Pervasive and Mobile Computing最新文献

Machine Learning (ML) is growing in popularity and is applied in numerous fields to solve complex problems. Opportunistic Networks are a type of Ad-hoc Network where a contemporaneous path does not always exist. So, forwarding methods that assume the availability of contemporaneous paths does not work. ML techniques are applied to resolve the fundamental problems in Opportunistic Networks, including contact probability, link prediction, making a forwarding decision, friendship strength, and dynamic topology. The paper summarises different ML techniques applied in Opportunistic Networks with their benefits, research challenges, and future opportunities. The study provides insight into the Opportunistic Networks with ML and motivates the researcher to apply ML techniques to overcome various challenges in Opportunistic Networks.

Function-as-a-Service (FaaS) has emerged as an evolution of traditional Cloud service models, allowing users to define and execute pieces of codes (i.e., functions) in a serverless manner, with the provider taking care of most operational issues. With the unending growth of resource availability in the Edge-to-Cloud Continuum, there is increasing interest in adopting FaaS near the Edge as well, to better support geo-distributed and pervasive applications. However, as the existing FaaS frameworks have mostly been designed with Cloud in mind, new architectures are necessary to cope with the additional challenges of the Continuum, such as higher heterogeneity, network latencies, limited computing capacity.

In this paper, we present an extended version of Serverledge, a FaaS framework designed to span Edge and Cloud computing landscapes. Serverledge relies on a decentralized architecture, where each FaaS node is able to autonomously schedule and execute functions. To take advantage of the computational capacity of the infrastructure, Serverledge nodes also rely on horizontal and vertical function offloading mechanisms. In this work we particularly focus on the design of mechanisms for function offloading and live function migration across nodes. We implement these mechanisms in Serverledge and evaluate their impact and performance considering different scenarios and functions.

People have high demands for comfort and technology in indoor environments. Gestures, as a natural and friendly human computer interaction (HCI) method, have received widespread attention and have been the subject of many research studies. Traditional approaches are based on wearable devices and cameras, which can be cumbersome to operate and infringe upon users’ privacy. Millimeter-wave (mmWave) radar avoids these problems by detecting gestures in a noninvasive manner. However, it encounters practical challenges in complex indoor environments, such as dynamic disturbance from surroundings, variable usage conditions and diverse gesture patterns, which conventionally require considerable manual effort to address. In this paper, we attempt to minimize human supervision and propose a noninvasive gesture recognition method named RaGe that involves a commercial mmWave indoor radar. First, a parameter optimization framework considering gesture prior constraints is proposed for radar configuration, which functions to weaken the disturbance from surroundings. Second, we alleviate data shortages in variable usage conditions and achieve low-cost data augmentation by applying affine transformations. Third, we combine deformable convolution operations with an unsupervised attention mechanism, thus exploring the intrinsic features involved in diverse gesture patterns. Experimental results show that RaGe is able to recognize 7 gestures with 99.3% accuracy and less human supervision, surpassing the state-of-the-art methods in comparative experiments.

With the soaring adoption of in-ear wearables, the research community has started investigating suitable in-ear heart rate detection systems. Heart rate is a key physiological marker of cardiovascular health and physical fitness. Continuous and reliable heart rate monitoring with wearable devices has therefore gained increasing attention in recent years. Existing heart rate detection systems in wearables mainly rely on photoplethysmography (PPG) sensors, however, these are notorious for poor performance in the presence of human motion. In this work, leveraging the occlusion effect that enhances low-frequency bone-conducted sounds in the ear canal, we investigate for the first time in-ear audio-based motion-resilient heart rate monitoring. We first collected heart rate-induced sounds in the ear canal using an in-ear microphone under seven stationary activities and two full-body motion activities (i.e., walking, and running). Then, we devised a novel deep learning based motion artefact (MA) mitigation framework to denoise the in-ear audio signals, followed by a heart rate estimation algorithm to extract heart rate. With data collected from 15 subjects over nine activities, we demonstrate that hEARt, our end-to-end approach, achieves a mean absolute error (MAE) of 1.88 ± 2.89 BPM, 6.83 ± 5.05 BPM, and 13.19 ± 11.37 BPM for stationary, walking, and running, respectively, opening the door to a new non-invasive and affordable heart rate monitoring with useable performance for daily activities. Not only does hEARt outperform previous in-ear heart rate monitoring work, but it outperforms reported in-ear PPG performance.

Empowered by Internet of Things (IoT) and cloud computing platforms, the concept of smart cities is making a transition from conceptual models to development and implementation phases. Multiple smart city initiatives and services such as Smart Grid and Smart Meters have emerged that have led to the accumulation of massive amounts of energy big data. Big data is typically characterized by five distinct features namely, volume, velocity, variety, veracity, and value. To gain insights and to monetize big data, data has to be collected, stored, processed, analyzed, mined, and visualized. This paper identifies the primary layers of a big data architecture with start-of-the-art communication, storage, and processing technologies that can be utilized to gain meaningful insights and intelligence from big data. In addition, this paper gives an in-depth overview for research and development who intend to explore the various techniques and technologies that can be implemented for harnessing big data value utilizing the recent big data specific processing and visualization tools. Finally, a use case model utilizing the above mentioned technologies for Smart Grid is presented to demonstrate the energy big data road map from generation to monetization. Our key findings highlight the significance of selecting the appropriate big data tools and technologies for each layer of big data architecture, detailing their advantages and disadvantages. We pinpoint the critical shortcomings of existing works, particularly the lack of a unified framework that effectively integrates these layers for smart city applications. This gap presents both a challenge and an opportunity for future research, suggesting a need for more holistic and interoperable solutions in big data management and utilization.

UAVs or vehicular networks have been extensively used in different domains. Such a system network consists of various heterogeneous and mobile devices operating autonomously and cooperatively to provide flexible services. However, ensuring devices’ runtime integrity has always been critical to such highly dynamic and disruptive networks. Collective attestation is a popular technique in ensuring service integrity on remote devices. However, the physical attacks pose significant threats to the enforcement of the runtime integrity, and the existing detection approaches raise a considerable number of false positives, which impede the robustness of the network. We propose LICAPA, a collective attestation framework for detecting physical attacks with high accuracy. LICAPA can detect a device under physical attack with the timestamps signed by other recently-attested devices. Such a proof-from-others mechanism provides more knowledge about the compromised device for physical attack detection. It reduces the potential false positives compared with the state-of-the-art approaches. LICAPA provides a physical-adversary-tolerant runtime device joining mechanism and a new attestation report aggregation scheme to reduce the storage and communication cost of the device. On the prototype implementation of the trust anchor, we evaluate LICAPA’s computational costs. The simulation results demonstrate LICAPA’s low communication cost and long resistance time against false detection on physical attack. LICAPA reduces the overall swarm attestation cost by over 20% compared with SALAD (Secure and Lightweight Attestation of Highly Dynamic and Disruptive Networks) and PASTA (Practical Attestation Protocol for Autonomous Embedded Systems).

As a special Internet of Things (IoT) application, the wireless body area network (WBAN) has gained widespread attention by medical institutions. However, existing schemes for WBAN data transmission lack heterogeneity support across certificateless cryptosystem (CLC) and public key infrastructure (PKI), resulting in issues like key escrow or complicated certificate management. In addition, for performance reasons, conventional signcryption protocols are unsuitable for WBAN applications. To address these gaps and enable secure and efficient sensitive data transmission from WBAN sensors to hospital servers, we design a heterogeneous online/offline signcryption scheme. Our scheme enables patients’ sensors implanted or worn to encrypt sensitive data in CLC and send it to the hospital server in PKI system. The CLC avoids key escrow issue while the PKI increases scalability. We minimize the online computational cost of WBAN sensors by dividing signcryption into offline and online phases, with time-consuming operations in the offline phase. Furthermore, we formally prove the security of our scheme and evaluate its performance. Results show our scheme has advantages in supporting heterogeneity across CLC and PKI with low computational costs, making it uniquely suitable for the protection of data privacy in WBAN applications compared to existing protocols.

In recent years, localizing or identifying the position of unknown sensor nodes has become an essential problem in Wireless Sensor Networks (WSN). The improvement in localization accuracy leads to obtaining the exact location of the dumb node. Among all localization algorithms, Amorphous localization is highly suggested for usage in many application domains due to its simplicity, viability, low cost, and zero additional hardware requirements. Position estimation of the dumb node in the Amorphous algorithm considers three different practical scenarios, such as the position of dumb nodes falling within the range of anchor nodes, the position of the dumb node being in the opposite direction of the anchor node, and the position of the dumb node not within the range of anchor node. However, the localization error generated by the Amorphous algorithm is high. To address the limitations of Amorphous algorithm we have proposed a PSO based Amorphous algorithm. The proposed work reduces the average hop size of anchor nodes and reduces the localization error. The simulation results demonstrate that, in comparison to other existing Amorphous algorithms, the proposed PSO based Amorphous localization algorithm produced a superior performance in terms of MAE, MSE and RMSE.

As the volume of data stored on local devices increases, users turn to edge devices to help with processing tasks. Developing offloading schemes is challenging due to the varying configurations of edge devices and user preferences. While traditional methods provide schemes for offloading in various scenarios, they face unavoidable challenges, including the requirement to manage device workloads in real-time, significant computational costs, and the difficulty of balancing multi-objectives in offloading schemes. To solve these problems, we propose the Task Offloading Decision Optimizer, which offers efficient multi-objective offloading schemes that consider real-time device workload and user preference. The proposed offloading scheme contains three goals: reducing task execution time, decreasing device energy consumption, and lowering rental costs. It comprises two essential parts: Scheme Maker and Scheme Assistor. Scheme Maker utilizes deep reinforcement learning, optimizes the internal architecture, and enhances the performance of the operation. It optimizes buffer storage to generate dependable multi-objective offloading schemes considering real-time environmental conditions. Scheme Assistor utilizes the data in the Scheme Maker buffer to enhance efficiency by reducing computational costs. Extensive experiments have proved that the proposed framework efficiently provides offloading schemes considering the real-time conditions of the devices and the users, and it offers offloading schemes that enhance task completion rate by 50%. Compared to the baseline, the task execution time is reduced by 12%, and the device energy consumption is reduced by 11.1%.

Aiming to large-scale data transmission requirements of resource-constrained IoT (Internet of Things) devices, the routing protocol for low power lossy network (RPL) is expected to handle the load imbalance and high energy consumption in heavy traffic scenarios. This paper proposes a novel RPL routing optimization Algorithm based on deep Reinforcement Learning (referred to as RARL), which employs the centralized training and decentralized execution architecture. Hence, the RARL can provide the intelligent parent selection policy for all nodes while improving the training efficiency of deep reinforcement learning (DRL) model. Furthermore, we integrate a new local observation into the RARL by exploiting multiple routing metrics and design a comprehensive reward function for enhancing the load-balance and energy efficiency. Meanwhile, we also optimize the Trickle timer mechanism for adaptively controlling the delivery of DIO messages, which further improves the interaction efficiency with environment of DRL model. Extensive simulation experiments are conducted to evaluate the effectiveness of RARL under various scenarios. Compared with some existing methods, the simulation results demonstrate the significant performance of RARL in terms of network lifetime, queue loss ratio, and packet reception ratio.