Human Activity Recognition (HAR) underpins applications in healthcare, rehabilitation, fitness tracking, and smart environments, yet existing deep learning approaches demand dataset-specific training, large labeled corpora, and significant computational resources. We introduce RAG-HAR, a training-free retrieval-augmented framework that leverages Large Language Models (LLMs) for HAR. RAG-HAR computes lightweight statistical descriptors, retrieves semantically similar samples from a vector database, and uses this contextual evidence to make LLM-based activity identification. We further enhance RAG-HAR by first applying prompt optimization and introducing an LLM-based activity descriptor that generates context-enriched vector databases for delivering accurate and highly relevant contextual information. Along with these mechanisms, RAG-HAR achieves state-of-the-art performance across six diverse HAR benchmarks. Most importantly, RAG-HAR attains these improvements without requiring model training or fine-tuning, emphasizing its robustness and practical applicability. RAG-HAR moves beyond known behaviors, enabling the recognition and meaningful labelling of multiple unseen human activities.

Hierarchical Federated Learning (HFL) frameworks place edge servers between IoT devices and the cloud server to reduce communication costs and preserve privacy. In practice, however, HFL must handle hierarchical non-IID data across both device and edge levels. At the edge-level, heterogeneity arises because devices connected to the same edge server often share geographic or contextual similarities, giving each server its own optimization goal aligned with its region-specific data distribution rather than with a shared global objective. Existing HFL methods largely ignore this distinction, focusing on training a single global model that can obscure severe underperformance at the edge-level with underrepresented data. Since edge servers often act as operational units, poor performance at an edge implies degraded service quality, undermining system reliability and user trust. We propose Personalized Hierarchical Edge-enabled Federated Learning (PHE-FL), a novel method that produces personalized edge models by adaptively integrating edge- and cloud-level knowledge based on the data distribution of each edge, without incurring additional computational overhead or compromising client privacy. We deploy edge-specific test sets at each edge to ensure its unique data distribution is accurately reflected during evaluation. To the best of our knowledge, this is the first work to explicitly address hierarchical data heterogeneity in a 3-level HFL framework, both in terms of personalization and evaluation. Extensive experiments show that PHE-FL achieves up to 83% higher accuracy than existing edge-accommodated FL methods and maintains robust performance across edge-level non-IIDness, with reduced accuracy fluctuations compared to the state-of-the-art FedAvg with two levels (edge and cloud) aggregation.

Artificial intelligence of things (AIoT) is an emerging paradigm integrating artificial intelligence (AI) technologies within the Internet of Things (IoT) paradigm. However, deploying deep-learning models on IoT devices is challenging due to their inherent computational, communications, and security constraints. To address these challenges, we propose Flexible Neuron Selection (FlexNS), a computation- and communication-efficient personalised multi-task transfer learning framework for AIoT.FlexNSenables IoT devices to train their private task-specific shallow models by leveraging a multi-task, deep-learning model pre-trained by a cloud server.FlexNSsignificantly reduces IoT devices’ computational and communications resource demands by selecting a subset of neurons in an early layer of the server’s public model to be connected to the private models of multiple IoT devices. The neurons need to be carefully selected to ensure effective and efficient knowledge transfer to the fine-tuned private models tailored to each IoT device’s specific task. Experimental results show thatFlexNS-based private models achieve 104.3% and 98.4% model accuracy compared to the public model for two datasets on network intrusion detection and image classification tasks, with 99.5% and 98.0% reduction in training and inference time.

Collecting multidimensional user data is essential for personalized services, yet it poses significant privacy risks. While privacy regulations like the GDPR and CPRA advocate for data minimization, attribute correlations can inadvertently amplify unintentional information disclosure, leading to correlation-induced information leakage (CIL). Although data collectors often possess rich prior knowledge of these correlations, existing Local Differential Privacy (LDP) mechanisms are inadequate for effectively leveraging this information to reduce CIL. In this paper, we propose CoP, a coordinated perturbation mechanism designed to mitigate CIL in multidimensional data collection while preserving utility. Unlike traditional LDP approaches, CoP explicitly incorporates prior distribution knowledge to coordinate the perturbation process across attributes. By optimizing the perturbation strategy based on known correlations, CoP achieves a better privacy-utility trade-off. Extensive evaluations across both synthetic and real-world datasets demonstrate that CoP significantly outperforms state-of-the-art LDP mechanisms in reducing disclosure while preserving analytical accuracy.

WiFi devices are ubiquitous in modern environments, from smartphones and laptops to IoT sensors and AR/VR headsets. Identifying device types/models within these populations enables crowd analysis, network optimization, and detection of unusual devices. Current identification methods struggle with MAC address randomization, require large training datasets, and perform poorly in real-world deployments. This paper introduces a device identification method based on Information Element (IE) attributes extracted from WiFi probe requests. We evaluate the approach using probe requests captured in the 2.4 GHz band. Evaluation across 70+ device types yields 99% precision, 98% recall, and 99% F1 score, exceeding deep learning approaches (92% F1 score) under similar training conditions. Our approach maintains accuracy despite MAC randomization and requires minimal training data. We demonstrate practical applicability through an operational dashboard tested in real-world scenarios for urban planning and network management. Case studies across diverse environments confirm the effectiveness of the method for operational use.

Federated Learning (FL) enables privacy-preserving machine learning by allowing clients to collaboratively train models without sharing raw data. Federated Learning as a Service (FLaaS) extends this approach to cloud infrastructures. However, conventional secure aggregation protocols, such as Google's SecAgg and SecAgg+, introduce high computation and communication overheads, particularly in large-scale FLaaS deployments where client dropout rates are limited. To address these challenges, we propose ACCESS-FL, a lightweight, secure aggregation method designed for honest-but-curious FLaaS scenarios with stable network conditions. ACCESS-FL eliminates double masking, Shamir's Secret Sharing, and excessive encryption/decryption by creating shared secrets only between two peers per client, which reduces computation and communication complexity to constant$O(1)$and makes the algorithm independent of network size and comparable to standard FL. ACCESS-FL preserves privacy against inversion attacks and maintains model accuracy equivalent to the FL, SecAgg, and SecAgg+ protocols, proving that reducing overhead does not compromise learning performance and achieves communication and computation costs comparable to standard FL. Experimental evaluations on benchmark datasets (MNIST, FMNIST, and CIFAR-10) demonstrate lower overhead, making ACCESS-FL practical for service-based stable FLaaS applications such as healthcare analytics.

Mixed Reality systems for IoT control often rely on fragmented interfaces, cloud infrastructure, and lack spatial awareness. We present ViewIoT: a fully on-device framework that enables intuitive and spatially grounded interaction with nearby IoT devices using a head-mounted display. ViewIoT discovers and localizes devices by combining real-time 3D spatial meshes with lightweight wireless signal sensing, removing the need for manual setup or external infrastructure. At the core is a unified multimodal intent recognition pipeline, powered by an embedded language model, that interprets hand gestures, voice commands, gaze, and proximity signals to resolve user intent in under 100 milliseconds. Running on a Meta Quest 3S, our prototype supports real-time IoT control through gaze-based panels and natural interaction without relying on cloud services or pre-installed anchors. This work provides a practical foundation for seamless and infrastructure-free management of smart environments in a Mixed Reality setting.

Multiview systems have become a key technology in modern computer vision, offering advanced capabilities in scene understanding and analysis. However, these systems face critical challenges in bandwidth limitations and computational constraints, particularly for resource-limited camera nodes. This paper presents a novel approach for communication-efficient distributed multiview detection and tracking using masked autoencoders (MAEs). We introduce a semantic-guided masking strategy that leverages pre-trained segmentation models and a tunable power function to prioritize informative image regions. This approach, combined with an MAE, reduces communication overhead while preserving essential visual information. We evaluate our method on both virtual and real-world multiview datasets, demonstrating comparable performance in terms of detection and tracking performance metrics compared to state-of-the-art techniques, even at high masking ratios. Our selective masking algorithm outperforms random masking, maintaining higher accuracy and precision as the masking ratio increases. Furthermore, our approach achieves a significant reduction in transmission data volume compared to baseline methods, thereby balancing multiview tracking performance with communication efficiency.

Recent cyberattacks have increasingly targeted distributed networking environments like IoT networks. To detect these attacks, hidden under network traffic encryption, many centralized Machine Learning (ML) based solutions have been introduced, which are not well suited for IoT networks. This work proposes PIFL a practical approach to secure IoT networks by combining federated learning, in-network ML using P4-enabled devices, software-defined networks, and binarized neural networks. PIFL detects compromised edge devices and isolates them into separate network slices based on trust parameters derived from their behavior. We demonstrate the feasibility of PIFL using an experimental testbed with three intelligent network devices and seven IoT devices implemented on Raspberry Pi devices.

WiFi has emerged as the standard method for local connectivity across various devices, including smart assistants, IoT devices, smart TVs, and AR/VR devices. Identifying WiFi devices in neighborhoods has implications for law enforcement, urban planning, and socio-economic analysis. This paper introduces a novel approach to constructing WiFi device-type signatures using Information Element attributes from wildcard WiFi probe requests. Our method accurately identifies device types even when dealing with randomized MAC addresses and requires minimal training data, thus addressing limitations of existing machine learning and deep learning approaches. We evaluate our approach using a dataset of 51,726 probe requests across 50 device types, achieving an average F1 score of 99%, precision of 99%, and recall of 98% in device-type identification. Importantly, our method outperforms deep learning methods with significantly less training data, achieving a 92% F1 score with only one training sample per device type.

In many federated learning (FL) models, a common strategy employed to ensure the progress in the training process, is to wait for at least M clients out of the total N clients to send back their local gradients based on a reporting deadline T, once the parameter server (PS) has broadcasted the global model. If enough clients do not report back within the deadline, the particular round is considered to be a failed round and the training round is restarted from scratch. If enough clients have responded back, the round is deemed successful and the local gradients of all the clients that responded back are used to update the global model. In either case, the clients that failed to report back an update within the deadline would have wasted their computational resources. Having a tighter deadline (small T) and waiting for a larger number of participating clients (large M) leads to a large number of failed rounds and therefore greater communication cost and computation resource wastage. However, having a larger T leads to longer round durations whereas smaller M may lead to noisy gradients. Therefore, there is a need to optimize the parameters M and T such that communication cost and the resource wastage is minimized while having an acceptable convergence rate. In this regard, we show that the average age of a client at the PS appears explicitly in the theoretical convergence bound, and therefore, can be used as a metric to quantify the convergence of the global model. We provide an analytical scheme to select the parameters M and T in this setting.

Wireless fingerprinting is a device identification approach which leverages hardware imperfections and wireless channel variations as unique user-centric signatures. Recent studies have also demonstrated that user behavior can be used as a signature by collecting network traffic data, e.g., packet length, without the need to decode/decrypt the payload. Inspired by these results, we propose a multi-layer fingerprinting framework that jointly combines the multi-layer signatures for improved identification performance. In contrast to previous works in the area, our multi-view learning approach is rooted in the common information framework developed by Wyner [1] and is able to exploit data with multiple forms to enable the extraction of the user-centric signatures shared among the multi-layer features without the need for labels (i.e., unsupervised learning setup). We further use variational inference to obtain a computationally efficient algorithm based on a tight surrogate bound on the loss function. Our evaluation framework is based on a dataset obtained by combining real-world video traffic with simulated physical layer characteristics. Finally, our empirical results show that our Wyner Variational Autoencoder significantly outper-forms the state-of-the-art baseline in the unsupervised wireless fingerprinting setting.

Immense growth of network usage and the associated proliferation of network, traffic, traffic classes, and diverse QoS requirements pose numerous challenges for network operators. Though data-driven approaches can provide better solutions for these challenges, limited data has been a barrier to developing those methods with high resiliency. In this work, we propose SyNIG (Synthetic Network Traffic Generation through Time Series Imaging), which utilizes Generative Adversarial Networks (GANs) for network traffic synthesis by converting time series data to a specific image format called GASF (Gramian Angular Summation Field). With GASF images we encode correlation between samples in 1D signals on a single 2D pixel map. Taking three types of network traffic; video streaming, accessing websites and IoT, we synthesize over 200,000 traces using over 40,000 original traces generalizing our method for different network traffic. We validate our method by demonstrating the fidelity of the synthetic data and applying them to several network related use cases showing improved performance.

With the rapid growth of Industrial Internet of Things (IIoT) devices, managing an extensive volume of IIoT data becomes a significant challenge. While the conventional cloud storage approaches with centralized data centers suffer from high latency for large-scale IIoT data storage due to increased communication and latency overheads, distributed storage frameworks, such as blockchains, have become promising solutions. In this article, we design and analyze a dual-blockchain framework for secure and scalable distributed data management in large-scale IIoT networks. The proposed framework, namedMapChain-D, consists of a data chain that is mapped to an index chain to provide efficient data storage and lookup.MapChain-Dis designed for practical IIoT applications with storage, latency, and communication constraints. Detailed data exchange protocols are presented for data insertion and retrieval operations inMapChain-D. Based on these, theoretical analyses are provided on the space, time, and communication complexities ofMapChain-Dcompared with conventional single-chain frameworks with local and distributed data storage. We implement ourMapChain-Dprototype using open-source LoRaWAN communications with multiple Raspberry Pi and Arduino devices, Kademlia-based distributed hash table, and Ethereum-based blockchain with proof-of-authority consensus. Experimental results from our prototype show thatMapChain-Dis more suitable to be deployed on resource-constrained IIoT devices. We also highlight the scalability and flexibility ofMapChain-Dwith different number of edge nodes in the system.

Manual annotation of large-scale point cloud dataset for varying tasks such as 3D object classification, segmentation and detection is often laborious owing to the irregular structure of point clouds. Self-supervised learning, which operates without any human labeling, is a promising approach to address this issue. We observe in the real world that humans are capable of mapping the visual concepts learnt from 2D images to understand the 3D world. Encouraged by this insight, we propose CrossPoint, a simple cross-modal contrastive learning approach to learn transferable 3D point cloud representations. It enables a 3D-2D correspondence of objects by maximizing agreement between point clouds and the corresponding rendered 2D image in the invariant space, while encouraging invariance to transformations in the point cloud modality. Our joint training objective combines the feature correspondences within and across modalities, thus ensembles a rich learning signal from both 3D point cloud and 2D image modalities in a self-supervised fashion. Experimental results show that our approach outperforms the previous unsupervised learning methods on a diverse range of downstream tasks including 3D object classification and segmentation. Further, the ablation studies validate the potency of our approach for a better point cloud understanding. Code and pretrained models are available at https://github.com/MohamedAfham/CrossPoint.

360° videos are getting popular providing an immersive streaming experience for the user, nevertheless, demand high bandwidth in mobile networks due to their larger spherical frames. In this preliminary work, we propose to combine frame interpolation and super resolution methods to optimize tile based 360° video delivery by streaming them at low qualities in network and increasing the quality leveraging Multi Access Edge Computing. We propose a mechanism to adaptively decide this quality conversion at the client side which improves average video quality by 30% and bandwidth saving by 43.3% compared to existing tile based streaming.

360° videos have become popular creating an immersive streaming experience for the user, nevertheless, these videos demand high bandwidth in operational networks and have strict latency requirements. Viewport (VP) aware streaming with variable tiling has been proposed as a promising solution to reduce bandwidth consumption while providing fine granularity to the user VP. Content caching at the edge has also been proposed to reduce the delivery latency. Though, combining these two mechanisms have potential advantages, applying conventional tile-based caching, which primarily tries to find identical tiles, is not feasible due to the high diversity in tile area and location in variable tiles. To this end, we propose OpCASH, an ILP based mechanism to devise optimal cache tile configuration at a MEC server to provide a non-overlapping tile cover for a given VP request in variable tiles, while minimizing the requests to remote servers and reducing the delivery latency. Experimental trace-driven simulation results show that we can achieve more than 95% of VP coverage from cache after just 24 views of the video. Compared to a baseline which represents conventional tile-based caching, OpCASH reduces the data fetched from the content servers by 85% and total content delivery time by 74%.

360° videos a.k.a. spherical videos are getting popular among users nevertheless, omnidirectional view of these videos demands high bandwidth and processing power at the end devices. Recently proposed viewport aware streaming mechanisms can reduce the amount of data transmitted by streaming a limited portion of the frame covering the current user viewport (VP). However, they still suffer from sending a high amount of redundant data, as the fixed tile mechanisms can not provide a finer granularity to the user VP. Though, making the tiles smaller can provide a finer granularity for user viewport, it will significantly increase encoding-decoding overhead. To overcome this trade-off, in this paper, we present a computational geometric approach based adaptive tiling mechanism named VASTile, which takes visual attention information on a 360° video frame as the input and provides a suitable non-overlapping variable size tile cover on the frame. Experimental results show that VASTile can save up to 31.1% of pixel redundancy before compression and 35.4% of bandwidth saving compared to recently proposed fixed tile configurations, providing tile schemes within 0.98 (±0.11) seconds time frame.

Unlike conventional videos, 360° videos give freedom to users to turn their heads, watch and interact with the content owing to its immersive spherical environment. Although these movements are arbitrary, similarities can be observed between viewport patterns of different users and different videos. Identifying such patterns can assist both content and network providers to enhance the 360° video streaming process, eventually increasing the end-user Quality of Experience (QoE). But a study on how viewport patterns display similarities across different video content, and their potential applications has not yet been done. In this paper, we present a comprehensive analysis of a dataset of 88 360° videos and propose a novel video categorization algorithm that is based on similarities of viewports. First, we propose a novel viewport clustering algorithm that outperforms the existing algorithms in terms of clustering viewports with similar positioning and speed. Next, we develop a novel and unique dynamic video segment categorization algorithm that shows notable improvement in similarity for viewport distributions within the clusters when compared to that of existing static video categorizations.

Streaming 360° video demands high bandwidth and low latency, and poses significant challenges to Internet Service Providers (ISPs) and Mobile Network Operators (MNOs). The identification of 360° video traffic can therefore benefits fixed and mobile carriers to optimize their network and provide better Quality of Experience (QoE) to the user. However, end-to-end encryption of network traffic has obstructed identifying those 360° videos from regular videos. As a solution this paper presents 360NorVic, a near-realtime and offline Machine Learning (ML) classification engine to distinguish 360° videos from regular videos when streamed from mobile devices. We collect packet and flow level data for over 800 video traces from YouTube & Facebook accounting for 200 unique videos under varying streaming conditions. Our results show that for near-realtime and offline classification at packet level, average accuracy exceeds 95%, and that for flow level, 360NorVic achieves more than 92% average accuracy. Finally, we pilot our solution in the commercial network of a large MNO showing the feasibility and effectiveness of 360NorVic in production settings.

Unlike the traditional Internet application such as web browsing and peer-to-peer(P2P), video streaming has been dominating the global network traffic for the past few years, raising many challenges for network providers. With the popularity of interactive videos, a.k.a 360° videos, resource requirement for video streaming has been further increased. Prior identification of these video traffic is useful for effective provisioning of network resources, yet it is difficult due to the end-to-end encryption of data. However, with the recent advances in Machine Learning (ML) methods, prior identification of these resource-demanding traffic types has become viable. Nonetheless, they require more training data, without which leads to poor performance. Collecting more training data may also pose issues related to delayed training time. To remedy this problem, in this paper, we propose a novel Generative Adversarial Network (GAN) based data generation solution to synthesise video streaming data targeting 360°/normal video classification. Taking over 600 actual video traces and generating ≈ 30000 new traces, our post-classification results show that we can achieve 5 - 15% of accuracy improvement compared to only having actual traces.

The security and privacy of the end-users are a few of the most important components of a communication network. Though end-to-end encryption (e.g., TLS/SSL) fulfils this requirement, it makes inspecting network traffic with legacy solutions such as Deep Packet Inspection difficult. Recent Machine Learning techniques have shown outstanding performance in encrypted traffic classification. Nevertheless, such approaches require efficient flow sampling at real enterprise-scale networks due to the sheer volume of transferred data. Through this paper, we propose a holistic architecture to extract flow information of encrypted data at multi Gbps line rate using sampling and sketching mechanisms, enabling network operators to estimate flow size distribution accurately and understand the behavior of VPN-obfuscated traffic. Using over 6000 video traffic traces, under three main evaluation scenarios based on trace duration and starting time point, we show that it is possible to achieve 99% accuracy for service provider classification and over 90% accuracy for content classification for a given service provider in the best case. We also deploy our solution at an operational enterprise-scale network leveraging kernel bypassing to demonstrate its capability to efficiently sample live traffic for analytics.

Network communications, despite being encrypted, leak crucial information via side channels. WiFi networks are more prone to such side-channel attacks since any attacker within the network’s range can passively eavesdrop the channel. With the increasing number of smart home devices and sensors connecting to private WiFi networks, it is essential to understand the inadvertent information leakage through WiFi side-channels. Our work demonstrates how fine-granular information on the activities happening inside a house can be inferred by passively monitoring WiFi network traffic. In particular, we were able to correctly classify various user interactions with simple IoT devices such as smart bulbs or power sockets as well as advanced voice-based intelligent assistants.

While end-to-end encryption brings security and privacy to the end-users, it makes legacy solutions such as Deep Packet Inspection ineffective. Despite the recent work in machine learning-based encrypted traffic classification, these new techniques would require, if they were to be deployed in real enterprise-scale networks, an enhanced flow sampling due to sheer volume of data being traversed. In this paper, we propose a holistic architecture that can cope with encryption and multi-Gbps line rate with sampling and sketching flow statistics, which allows network operators to both accurately estimate the flow size distribution and identify the nature of VPN-obfuscated traffic. With over 6000 video traffic traces, we show that it is possible to achieve 99% accuracy for service provider classification even with sampled possibly inaccurate data.

This paper focuses on reporting of Internet malicious activity (or mal-activity in short) by public blacklists with the objective of providing a systematic characterization of what has been reported over the years, and more importantly, the evolution of reported activities. Using an initial seed of 22 blacklists, covering the period from January 2007 to June 2017, we collect more than 51 million mal-activity reports involving 662K unique IP addresses worldwide. Leveraging the Wayback Machine, antivirus (AV) tool reports and several additional public datasets (e.g., BGP Route Views and Internet registries) we enrich the data with historical meta-information including geo-locations (countries), autonomous system (AS) numbers and types of mal-activity. Furthermore, we use the initially labelled dataset of ~1.57 million mal-activities (obtained from public blacklists) to train a machine learning classifier to classify the remaining unlabeled dataset of ~44 million mal-activities obtained through additional sources. We make our unique collected dataset (and scripts used) publicly available for further research. The main contributions of the paper are a novel means of report collection, with a machine learning approach to classify reported activities, characterization of the dataset and, most importantly, temporal analysis of mal-activity reporting behavior. Inspired by P2P behavior modeling, our analysis shows that some classes of mal-activities (e.g., phishing) and a small number of mal-activity sources are persistent, suggesting that either blacklist-based prevention systems are ineffective or have unreasonably long update periods. Our analysis also indicates that resources can be better utilized by focusing on heavy mal-activity contributors, which constitute the bulk of mal-activities.

Mobile vision technologies have paved the way for augmented (AR) and mixed reality (MR) applications to be realizable on mobile devices. Mobile platforms such as Android and iOS have recently demonstrated the early opportunities for AR/MR applications using their devices. Now, while these technologies can still be considered in its infancy, it is opportune to start thinking about privacy and security while their functionalities are slowly being revealed to us. In this work, we present a visual access control mechanism in the form of object-level abstraction. Using readily-available object detection algorithms, we are able to demonstrate a proof-of-concept object-level abstraction for fine-grained access control in a mobile device. Furthermore, aside from the inherent confidentiality and content awareness guarantee of abstraction, reduction in execution times from visual processing resource sharing is another consequential benefit of abstraction without any energy consumption impact.

People may perform synchronized activities in a group setting. It is helpful to provide notifications to users and also the group leader whether people are in sync. This work aims to provide this support via analyzing motion data collected from wearable devices. We collected experimental data from smart watches worn by people, applied signal processing algorithms in both time and frequency domains for identification of the fine-grained group gesture status. We further developed a prototype system consisting of a smart watch, a smartphone, and a server. Our simulation results and actual system implementation demonstrate the feasibility of our approaches.

Mobile data traffic is growing exponentially and it is even more challenging to distribute content efficiently while users are “on the move” such as in public transport. The use of mobile devices for accessing content (e.g., videos) while commuting are both expensive and unreliable, although it is becoming common practice worldwide. Leveraging on the spatial and temporal correlation of content popularity and users' diverse network connectivity, we propose a novel content distribution system, uStash, which guarantees better QoE with regards to access delays and cost of usage. The proposed collaborative download and content stashing schemes provide the uStash provider the flexibility to control the cost of content access via cellular networks. We model the uStash system in a probabilistic framework and thereby analytically derive the optimal portions for collaborative downloading. Then, we validate the proposed models using real-life trace driven simulations. In particular, we use dataset from 22 inter-city buses running on six different routes and from a mobile VoD service provider to show that uStash reduces the cost of monthly cellular data by approximately 50 percent and the expected delay for content access by 60 percent compared to content downloaded via users' cellular network connections.

The prevalence of smart wearable devices is increasing exponentially and we are witnessing a wide variety of fascinating new services that leverage the capabilities of these wearables. Wearables are truly changing the way mobile computing is deployed and mobile apps are being developed. It is possible to leverage the capabilities such as connectivity, processing, and sensing of wearable devices in an adaptive manner for efficient resource usage and information accuracy within the personal area network. We show that app developers are not yet taking advantage of these cross-device capabilities, however, instead using wearables as passive sensors or simple end displays to provide notifications to the user. We thus design Application Function Virtualization (AFV), an architecture enabling automated dynamic function virtualization and scheduling across devices in a personal area network, simplifying the development of the apps that are adaptive to context changes. AFV provides a simple set of APIs hiding complex architectural tasks from app developers whilst continuously monitoring the user, device, and network context, to enable the adaptive invocation of functions across devices. We show the feasibility of our design by implementing AFV on Android, and the benefits for the user in terms of resource efficiency, especially in saving energy consumption, and quality of experience with multiple use cases.

Fitness related wearables have become ubiquitous in the recent past. Nevertheless, short battery life of these devices is still a pressing issue. Limited battery capacity in small form factor and power hungry continuous monitoring of accelerometer have been significant concerns in this regard. To address these issues we propose a novel low-power step counting solution based on an Electromagnetic energy harvesting mechanism. Extremely simple nature of the step counter removes the requirement of any step detection algorithm, thereby reducing the power consumption, while the energy harvester generates a portion of energy requirement prolonging the battery life.

The use of directional antennas in mobile wireless networks has emerged recently due to their numerous advantages such as high gain, low interference and low transmission power to reach the same distance as omni antennas. However, directional antennas limit the coverage and connectivity in the network. In this paper, we propose an algorithm to calculate a topology preserving map for wireless sensor networks using directional antennas. In the literature, topology mapping have been proposed for sensor networks equipped with omni antennas; however, to the best of our knowledge none of the existing work has considered the use of directional antennas in sensor network topology mapping. The application of existing algorithms in directional sensor networks will consume significantly high energy, as it requires transmitting the number of sectors times more beacon messages to cover the omini antenna range. Thus in this paper, a novel beacon forwarding algorithm limiting the number of forwarding sectors based on packet received direction is proposed. The simulation results show that it is possible to generate an accurate topology preserving map can be generated by only forwarding beacons in total of 135° angle. Moreover, the energy and the number of edges used to generate the topology map can be reduced by nearly 50% compared to the topology map generation with omni antennas based on received signal strength and hop count.

The smart devices usage is growing rapidly driven by innovative new smart wearables and service offerings. This has led to applications that utilize multiple devices around the body to provide immersive environments such as mixed reality that rely on a number of different types of functions and require considerable resources. Thus one of the major challenges in supporting these applications is dependent on the battery lifetime of devices that provide the necessary functionality. The focus of this paper is to improve the battery efficiency through intelligent resources utilization. We show that, when the same resource is available on multiple devices that form part of the wearable system, it is possible to consider them as a resource pool and further utilize them intelligently to improve the system lifetime via function virtualization. We formulate the intelligent function allocation algorithm as a Mixed Integer Linear Programming (MILP) optimization problem and propose an efficient heuristic solution. Next, we demonstrate the orchestration of the virtualized functions in order to achieve specific functionalities. The experimental data driven simulation results show that approximately 40-50% system battery life improvement can be achieved with proper function allocation and orchestration.

© 2018 IEEE. The proliferation of smart devices has led to an exponential growth in digital media consumption, especially mobile video for content marketing. The vast majority of the associated Internet traffic is now end-to-end encrypted, and while encryption provides better user privacy and security, it has made network surveillance an impossible task. The result is an unchecked environment for exploiters and attackers to distribute content such as fake, radical and propaganda videos. Recent advances in machine learning techniques have shown great promise in characterising encrypted traffic captured at the end points. However, video fingerprinting from passively listening to encrypted traffic, especially wireless traffic, has been reported as a challenging task due to the difficulty in distinguishing retransmissions and multiple flows on the same link. We show the potential of fingerprinting videos by passively sniffing WiFi frames in air, even without connecting to the WiFi network. We have developed Multi-Layer Perceptron (MLP) and Recurrent Neural Networks (RNNs) that are able to identify streamed YouTube videos from a closed set, by sniffing WiFi traffic encrypted at both Media Access Control (MAC) and Network layers. We compare these models to the state-of-the-art wired traffic classifier based on Convolutional Neural Networks (CNNs), and show that our models obtain similar results while requiring significantly less computational power and time (approximately a threefold reduction).

Cash-less payment via a variety of credit, debit or prepaid cards is pervasive in our interconnected society, but not so ubiquitous in remote rural regions where network connectivity is intermittent. We proposed a cash-less payment scheme for remote villages based on blockchains that allow maintaining a record of verifiable transactions in a distributed manner. We overcome the limitations of intermittent network connectivity by solely relying on blockchain mining nodes in the village for transaction processing and verification. The bank joins as a peer and monitors node behaviors, rewards miners and processes currency exchanges whenever the connectivity is available. We take advantage of the Ethereum network to develop our solution and demonstrate the feasibility of the proposed system on off-the-shelf computing devices. We emulate a remote village scenario with intermittent network connectivity and show the robustness and reliability of the proposed system.

Accurate estimation of crowd dynamics is difficult, especially when it comes to fine-grained spatial and temporal predictions. A deep understanding of these fine-grained dynamics is crucial during a major disaster, as it guides efficient disaster managements. However, it is particularly challenging as these fine-grained dynamics are mainly caused by high-dimensional individual movement and evacuation. Furthermore, abnormal user behavior during disasters makes the problem of accurate prediction even more acute. Traditional models have difficulties in dealing with these high dimensional patterns caused by disruptive events. For example, the 2016 Kumamoto earthquakes disrupted normal crowd dynamics patterns significantly in the affected regions. We first perform a thorough analysis of a crowd population distribution dataset during Kumamoto earthquakes collected by a major mobile network operator in Japan, which shows strong fine-grained temporal autocorrelation and spatial correlation among geographically neighboring grids. It is also demonstrated that temporal autocorrelation during disasters is more than simple diurnal patterns. Moreover, there are many factors that could potentially influence spatial correlations and affect the dynamics patterns. Then, we illustrate how a spatial-temporal Long-Short-Term-Memory (LSTM) deep neural network could be applied to boost the prediction power. It is shown that the error in terms of Mean Square Error (MSE) is reduced by as much as 55.1-69.4% compared to regressive models such as AR, ARIMA and SVR. Furthermore, LSTM outperforms the aforementioned models significantly even when little training data is available right after the mainshock. Finally, we also show a Region-aware LSTM does not necessarily outperform a regular LSTM.

Millimeter wave communication shows promise in realizing next generation wireless sensor networks for bandwidth demanding applications. Despite its support of multi Gbps data rates, MmWaves requires unobstructed line-of-sight and suffers from heavy path losses. Overcoming these in complex 3D environments requires sectored antenna arrays with narrow beam widths and adaptive beamforming. Therefore, network topology maps would be significant than ever in millimeter wave sensor networks. Traditional topology mapping algorithms rely on omnidirectional transmission and reception and are thus not tailored to such networks. A novel topology mapping algorithm, Millimeter Wave Topology Map (MmTM) is proposed for 3D deployments, which take advantage of the directional information available from beamforming antennas as well as their beam steering capability. An autonomous robot traverses the network recording the packet reception from different nodes along with the receiving antenna sector ID that delivers the packet with highest signal quality. The techniques used in standard IEEE 802.11ad protocol are used for the optimum sector selection and collision avoidance. MmTM is evaluated using two realistic sensor network environments and compared with prominent localization approaches based on received signal strength and hop count. The results show that proposed algorithm has a less than 0.7m distance error and more than 50% of nodes are located in the correct direction, which is 7m and 35% improvement in distance error and sector displacement matrices compared to other algorithms.

The majority of available wearable computing devices require communication with Internet servers for data analysis and storage, and rely on a paired smartphone to enable secure communication. However, many wearables are equipped with WiFi network interfaces, enabling direct communication with the Internet. Secure communication protocols could then run on these wearables themselves, yet it is not clear if they can be efficiently supported.,,,,In this paper, we show that wearables are ready for direct and secure Internet communication by means of experiments with both controlled local web servers and Internet servers. We observe that the overall energy consumption and communication delay can be reduced with direct Internet connection via WiFi from wearables compared to using smartphones as relays via Bluetooth. We also show that the additional HTTPS cost caused by TLS handshake and encryption is closely related to the number of parallel connections, and has the same relative impact on wearables and smartphones.

Dynamic Adaptive Streaming over HTTP (DASH) is one of the most popular ways to stream videos at present. In this work, we propose a DASH player energy-aware plugin (eDASH) for mobile devices which help reduce the battery consumption of the device. The eDASH player utilises a novel bitrate and video brightness adaptation algorithm to determine the next chunk to download. This algorithm utilises an energy-aware QoE model which factors in power consumption of the device in conjunction with existing bitrate adaptation logic to determine the next chunk. We also propose a new DASH architecture which could be easily integrated with the existing one. Macro-benchmarking of energy consumption of a mobile device while streaming and playing back video is conducted to obtain energy profiles of various video qualities. This energy data is then used along with real world network traces to drive simulations to evaluate energy savings that could be achieved using eDASH. We observe that up to 45% energy savings could be achieved with minimal reduction is QoE. We also find that up to 80% data transfer savings could also be achieved with an eDASH client.

Opportunistic communication, off-loading, and decentrlaized distribution have been proposed as a means of cost efficient disseminating content when users are geographically clustered into communities. Despite its promise, none of the proposed systems have not been widely adopted due to unbounded high content delivery latency, security, and privacy concerns. This paper, presents a novel hybrid content storage and distribution system addressing the trust and privacy concerns of users, lowering the cost of content distribution and storage, and shows how they can be combined uniquely to develop mobile social networking services. The system exploit the fact that users will trust their friends, and by replicating content on friends' devices who are likely to consume that content it will be possible to disseminate it to other friends when connected to low cost networks. The paper provides a formal definition of this content replication problem, and show that it is NP hard. Then, it presents a community based greedy heuristic algorithm with novel dynamic centrality metrics that replicates the content on a minimum number of friends' devices, to maximize availability. Then using both real world and synthetic datasets, the effectiveness of the proposed scheme is demonstrated. The practicality of the proposed system, is demonstrated through an implementation on Android smartphones.

Smart wearable devices are widely available today and changing the way mobile applications are being developed. Applications can dynamically leverage the capabilities of wearable devices worn by the user for optimal resource usage and information accuracy, depending on the user/device context and application requirements. However, application developers are not yet taking advantage of these cross-device capabilities.

Mobile video traffic is exploding and it is particularly challenging to stream video when high density of users are "on the move", e.g., in public transport systems. It becomes increasingly problematic as video traffic is predicted to account for more than 80% of Internet traffic by 2019. This will be exacerbated by factors such as cellular network coverage issues and unstable network throughput due to high speed mobility. By exploiting the predictable public transport mobility patterns, spatio-temporal correlation of user interests and users' video viewing behaviors, we proposed TransFetch which uses intelligent caching on-board the public transport vehicles as well as a novel video chunk placement algorithm. We show through extensive simulations, that TransFetch reduces the system cellular data usage by up to 45% and improves the quality of video streaming by up to 35%. Finally, we demonstrate the practical feasibility of TransFetch by implementing caching units on a Raspberry-Pi and a mobile app on an Android device.

Yalut is a novel user-centric hybrid content sharing overlay for social networking. Yalut enables the users to retain control over their own data and preserve their privacy, whilst still using the popular centralized services. In this demonstration, we show the feasibility of Yalut by integrating the service with the popular social networking apps on Android devices, Mac and Windows desktop platforms. We show that it is possible to provide the benefits of distributed content sharing on top of the existing centralized services with minimal changes to the content sharing process.

Crowd-Cache is a novel crowd-sourced content caching system which provides cheap and convenient content access for mobile users. Our system exploits both transient colocation of devices and the spatial temporal correlation of content popularity, where users in a particular location and at specific times would be likely interested in similar content. We demonstrate the feasibility of Crowd-Cache system through a prototype implementation on Android smartphones.

Distributed social networking services show promise to solve data ownership and privacy problems associated with centralized approaches. Smartphones could be used for hosting and sharing users data in a distributed manner, if the associated high communication costs and battery usage issues of the distributed systems could be mitigated. We propose a novel mechanism for reducing these costs to a level comparable with centralized systems by using a connectivity aware replication strategy. We develop an algorithm for grouping devices into tribes for content replication among intended content consumers and serve it using low-cost network connections. We evaluate the performance of the algorithm using three real world trace data sets. The results show that a persistent low-cost network availability can be achieved with an average of two replicas per content. Additionally, cellular bandwidth consumption and energy consumption of users are evaluated analytically using user content creation and consumption modeling. The results show that the proposed mechanism lowers monetary and energy costs for users compared to non-mobile-optimized distributed systems irrespective of the content demand model.

We focus on dissemination of content for delay tolerant applications, (i.e. content sharing, advertisement propagation, etc.) where users are geographically clustered into communities. We propose a novel architecture that addresses the issues of lack of trust, delivery latency, loss of user control, and privacy-aware distributed mobile social networking by combining the advantages of decentralized storage and opportunistic communications. The content is to be replicated on friends' devices who are likely to consume the content. The fundamental challenge is to minimize the number of replicas whilst ensuring high and timely availability. We propose a greedy heuristic algorithm for computationally hard content replication problem to replicate content in well-selected users, to maximize the content dissemination with limited number of replication. Using both real world and synthetic traces, we show the viability of the proposed scheme.

We proposed MobiTribe which enables device centric social networking on smart mobile devices while reducing high communication costs and battery usage normally associated with mobile distributed systems. In this paper, we demonstrate the feasibility of MobiTribe by integrating the service with the popular social networking application Facebook. We show that it is possible to provide the benefits of distributed content sharing on top of the existing centralized social networking services with minimal changes to the content sharing process.

Distributed social networking services show promise to solve data ownership and privacy problems associated with centralised approaches. Smartphones could be used for hosting and sharing users data in a distributed manner, if the associated high communication costs and battery usage issues of the distributed systems could be mitigated. We propose a novel mechanism for reducing these costs to a level comparable with centralised systems by using a connectivity aware replication strategy. To this end, we develop an algorithm based on a combination of bipartite b-matching and a greedy heuristics for grouping devices into tribes among intended content consumers. The tribes replicate content and serve it using low-cost network connections by exploiting time elasticity of user generated content sharing. The performance is evaluated using three real world trace data sets. The results show that a persistent low-cost network availability can be achieved with an average of two replicas per content. Additionally, a content creator can reduce 3G traffic by up to 43% and device energy use by up to 41% on average compared to content sharing in non-mobile-optimised distributed social networking approaches. Moreover, the results show that the proposed mechanism can provide the benefits of a distributed content sharing system for monetary and energy costs comparable to those of a centralised server based system.

An increasing portion of traffic in mobile networks conies from users creating content and uploading it to the Internet to share it. The capacity of mobile networks is a limited resource and uploading high resolution content consumes a large part of it. We introduce MobiTribe, a distributed storage cloud consisting of mobile devices for storing the content created on the phones. It can serve requests for content and take advantage of networks with spare capacity to deliver the content at a lower cost. We propose a content distribution and replication algorithm which achieves this goal. The performance of the algorithm is evaluated using empirical data traces of WLAN availability patterns of mobile devices, showing that it is possible to achieve 99.98% availability of a content via WLAN while minimising content distribution to an average of 2.69 replicas.