Pervasive and Mobile Computing最新文献

The pervasiveness and high number of Internet of Things (IoT) applications in people’s daily lives make this context a very critical attack surface for cyber threats. The high heterogeneity of involved entities, both in terms of hardware and software characteristics, does not allow the definition of uniform, global, and efficient security solutions. Therefore, researchers have started to investigate novel mechanisms, in which a super node (a gateway, a hub, or a router) analyzes the interactions of the target node with other peers in the network, to detect possible anomalies. The most recent of these strategies base such an analysis on the modeling of the fingerprint of a node behavior in an IoT; nevertheless, existing solutions do not cope with the fully distributed nature of the referring scenario.

In this paper, we try to provide a contribution in this setting, by designing a novel and fully distributed trust model exploiting point-to-point devices’ behavioral fingerprints, a distributed consensus mechanism, and Blockchain technology. In our solution we tackle the non-trivial issue of equipping smart things with a secure mechanism to evaluate, also through their neighbors, the trustworthiness of an object in the network before interacting with it. Beyond the detailed description of our framework, we also illustrate the security model associated with it and the tests carried out to evaluate its correctness and performance.

In the last decade, the investigation of mobility features has gained enormous significance in many scenarios as a result of the significant diffusion and deployment of mobile devices covered by high-speed technologies (e.g., 5G). Many contributions in the literature have attempted to discover mobility properties, but most studies are based on the time features of the mobility process. No study has yet considered the effects of setting a proper sampling frequency (generally set to 1 s), in order to avoid information loss. Following our previous works, we propose a novel predictive spectral approach for mobility sampling based on the concept of a predictive wavelet. With this method, the choice of sampling frequency is governed by the current spectral components of the mobility process and derived from an analysis of future, predicted components. To assess whether our proposal may yield a helpful method, we conducted several simulation campaigns to test sampling accuracy and obtained results that confirmed our expectations.

Several behavioral biometric authentication frameworks have been proposed to authenticate smartphone users based on the analysis of sensors and services. These authentication frameworks verify the user identity by extracting a set of behavioral traits such as touch, sensors and keystroke dynamics, and use machine learning and deep learning techniques to develop the authentication models. Unfortunately, it is not clear how these frameworks perform in the real world deployment and most of the experiments in the literature have been conducted with cooperative users in a controlled environment. In this paper, we present a novel behavioral biometric authentication framework, called B2auth, designed specifically for smartphone users. The framework leverages raw data collected from touchscreen on smartphone to extract behavioral traits for authentication. A Multilayer Perceptron (MLP) neural network is employed to develop authentication models. Unlike many existing experiments conducted in controlled environments with cooperative users, we focused on real-world deployment scenarios, collecting data from 60 participants using smartphones in an uncontrolled environment. The framework achieves promising results in differentiating the legitimate owner and an attacker across various app contexts, showcasing its potential in practical use cases. By utilizing minimalist data collection and cloud-based model generation, the B2auth framework offers an efficient and effective approach to behavioral biometric authentication for smartphones.

Federated learning based participatory sensing has gained much attention lately for the vital task of urban sensing due to privacy and security issues in conventional machine learning. However, inference attacks by the honest-but-curious application server or a malicious adversary can leak the personal attributes of the participants, such as their home and workplace locations, routines, and habits. Approaches proposed in the literature to prevent such information leakage, such as secure multi-party computation and homomorphic encryption, are infeasible for urban sensing applications owing to high communication and computation costs due to multiple rounds of communication between the user and the server. Moreover, for effective modeling of urban sensing phenomenon, the application model needs to be updated frequently — every few minutes or hours, resulting in periodic data-intensive updates by the participants, which severely strains the already limited resources of their mobile devices. This paper proposes a novel low-cost privacy-preserving framework for enhanced protection against the inference of participants’ personal and private attributes from the data leaked through inference attacks. We propose a novel approach of strategically leaking selected location traces by providing computation and communication-light direct (local) model updates, whereas the rest of the model updates (when the user is at sensitive locations) are provided using secure multi-party computation. We propose two new methods based on spatiotemporal entropy and Kullback–Leibler divergence for automatically deciding which model updates need to be sent through secure multi-party computation and which can be sent directly. The proposed approach significantly reduces the computation and communication overhead for participants compared to the fully secure multi-party computation protocols. It provides enhanced protection against the deduction of personal attributes from inferred location traces compared to the direct model updates by confusing the application server or malicious adversary while inferring personal attributes from location traces. Numerical experiments on the popular Geolife GPS trajectories dataset validate our proposed approach by reducing the computation and communication requirements by the participants significantly and, at the same time, enhancing privacy by decreasing the number of inferred sensitive and private locations of participants.

Early detection of agitation in individuals with dementia can lead to timely interventions, preventing the worsening of situations and enhancing their quality of life. The emergence of multi-modal sensing and advances in artificial intelligence make it feasible to explore and apply technology for this goal. We conducted a literature review to understand the current technical developments and challenges of its integration in caregiving institutions. Our systematic review used the Pubmed and IEEE scientific databases, considering studies from 2017 onwards. We included studies focusing on linking sensor data to vocal and/or physical manifestations of agitation. Out of 1622 identified studies, 12 were selected for the final review. Analysis was conducted on study design, technology, decisional data, and data analytics. We identified a gap in the standardized semantic representation of both behavioral descriptions and system event generation configurations. This research highlighted initiatives that leverage existing information in a caregiver's routine, such as correlating electronic health records with sensor data. As predictive systems become more integrated into caregiving routines, false positive reduction needs to be addressed as those will discourage their adoption. Therefore, to ensure adaptive predictive capacity and personalized system re-configuration, we suggest future work to evaluate a framework that incorporates a human-in-the-loop approach for detecting and predicting agitation.

Monkeypox (mpox) virus has become a “public health emergency of international concern” in the last few months, as declared by the World Health Organization, especially for low-income countries. A symptom of mpox infection is the appearance of rashes and skin eruptions, which can lead people to seek medical advice. A technology that might help perform a preliminary screening based on the aspect of skin lesions is the use of Machine Learning for image classification. However, to make this technology suitable on a large scale, it should be usable directly on people mobile devices, with a possible notification to a remote medical expert.

In this work, we investigate the adoption of Deep Learning to detect mpox from skin lesion images derived from smartphone cameras. The proposal leverages Transfer Learning to cope with the scarce availability of mpox image datasets. As a first step, a homogeneous, unpolluted, dataset was produced by manual selection and preprocessing of available image data, publicly released for research purposes. Subsequently, we compared multiple Convolutional Neural Networks (CNNs) using a rigorous 10-fold stratified cross-validation approach and we conducted an analysis to evaluate the models’ fairness towards different skin tones. The best models have been then optimized through quantization for use on mobile devices; measures of classification quality, memory footprint, and processing times validated the feasibility of our proposal. The most favorable outcomes have been achieved by MobileNetV3Large, attaining an F-1 score of 0.928 in the binary task and 0.879 in the multi-class task. Furthermore, the application of quantization led to a reduction in the model size to less than one-third, while simultaneously decreasing the inference time from 0.016 to 0.014 s, with only a marginal loss of 0.004 in F-1 score. Additionally, the use of eXplainable AI has been investigated as a suitable instrument to both technically and clinically validate classification outcomes.

Numerous users from diverse domains access information and perform various operations in multi-domain environments. These users have complex permissions that increase the risk of identity falsification, unauthorized access, and privacy breaches during cross-domain interactions. Consequently, implementing an access control architecture to prevent users from engaging in illicit activities is imperative. This paper proposes a novel blockchain-based access control architecture for multi-domain environments. By integrating the multi-domain environment within a federated chain, the architecture utilizes Decentralized Identifiers (DIDs) for user identification and relies on public/secret key pairs for operational execution. Verifiable credentials are used to authorize permissions and release resources, thereby ensuring authentication and preventing tampering and forgery. In addition, the architecture automates the authorization and access control processes through smart contracts, thereby eliminating human intervention. Finally, we performed a simulation evaluation of the architecture. The most time-consuming process had a runtime of 1074 ms, primarily attributed to interactions with the blockchain. Concurrent testing revealed that with a concurrency level of 2000 demonstrated that the response times for read and write operations were maintained within 1000 ms and 4600 ms, respectively. In terms of storage efficiency, except for user registration, which incurred two gas charges, all the other processes required only one charge.

In the Industrial Internet of Things (IIoT), haptic control of machines or robots can be managed remotely. However, with the emergence of Tactile Industrial Internet of Things (TIIoT), the transmission of haptic data over public channels has raised security and privacy concerns. In such an environment, mutual authentication between haptic users and remotely controlled entities is crucial to prevent illegal control by adversaries. Therefore, we propose an end-to-end authentication scheme, SecTIIoT, to establish secure communication between haptic users and remote IoT devices. The scheme addresses security issues by using lightweight hash cryptographic primitives and employs a useful piggyback strategy to improve authentication efficiency. We demonstrate that SecTIIoT is resilient to various known attacks with formal security proofs and informal security analysis. Furthermore, our detailed performance analysis shows that SecTIIoT outperforms existing lightweight authentication schemes as it provides more security features while reducing computation and communication costs.

The goal of continuous smartphone authentication is to detect when the adversary has gained possession of the user’s device post-login. This is achieved by triggering re-authentication at fixed, frequent intervals. However, these intervals do not take into account external information that might indicate that the impostor has gained physical access to the user’s device. Continuous smartphone authentication typically relies on behavioral cues, such as hand movement and touchscreen swipes, that can be collected without interrupting the user’s activity. Because these behavioral signals are characterized by relatively high error rates compared to physiological biometrics, their use at fixed intervals leads to unnecessary interruptions to the user’s activity in case of a false reject, and to not recognizing the impostor in case of a false accept.

To address these issues, in this paper we introduce a novel framework called SMARTCOPE: Smartphone Change Of Possession Evaluation. In this work, SMARTCOPE leverages smartphone movement signals collected during user activity to determine when the smartphone is no longer in the hands of its owner. When this occurs, SMARTCOPE triggers re-authentication. By using these signals, we are able to reduce the total number of re-authentication points while simultaneously lowering re-authentication error rates. Our analysis shows that our technique can reduce equal error rates by over 40%, from 7.8% to 4.6% using movement and keystroke features. Further, we show that SMARTCOPE can be used to transform a static (login-time) authentication system, such as face recognition, to a continuous re-authentication system, with a significant increase in security and limited impact on usability.

The IoT devices with embedded mobile devices create the Internet of Mobile Things (IoMT) paradigm. Mobility is not supported by the routing protocol for low-power and lossy networks (RPL) created for static networks. IoMT has raised routing challenges such as link failure, instability, energy depletion, packet loss, and handover delay in the network. In this context, IoMT Fuzzy-based RPL (IoMT-FRPL) is proposed in this paper to enhance routing performance. Receiving Signal Strength Indicator (RSSI), Euclidean distance, Hop Count, and Expected Transmission Count (ETX) metrics are built into the fuzzy interface system for the mobile nodes in the network to conserve energy. The IoMT-FRPL consists of the following three key steps: The first steps are data transmission and motion investigation, the second is fuzzy-based prediction of a new static parent for the mobile node, and the third is verifying the unique attachment point. When conventional RPL, mRPL, and EMA-RPL were compared to IoMT-performance FRPL's in Cooja/Contiki 2.7, the simulation results revealed improvements in energy consumption, handover delay, packet delivery rate (PDR), and signaling cost.