. In this article, we make an in-depth study crash events regarding ten very popular apps different genres, based...

Android applications have emerged as a prime target for hackers. malware detection stands pivotal technology, crucial safeguarding network security and thwarting anomalies. However, traditional static analysis makes it difficult to analyze new malicious applications, while dynamic requires higher system resources. We propose novel lightweight deep-learning framework based on attention temporal networks. This study delves into the Dalvik opcode sequences of malware, employing N-gram algorithm...

Step counting is a fundamental unit of human locomotion, and preferred metric for quantifying physical activity. However, the existing step counters are too inconvenient to wear treadmill can not count steps. Recently, commercial Wi-Fi based device-free sensing shows promising future ubiquitous motion-based interactions provides possibility device free counting. Previous research activity with Wi- Fi mainly focuses on single person recognition. The primary challenge multi-person recognition...

Despite being crucial to today's mobile ecosystem, app markets have meanwhile become a natural, convenient malware delivery channel as they actually "lend credibility" malicious apps. In the past few years, machine learning (ML) techniques been widely explored for automated, robust detection, but till now we not seen an ML-based detection solution applied at market scales. To systematically understand real-world challenges, conduct collaborative study with T-Market, popular Android that...

Contactless gesture recognition is an emerging interactive technique in ubiquitous and mobile computing. It combines the linguistics with wireless signals to analyze, judge, integrate human gestures by usage of intelligent algorithms. The existing contactless studies can achieve machine learning technologies. But practice, some objective factors, such as user's position, non-line sight condition, seriously affect performance these systems. In this paper, we propose robust using physical...

With the explosive usage of smart mobile devices, sustainable access to wireless networks (e.g., Wi-Fi) has become a pervasive demand. Most users expect seamless network connection with low cost. Indeed, this can be achieved by using an accurate received signal strength (RSS) map points. While existing methods are either costly or unscalable, recently emerged crowdsensing (MCS) paradigm is promising technique for building RSS maps. MCS applications leverage devices collaboratively collect...

Doors are important landmarks for indoor positioning systems. Hence an accurate and light-weight door detection approach is highly desired. The state-of-the-art solutions either vision based or infrastructure based, which incur nontrivial device management cost. This paper presents a novel approach, Light-weight Magnetic-based Door Detection (LMDD), only relies on the information from built-in sensors of smartphone. LMDD detects by analyzing change magnetic signal extracting special features...

Android <i>overlay</i> enables one app to draw over other apps by creating an extra <monospace>View</monospace> layer atop the host , which nevertheless can be exploited malicious (malware) attack users. To combat this threat, prior countermeasures concentrate on restricting capabilities of overlays at OS level while sacrificing overlays' usability; recently, overlay mechanism has been substantially updated prevent a variety attacks, however still evaded considerable adversaries. address...