Stress detection has been widely studied using physiological signals. However, most research considers stress as a categorical level, limiting the ability to uncover stress’s latent continuous psychological structure. A quantitative representation of stress accommodates inter-individual differences, enhances measurement reliability and sensitivity to changes, and facilitates adaptive real-time applications. This study proposes a framework for estimating a Quantitative Stress Index (QSI), a continuous stress score derived from self-report questionnaires and estimated using physiological features. These features are extracted from Electrodermal Activity (EDA) and Heart Rate Variability (HRV) obtained from Blood Volume Pulse (BVP) signals collected via the Empatica E4 wrist-worn device. The results indicate that the physiological QSI model effectively differentiates between stress and relaxation states, showing improved performance compared to established approaches and demonstrating its potential for quantitative, real-time stress monitoring using wearable sensors.

Mobility models have significantly evolved in recent years to capture vehicular and pedestrian trajectories. However, modeling the mobility of manual and low-powered vehicles, such as e-bikes, is highly dependent on the terrains they traverse. Variations in terrain can significantly impact their acceleration and displacement along their trajectories. Accurately estimating the location of these vehicles is critical to ensuring that the underlying communication infrastructure adequately covers them. Maintaining reliable connectivity and coverage is pivotal in enabling communication during emergencies, and supporting vital tracking services. To achieve this, it is crucial to develop an accurate trajectory model that considers terrain variations, particularly how elevation changes affect a vehicle's displacement. This paper presents a terrain-aware mobility model, named Terrain-Aware Mobility (TaM), designed to enhance the accuracy of trajectory predictions based on the Gauss-Markov model. The goal of TaM is to effectively model mobility across varying terrain slopes, thereby enabling the optimal placement of wireless access points (WAPs) in under-served areas. TaM is also designed to factor in potential errors in displacement, utilizing a measure of error covariance to mitigate compounding errors. The results of adopting different mobility models are contrasted in this paper, and further insights on building probabilistic mobility models are presented.

In the face of growing demands for high-quality digital media, enhancing video streaming quality remains a significant challenge, particularly in environments with diverse internet connectivity and limited bandwidth. This paper proposes the Edge-enhanced Streaming (EES) scheme, which leverages edge computing and machine learning to upscale low-bitrate video frames to higher resolutions. Utilizing the distributed computational power of edge devices such as smartphones and laptops, our methodology involves segmenting each video frame into smaller sub-frames. These sub-frames are then processed using a Super-Resolution (SR) machine learning model across available edge devices within the network. This approach optimizes underutilized computational resources, improves processing times, and reduces energy consumption, making it a highly suitable approach for real-time video streaming applications. Furthermore, to address the challenge of device reliability, we incorporate a task replication strategy, ensuring consistent quality improvements even with potential fluctuations in device availability. We evaluate our proposed scheme using the PRIM dataset from the PIRM-SR Challenge. Extensive simulations demonstrate significant enhancements in video quality, confirming the effectiveness of our distributed SR technique in overcoming bandwidth constraints and improving user experience.

Edge computing can satisfy the requirements of latency-critical and data-intensive applications by exploiting com-putational resources of end devices. However, such devices inherently suffer from dynamic user behavior, cyclic task-switching, varying link qualities which often impact their reliability. In addition, in incentivized systems, they may often over-estimate their advertised capabilities and consequently fail on delivering. In this paper, we propose the Reputation-based Task Assignment and Replication (RTAR) scheme. RTAR is the first scheme that uses a black box approach to perform cost-efficient task replication that accounts for workers' reliability and preserves workers' privacy by not requiring or soliciting any information about their devices. RTAR incorporates a reputation model using beta distribution to estimate the worker's reputation based on past performance. We formulate the problem as an Integer Linear Program (ILP) that strives to maximize the overall reputation of recruited workers, while abiding by a certain budget limit for each task. We also propose the RTAR-Heuristic (RTAR-H) scheme. RTAR-H uses matching theory to solve the optimization problem in a time-efficient manner. Extensive evaluations show that RTAR yields 63% and 68% reduction in recruitment cost and number of replicas, respectively, compared to a baseline scheme that blindly maximizes the number of replicas. Moreover, RTAR-H closely approaches the optimal solution, rendering a small gap of up to 1% and 1.2% in terms of task drop rate and recruitment cost, respectively.

Improving Intrusion detection systems (IDS) is pivotal for securing networks from various elusive attacks, including DDoS, packet injection, and unauthorized access. While many IDS systems are successful in fending off attacks, they often suffer from either low accuracy or significant latency. This paper proposes an IDS framework that balances higher accuracy with computational complexity, focusing on decreasing dimensionality by utilizing clustering methods, along with Ant Colony Optimization, and Rough k-means algorithms. Additionally, the paper enhances existing feature selection algorithms through multi-objective optimization, eliminating superfluous features that do not contribute significantly to intrusion detection and reducing problem dimensions, which increases speed. The combination of rough clustering and optimization techniques leads to a desirable accuracy of 87% in low dimensions. The proposed framework is evaluated using two benchmark datasets, NSL-KDD and ISCX 2012, and compared to two leading approaches, namely ICA-BP and GA-BP, where our model surpassed their accuracy levels.

Dynamic cache allocation and pricing in Information-Centric Networks (ICNs) is a challenging problem, especially when multiple content producers and competing ICN cache service providers are involved. Many existing ICN caching schemes generalize their frameworks as producer-agnostic architectures while considering a single ICN cache service provider. Realistically, as ICNs grow, multiple cache providers will compete for valuable content that would generate higher cache hits, and the ecosystem will inevitably become market-driven. In this paper, we investigate the dynamic cache allocation and price determination problem considering a caching system consisting of multiple content producers who act as the buyers and multiple competing ICN cache providers who act as the sellers of the caching resources. We propose a novel reverse auction-based caching and pricing scheme named SEMRA that aims to maximize the caching benefits of content producers. Simulation results demonstrate how the proposed scheme improved ICN caching over several caching metrics across varying cache sizes and popularity skewness values. Future work in this domain is highlighted in the conclusion.

The span and accuracy of sensors being developed for Body Sensor Networks (BSNs) are rapidly improving, aiding well-being services, and paving the way for real-time Affective Communication. While traditionally accurate sensing techniques were either highly invasive or rudimentary in their data, the roadmap of developments ahead shall enable a multitude of dependable analysis of patient monitoring. However, given the spectrum of techniques and tools available, researchers in this domain may find it challenging to balance accuracy, invasiveness, responsiveness and other design parameters of health-centric BSNs. In this paper we analyze and contrast different sensor types and sensing inference techniques, along with the response metrics that they are able to deliver. We further elaborate on their association to responses from the nervous system, and how they correlate to different measures of well-being. To aid future research on AI-assisted patient monitoring, we present current benchmarks for assessing BSN data accuracy, and their cross-validation across physiological signals. The goal of this paper is to assist researchers and practitioners in identifying the best subset of sensors and methods that can be used for a BSN for Affective Communication, and enable future immersive health-interactions over the Tactile Internet.

IoT systems are advancing to enable higher levels of engagement and omnipresence. A critical yet uncharted domain lies in communicating affect across participants, especially in medical settings where emotions and expressions are pivotal, and eXtended Reality (XR) systems that rely on immersive virtualization of the participating parties. However, IoT systems seldom have the bandwidth or reliability to enable such services. In this demo, we present an experiment that leverages Edge Intelligence and Artificial Intelligence to extract and encode emotions at one edge, and communicate a low-footprint encapsulation of such emotions at the other edge. The proposed architecture is designed to reduce overall traffic and build on low-power video and display equipment, to realize Affective Semantic Communication (AffSeC). This demonstration shall represent AffSeC in a medical setting, where a patient interacts with a physician over a low-BW E2E route. The proposed scheme will be contrasted to standard video compression to demonstrate the efficacy and promise of this model.

Affective computing is an emerging field that aims to develop technologies capable of recognizing and responding to human emotions. However, during communication sessions, the exchange of a high volume of data can cause high latency. One approach to mitigating this issue is semantic communication, which may reduce the amount of data exchanged. Hereby, we propose a novel idea that utilizes semantic communication in affective computing by minimizing the amount of information exchanged between endpoints. Specifically, we examine a use case of a remote doctor application, where a patient’s emotions are captured, and their vital signs are obtained using wearable devices, with this information reported to a remote doctor. To reduce data exchange, we utilize semantic communication to extract the meaning of the conveyed information, rather than transmitting the raw information itself. This approach can enhance the efficiency of communication in URLLC applications and has the potential to improve patient outcomes.

The phonocardiogram (PCG) is a crucial acoustic tool for early detection and diagnosis of cardiac abnormalities. However, its use it typically limited to a physician’s presence, especially given its reliance on accurate signal acquisition. In this demonstration, we will show an IoT PCG setup that is designed for low-power IoT operation and communication. The designed system, named HeartState Deep will employ machine learning to generate a cardiac assessment along with the PCG output over an IoT network. This framework is designed to improve the accuracy of classification by integrating spatial features using a convolutional neural network and temporal features by using recurrent neural networks to capture diverse information from the PCG signal. This method is evaluated using the PhysioNet Challenge2016 as a benchmark dataset. The demo will demonstrate the efficacy of the system in utilizing Edge intelligence and IoT in delivering critical PCG data and assessments.

Edge Computing (EC) has been evolving towards harvesting latent yet underutilized computational resources of the Extreme Edge Devices (EEDs), such as autonomous vehicles, smartphones, and tablets. However, EEDs tend to be user-owned devices. This triggers a high level of uncertainty, the impact of which is mostly overlooked. Such uncertainty can stem from the potential loss of network connectivity, battery depletion, as well as the dynamic user access behavior that can affect the computational capability of EEDs and compromise the convenience of users. This uncertainty can profoundly impact the devices' reliability of executing the offloaded tasks. In this context, we propose the Replica Maximization at the Extreme Edge (RMEE) scheme. RMEE employs task replication to achieve maximum reliability and improve successful task execution while abiding by certain QoS requirements. Towards that end, RMEE aims to maximize the number of offloaded replicas for each task, while ensuring that the task execution delay is kept within a certain threshold. We formulate the task replication optimization problem as a Mixed-Integer Linear Program (MILP) and devise an analytical solution using the Karush-Kuhn-Tucker (KKT) conditions and Lagrangian analysis. Extensive simulations have shown that RMEE outperforms other baseline schemes that involve single and fixed number of replicas, in terms of drop rate, satisfaction ratio, and the number of replicas by up to 100%, 100% and 60%, and 95.1 % and 85.4%, respectively.

Fire-fighting drones will play a key role in aiding fire-fighting efforts, not only in forest surveillance but also in adopting an agile aid in wildfire fire extinguishing. One of the key challenges in scaling-up the use of fire-fighting drones is their ability to autonomously carry out their tasks, so that swarms of them could be rapidly dispersed. However, in certain situations, it is inevitable that a remote controller needs to be involved in steering/adjusting their operation due to the highly volatile operation environment, especially with heat columns. We build on the potential of the Tactile Internet (TI) in delivering near real-time haptic interaction with drones. We build on the IEEE P1918.1 standard TI architecture and present a deterministic state machine following detailing remote interaction with such drones, and present the message exchanges to support their remote operation. Furthermore, we present a fire-fighting drone use case focused on flight autonomy, remote TI control, and identifying the characteristics of drone in service.

Fire-fighting drones will play a key role in aiding fire-fighting efforts, not only in forest surveillance but also in adopting an agile aid in wildfire fire extinguishing. One of the key challenges in scaling-up the use of fire-fighting drones is their ability to autonomously carry out their tasks, so that swarms of them could be rapidly dispersed. However, in certain situations, it is inevitable that a remote controller needs to be involved in steering/adjusting their operation due to the highly volatile operation environment, especially with heat columns. We build on the potential of the Tactile Internet (TI) in delivering near real-time haptic interaction with drones. We build on the IEEE P1918.1 standard TI architecture and present a deterministic state machine following detailing remote interaction with such drones, and present the message exchanges to support their remote operation. Furthermore, we present a fire-fighting drone use case focused on flight autonomy, remote TI control, and identifying the characteristics of drone in service.

Edge networks provide ample resources for low-latency service recruitment, unlike remote resources in the Cloud. As such, smart devices and Internet of Things (IoT) nodes form a pool of Extreme Edge Devices (EED) that are within reach of Mist and Fog networks, providing significant advantages in latency, geographic cognizance, and reduced communication costs. EEDs are often recruited in Edge networks assuming they are reliable in their commitment to tasks. However, many EEDs may fail to fulfill their tasks because they operate under opportunistic approaches and are prone to intermittent connectivity. To ameliorate task failure, we aim to optimize task allocation under the assumption of failure. Additionally, we optimize CPU utilization to engage reliable EEDs, resorting to replication when needed to exceed a tunable reliability margin. We demonstrate the efficacy of our model in multiple scenarios and present future work in EED utilization.

In Information-Centric Networks (ICNs), caching decisions are mostly driven by request-centric mechanisms. Fluctuations in request rates and cache capacities typically have the most impact on where content would be cached. However, as ICNs expand in scale, producers may prefer certain nodes to cache their contents based on favorable properties, such as topological centrality, closeness of cache locations with respect to consumers, security measures, and service up-time. These factors would impact the valuation of caching nodes. Nevertheless, maximizing cache utilization via dynamic cache valuation is a challenging task in ICNs, largely due to inter-dependencies in model parameters. In this paper, we propose a novel caching model where content producers aim to dynamically valuate cache nodes to optimize caching. The model is built on a value-based utility function that considers dynamic and topological attributes of cache nodes, enabling a dynamic novel caching scheme named Max-Node Utility that aims to maximize caching utility. Simulation results demonstrate that Max-Node Utility outperforms current state-of-the art caching schemes by providing better caching utility, reducing access delay and increasing cache hit ratios across varying cache sizes and popularity skewness values. An outlook on the premise of producer-driven caching schemes is presented in the conclusion, to emphasize future directions in similar caching models.

While the Internet infrastructure is growing in reliability and responsiveness, the exponential growth of users, services and Internet of Things (IoT) systems are inevitably going to stretch the resources of Mobile Edge Computing (MEC). Careful planning and deployment of computing and networking resources could significantly aid service proliferation, but they would not suffice. That is, IoT systems are inherently tied to preemptive deployment in areas of interest, and seldom planned for agile deployment in areas of interest (e.g. forest fires) after it's service use has occurred. In remedy, this work will explore the utility of “summoning” drones that could augment MEC resources to improve responsiveness (e.g. introduce computational power within a single hop) or reduce the load on resource constrained devices in the field. The operation and coordination of such drones would build on established literature in Flying Ad Hoc Networks (FANets). This paper will focus on improving the agility of EC via drones in a coordinated FANet, aiming to enable rapid-dissemination of resources to EC systems that are experiencing high operational loads and/or lack of required resources. This is a building block towards systems that would pro-actively leverage “mobile resources” in anticipation of demand, and relinquish such resources when their local resources and accessible backbone would not suffice in meeting their SLAs.

Remote interaction and manipulation of physical objects is one of the promising directions in networked services across the connected world. In formalizing an enabling architecture, the IEEE P1918.1 Standard for the Tactile Internet (TI) is being developed to outline tactile interactions and democratizing skills. For the Tactile Internet to become a reality, advancements in network technologies, orchestrated with haptic-technologies and artificial intelligence, are necessary. The standardization of 5G, rolling out as the next generation of mobile communications, is paving the way for supporting the Tactile Internet vision. The IEEE P1918.1 Working Group (WG) has been developing the architecture, objectives, reasoning and future use cases of the Tactile Internet. This paper focuses on the potential for enabling TI-based remote surgery, and introduces a specific use case (cholecystectomy) that is mapped to the IEEE P1918.1 standard reference architecture. Challenging issues such as network design, Quality of Service (QoS), and multiplexing are taken into account and discussed. This work also emphasizes the role of deterministic message exchanges that would support Tactile Internet in carrying out a remote operation.

Recent studies into streaming media delivery suggest that performance gains from ubiquitous caching in Information-Centric Networks (ICN) may be negated by Dynamic Adaptive Streaming (DAS), the de facto method for retrieving multimedia content. Bitrate adaptation mechanisms, that drive video streaming, clash with caching mechanisms in ways that affect users' Quality of Experience (QoE). Cache performance also diminishes as consumers dynamically select content encoded at different bitrates. In this article we use this evidence to draw a novel insight: in adaptive streaming over ICN, bitrates should be prioritized alongside popularity and hit rates. We build on this insight to propose RippleCache as a family of cache placement schemes that safeguard high-bitrate content at the edge and push low-bitrate content into the network core. Doing so reduces contention of cache resources, as well as congestion in the network. To validate RippleCache claims we construct two separate implementations. We design RippleClassic as a benchmark solution that optimizes content placement by maximizing a measure for ICNs shown to have high correlation with QoE. In addition, our lighter-weight RippleFinder is then re-designed with distributed execution for application in large-scale systems. RippleCache performance gains are reinforced by evaluations in NS-3 against state-of-the-art baseline approaches, using standard measures of QoE as defined by the DASH Industry Forum. Our results demonstrate that RippleClassic and RippleFinder deliver content that suffers less oscillation and rebuffering, all while achieving the highest levels of video quality; thus indicating overall improvements to QoE.

The term Tactile Internet broadly refers to a communication network that is capable of delivering control, touch, and sensing/actuation information in real-time. The Tactile Internet is currently a topic of interest for various standardization bodies. The emerging IEEE P1918.1 standards working group is focusing on defining a framework for the Tactile Internet. The main objective of this work is to present the IEEE P1918.1 reference architecture framework for the Tactile Internet. The paper provides an indepth survey of various architectural aspects including the key entities, the interfaces, the functional capabilities, and the protocol stack. A case study has been presented as a manifestation of the architecture. Performance evaluation demonstrates the impact of functional capabilities and the underlying enablers on user-level utility pertaining to a generic Tactile Internet application.

Information-Centric Networks (ICN) are aiming to shift the current host-oriented Internet model towards a content-centric one, by focusing on highly scalable and efficient content distribution and retrieval. Content caching is a fundamental building block in ICN, optimized to enable fast, reliable, and scalable content distribution and delivery. In this paper, we propose a novel utility value-based caching scheme, named Max- Utility for maximizing the aggregated utility- value of an ICN cache service provider. This novel approach considers attributes of both the content, and its producer, to determine the aggregated value, and builds on a dynamic caching algorithm that aims to maximize the aggregated utility value to guide cache placement and replacement decisions. Simulation results demonstrate that, Max-Utility outperforms current state-of-the art caching schemes by providing better caching utility while significantly eliminating caching redundancy and incurring less access delay to retrieve good quality contents across varying cache sizes and popularity skewness values.

Video traffic is growing in dominance in today's Internet, prompting new challenges in timely delivery of video content. As Dynamic Adaptive Streaming (DAS) is becoming the de facto paradigm for video delivery, there is growing evidence on how caching improves users' Quality of Experience (QoE) in DAS. However, there is no consensus on how to maximize the utilization of in-network caching resources. Specifically, there are conflicting proposals on the impact of caching based on its distance (in hops) from the network edge. That is, contrasting ubiquitous network-wide caching to edge-caching. Supporters of the ubiquitous caching paradigm propose bitrate-aware caching schemes for optimizing video streaming, while counter-proposals suggest that edge-caching only the highest bitrate with online transcoding, may offer superior performance to ubiquitous caching. In this paper, we answer a contentious question: Can transcoding at the edge outperform bitrate-aware ubiquitous caching for DAS? We devise an extensive simulation environment using NS-3 to contrast both paradigms, experimenting with different bandwidth fluctuation patterns, under the FESTIVE user-based bitrate adaptation protocol. Caching performance was evaluated under five established QoE metrics, gauging delivered video quality, playback freezing and bitrate oscillation. We further assume zero processing delay for online transcoding at the network edge, to contrast to an upper bound performance from the edge-caching paradigm. Our experiments demonstrate that neither transcoding nor bitrate-aware caching offer a silver bullet for all cases. We present our insights on networking scenarios where each model would dominate in performance, and present our concluding remarks on their development.

The IEEE “Tactile Internet” (TI) Standards working group (WG), designated the numbering IEEE 1918.1, undertakes pioneering work on the development of standards for the TI. This paper describes the WG, its intentions, and its developing baseline standard and the associated reasoning behind that and touches on a further standard already initiated under its scope: IEEE 1918.1.1 on “Haptic Codecs for the TI.” IEEE 1918.1 and its baseline standard aim to set the framework and act as the foundations for the TI, thereby also serving as a basis for further standards developed on TI within the WG. This paper discusses the aspects of the framework such as its created TI architecture, including the elements, functions, interfaces, and other considerations therein, as well as the novel aspects and differentiating factors compared with, e.g., 5G Ultra-Reliable Low-Latency Communication, where it is noted that the TI will likely operate as an overlay on other networks or combinations of networks. Key foundations of the WG and its baseline standard are also highlighted, including the intended use cases and associated requirements that the standard must serve, and the TI's fundamental definition and assumptions as understood by the WG, among other aspects.

The Tactile Internet (TI) is building on the premise of remote operation in perceived real-time, and enables a plethora of applications that involve immersive interactions. As we build a future for globalizing skills, delivering haptic feedback across continents, and immersing users in remote environments, we are faced with significant challenges in understanding the context of Tactile Internet interactions, which we refer to as tactile cognizance. The challenge of understanding a remote terminals' context impacts not only the quality and depth of haptic feedback, but our ability to deliver perceived real-time operation. That is, as we develop AI techniques to compensate for the inevitable delay in remote operation, we need more information about a terminal's context and interactions to improve our prediction of movement and feedback. The Internet of Things (IoT) is promising to interconnect billions of sensors, and augment multiple tiers of cognition to expedite and fine-tune sensory acquisition from heterogeneous contexts. In this paper, we will survey recent developments in the IoT, and novel techniques for cloudlet-based cyber foraging (i.e., edge computing) to project how Tactile Internet interactions could benefit from IoT contextualization. We present a taxonomy of edge IoT systems designed for rapid data acquisition, with an emphasis on systems that prioritize stringent reliability and latency mandates. This paper builds on edge computing techniques to propose a framework for multi-tiered cognition in the Tactile Internet to feed its signaling systems, and how future TI codecs could embed contextual information in haptic feedback.

Recent studies suggest that performance gains for content delivery over Information-centric Networks (ICNs) may be negated by Dynamic Adaptive Streaming (DAS), the de facto method for retrieval of multimedia content. The bitrate adaptation mechanism that drives video streaming appears to clash with generic ICN caching techniques in ways that affect users' Quality of Experience (QoE). Cache performance diminishes as video consumers dynamically select content encoded at different bitrates. Motivated by preliminary evidence suggesting the merits of bitrate-based cache partitioning, we introduce a scheme to dissect the cache capacity of routers along a forwarding path according to dedicated bitrates. To facilitate this partitioning, we propose a guiding principle RippleCache, which stabilizes bandwidth fluctuation while achieving high cache utilization by safeguarding high-bitrate content on the edge and pushing low-bitrate content into the network core. We further propose a cache placement scheme, RippleFinder, to realize this RippleCache principle and highlight its impact on users' QoE by cache partitioning. The performance gains are reinforced by evaluations in NS-3. Measurements show RippleFinder can significantly reduce bitrate oscillation, while ensuring high video quality, indicating overall improvement to QoE.

Internet of Things (IoT) systems are inherently built on data gathered from heterogeneous sources. In the quest to gather more data for better analytics, many IoT systems are instigating significant challenges. First, the sheer volume and velocity of data generated by IoT systems are burdening our networking infrastructure, especially at the edge. The mobility and intermittent connectivity of edge IoT nodes are further hampering real-time access and reporting of IoT data. As we attempt to synergize IoT systems to leverage resource discovery and remedy some of these challenges, the rising challenges of Quality of Information (QoI) and Quality of Resource (QoR) calibration, render many IoT interoperability attempts far-fetched. We survey a number of challenges in realizing IoT interoperability, and advocate for a uniform view of data management in IoT systems. We delve into three planes that encompass Big Sensed Data (BSD) research directions, presenting a building block for future research efforts in IoT data management.

The growing demand for video streaming is straining the Internet, and mandating a fundamental change in future networking paradigms. Current advancements in Information-centric Networks (ICN) promise a novel approach to intrinsically handling content dissemination, caching and retrieval. While streaming technologies are converging towards Dynamic Adaptive Streaming (DAS), in-network caching in ICN facilitates serving users with better video qualities, potentially beyond their actual bandwidth-mandated throughput. In this paper, we propose an assessment framework for adaptive video streaming to evaluate the performance of ICN caching schemes, and measure their impact on improving users' streaming experience. We present a thorough study of core performance metrics and adopt those metrics which are designed for video caching. We conduct experiments on a NS-3 based simulator, ndnSIM, and propose insights which will aid the development of future caching schemes that cater to inevitable bitrate variations.

As competing industries delve into the Internet of Things (IoT), a growing challenge of interoperability and redundant deployments is magnified. Specifically, as we augment more “things” in the IoT fabric, how will these components interact across their heterogeneity, let alone collaborate. In this paper, we address the core issue of component interaction and operation under the IoT umbrella. We present our contribution in the framework of wireless sensor networks (WSNs), as a founding block in the IoT. More importantly, we present a novel paradigm in the design of WSNs, to build a resilient architecture that decouples operational mandates from the nodes. We abstract IoT things as wirelessly interfaced components, which introduce functionality physically decoupled from their devices; boosting resilience, dynamicity, and resource utilization. This approach dissects the study of any IoT nodal capacity to its “connected” components, and empowers dynamic associativity between things to serve varying functional requirements and levels. It also enables reintroducing only the components required to suffice for network operation, or only those needed to meet a new requirement. More importantly, critical resources in the network will be shared within their neighborhoods. Thus network lifetime will relate to functional cliques of dynamic IoT nodes, rather than individual networks. We evaluate the cost effectiveness and resilience of our paradigm via simulations.

The growing demand for video content is reshaping our view of the current Internet, and mandating a fundamental change for future Internet paradigms. A current focus on Information-Centric Networks (ICN) promises a novel approach to intrinsically handling large content dissemination, caching and retrieval. While ubiquitous in-network caching in ICNs can expedite video delivery, a pressing challenge lies in provisioning scalable video streaming over adaptive requests for different bit rates. In this paper, we propose novel video caching schemes in ICN, to address variable bit rates and content sizes for best cache utilization. Our objective is to maximize overall throughput to improve the Quality of Service (QoS). In order to achieve this goal, we model the dynamic characteristics of rate adaptation, deriving caps on average delay, and propose DaCPlace which optimizes cache placement decisions. Building on DaCPlace, we further present a heuristic scheme, StreamCache, for low-overhead adaptive video caching. We conduct comprehensive simulations on NS-3 (specifically under the ndnSIM module). Results demonstrate how DaCPlace enables users to achieve the least delay per bit and StreamCache outperforms existing schemes, achieving near-optimal performance to DaCPlace.

The rising tide of data from Sensor Networks, Internet of Things devices and novel sensing systems (e.g. smart devices and wearable technology) are introducing a number of opportunities as well as challenges. While the diversity and abundance of sensing resources are providing a wealth of data for Information Services, we are facing growing challenges in coping with the volume, diversity, and inconsistency of data, both in reported values and measures of accuracy, in addition to challenges in interoperability among these systems to truly realize ubiquitous services that can harness information from aggregated data. At a time when real-time access to data is critical to many applications, especially decision-making processes, the status quo in coping with Big Sensed Data (BSD) is faltering. In this paper we build on recent advancements in interoperability, and frameworks for managing IoT, to present a framework for heterogeneous sensing in BSD (HetSense-BSD), which adopts a multi-phase approach in soliciting heterogeneous resources to serve Information services. We introduce a novel Selective Sensor Fusion (S2F) algorithm for pruning superfluous data at the source, to the reduce communication footprint of data with inferior quality, and better utilize access networks for delivering the best possible data from available resources to the services. We present a use case for HetSense-BSD, and elaborate on the design of this framework as a first milestone in harnessing the aggregated potential of ubiquitously available resources in novel sensing systems.

The growing demand for video streaming is straining the current Internet, and mandating a novel approach to future Internet paradigms. The advent of Information-Centric Networks (ICN) promises a novel architecture for addressing this exponential growth in data traffic, with ubiquitous caching to facilitate video delivery. In this paper, we present a novel in-network video caching policy in ICN, named StreamCache, catering to variable video contents with different sizes and bit rates. Our objective is improving the average throughput of users which consequently enhances the Quality of Experience (QoE), under the heterogeneity of users' devices and network conditions. StreamCache is a popularity-based policy, which operates distributively at routers, designed for the online processing in order to narrow the gap between the offline theoretical optimal solution and the real-world application. StreamCache operates in rounds, making caching decisions based on video request statistics and minimal cache coordination. We show that, StreamCache achieves near-optimal performance compared with the offline benchmark scheme, DASCache and outperforms current state-of-the-art protocols, such as ProbCache by presenting an elaborate evaluation carried out on ndnSIM, over NS-3.

Robust and prompt emergency response is a crucial service that smart cities should provide to citizens, communities, and corporations. Emergency management strategies that are currently supported by cities yield pre-determined protocols that can only handle well-understood incidents. However, there are incidents whose nature, shape, scale, and timing are not as predictable. The lack of adequate data management platforms to harvest emergency-related data from the proliferation of data sources scattered around a city is a major shortfall in current emergency response and risk assessment processes. We propose an improved information infrastructure to assist emergency personnel in responding effectively and proportionally to large-scale, distributed, unstructured natural and man-made hazards such as multi-vehicle accidents, outbreaks of human or animal diseases, major weather events, large fires, and terrorist attacks. The proposed infrastructure will crowdsource the multitude of human and physical sensing resources that can generate data about incidents (e.g. smartphones, sensors, vehicles, etc.) in order to build a comprehensive understanding of emergency situations and provide situational awareness and recommendations to emergency teams on the scene. Our infrastructure consists of three components: (1) large-scale crowdsensing and data quality valuation, (2) heterogeneous data integration and analytics, and (3) decision making, alternative generation and recommendations. Leveraging crowdsensing and heterogeneous data analytics will improve the response coordination to critical incidents and real-time incident management, which will contribute to saving lives and reducing injuries, improving the quality of life, and saving resources by deploying them more effectively.

Location-Based services are gaining momentum as an important advancement in context aware services. That is, empowering users to identify potential services in their current space, and the prospect for services that are able to target local users, are pushing interest in research and industry alike. This paper explores the use of non-audible sound as a communication medium to tag and access location based services and gain access to their pertinent information. We propose and demonstrate the indoor implementation of a prototype of a location-based service-enabling system for hand-held devices. The system allows users to use their hand-held devices to search and interact with available services in their surroundings. A beacon placed in the service location broadcasts a service code mappable to the services particular to that location, and encoded via an ultrasound signal. The hand-held device can then identify that signal and prompt the user with the available services. We detail the novel system design and the ensuing architecture, and demonstrate the viability of the system which is tested over a variety of environments and scenarios. We conclude with an overview of the wide range of applications of this system, and note how it can enhance the way clients access location based services.

The growing demand for video streaming is straining the current Internet, and mandating a novel approach to future Internet paradigms. The advent of Information-Centric Networks (ICN) promises a novel architecture for addressing this exponential growth in data-intensive services, of which video streaming is projected to dominate (in traffic size). In this paper, we present a novel strategy in ICNs for adaptive caching of variable video contents tailored to different sizes and bit rates. Our objective is to achieve optimal video caching to reduce access time for the maximal requested bit rate for every user. At its core, our approach capitalizes on a rigorous delay analysis and potentiates maximal serviceability for each user. We incorporate predictors for requested video objects based on a popularity index (Zipf distribution). In our proposed model, named DASCache, we present delay queuing analysis for cached objects, providing a cap on expected delay in accessing video content. In DASCache, we present a Binary Integer Programming (BIP) formulation for the cache assignment problem, which operates in rounds based on changes in content requests and popularity scores. DASCache reacts to changes in network dynamics that impact bit rate choices by heterogeneous users and enables users to stream videos, maximizing Quality of Experience (QoE). To evaluate the performance of DASCache, in contrast to current benchmarks in video caching, we present an elaborate performance evaluation carried out on ndnSIM, over NS-3.

Maintenance is a core challenge in all routing protocols. The utilization of IPv6 for Low Power and Lossy Networks (6LowPAN) resulted in the recent standardization of a dedicated routing protocol called RPL (Routing Protocol for Low Power and Losy Neworks). In 6LowPAN, a critical challenge exists in decreasing packet loss under the stringent energy-efficiency mandate to increase network longevity. Moreover, the challenge of failed nodes/links, and operating in a lossy environment where connections require rapid maintenance, present significant challenges. Recent attempts at routing maintenance in RPL presented advancements in handling failures, yet under reactive mechanisms that respond to failures and attempt to reduce network down-time. In this paper we design and implement a proactive RPL (Pro-RPL) maintenance scheme that enables the network to selectively predict and mitigate failures before they impact network connectivity. In Pro-RPL we capitalize on a suffering index that is associated with RPL nodes, and monitors their tendency to result in a failure. This dynamic monitoring is decentralized in nature, and presents a conforming yardstick across RPL nodes, to eliminate overhead in implementation and potential control-traffic over the network. We evaluate the efficiency of Pro-RPL in reducing packet loss, energy consumption and extending network lifetime via extensive simulations with the Cooja Simulator over the Contiki OS.

Peer-to-peer (P2P) systems present a unique medium for resource sharing among cliques of participants (peers) in a distributed and self-organized manner. With the advent of mobile users and the increasing power of mobile devices, the spectrum of P2P capabilities should scale. Peers establish transient or persistent relationships with other peers based on mutual interest. Communicating peers may use intermediary peers to forward communication messages, if a direct link is beyond their communication range. A critical design parameter is establishing a resilient communication topology, yet reduce the overhead of control messages required to instill and maintain it. This rises as a significant hindrance in mobile environments, which pose additional challenges on P2P networks due to the heterogeneity of nodes, limited resources, dynamic contexts in addition to the inherited wireless network stringencies. Thus far, efforts in establishing P2P networks via super peers (SPs) have been capped by considering a subset of peer properties to evaluate their candidacy. This paper presents RobP2P, a robust architecture to construct mobile P2P networks and efficiently maintain network state. RobP2P introduces a SP selection protocol based on a dynamic score function that takes into account peers’ capabilities and context, such as location and quality of connectivity. The paper also presents an agile utility function through which SPs can delegate monitoring responsibilities to comparably powerful and stable peers to ensure self-healing topology maintenance. We present an elaborate performance evaluation of RobP2P implemented on Network Simulator NS-3. Our results illustrate the efficiency of RobP2P, its resilience to failures, and the improvements in lowering overhead traffic while reliably maintaining the consistency of network state.

Much has been deliberated lately on the adaptability of Wireless Sensor Networks (WSNs) to transition into a Cloud-based paradigm. This divergence has been mainly attributed to enabling a dynamic design, larger spread and a more distributed control scheme for WSNs that are inherently static and data-centric. Thus, transitioning into a service-centric paradigm, with the “Cloud” as an enabler, seems appealing. In this paper we argue against the seemingly straight forward transition, and emphasize the pitfalls in transitioning WSNs to an inherently distributed architecture. We articulate on four grounds, temporal and spatial limitations, resilience measures, energy efficiency and functional decomposition. Sheer connectivity, as an intrinsic property that presents hindrances in all these factors, is addressed in light of each. Finally, we present insights into future progressions of WSNs that boosts their dynamic presence without impacting intrinsic design dimensions. This paper serves both as an analytic overview of current directions and hindrances, and an overview to where we can go next in remedy to current and projected bottlenecks in Cloud-based sensing systems.

We advocate for a novel paradigm in Wireless Sensor Networks (WSNs). As a technology, it has evolved to a scalable networking paradigm with minimalistic operational mandates. However, inherited design principles of static functionality, that are pre-determined at design stage, hinder WSN evolvement. More importantly, while we design WSNs to endure harsh environments and scale in both urban and remote settings, we neglect two major factors. The over-deployment of WSNs renders many sensing nodes redundant in functionality, and inflates the cost of running applications; not to mention the resulting medium contention. In this paper we present a novel approach to expanding the operational scale of WSNs by adapting to the environment in which it is deployed. That is, capitalizing on an organic approach in thriving on available resources in the region of interest to reduce deployment cost, and solicit incentivized interaction among communicating resources to deliver dynamic sensing. Not only does this span a new dimension of reliability, over garnered resources, but presents a novel approach to assigning sensing tasks to available resources in correlation to their abundance and serviceability. We present our performance evaluation of reduction in operational costs, and the uptake of sensing tasks by neighboring resources via extensive simulations. We aim to benchmark WSN operational versatility and present a rigorous basis for evaluating the ability of WSNs to resiliently scale to new applications as well as handle intermittent and permanent failures.

The abundance of sensed data and its correlation between wireless entities has recently increased significantly. Understanding the context of each entity in a given environment is not-trivial. Mainly due to the need of realizing an efficient scheme for context fusion over multiple sensing/polling technologies. In this paper we present a selective context fusion model that utilizes an integrated architecture which encompasses both RFID systems and information collected from Wireless Sensor Networks (WSNs). We integrate the identification capabilities of the former with the group intelligence sustained by the latter. A mediator acts as both the reader and relay (RR) node to communicate both technologies respectively. Thus, collecting context information from sensors and tags, then aggregating, filtering and carrying out analysis to selectively enhance the quality of context collected in its vicinity. As such, the network will fuse information over a multiplicity of devices. The goal of this system is to utilize contextual information about the devices generating the data to better the selection process. The filtration process eliminates irregularities in the data as well as redundancy. Moreover, a weighted function stresses the value of data generated by higher-end nodes. Weight is also attributed to log-based evaluation protocols that identify a reliability metric. To further strengthen the fusion approach, local RRs will collect and aggregate context information from neighboring RR nodes, as well as knowledge databases over the Internet. As a load balancing measure, and to avoid resource draining, participating nodes will have an inversely proportional likelihood of participation in providing context information as their contribution count increases. Our system is further elaborated upon via an extensive use case.

Wireless Sensor Networks (WSNs) are approaching an operational stalemate. While sheer emphasis on energy efficiency and resilience aid network longevity, WSNs face many hindering design principals. Prominently, an application-oriented view that isolates WSNs from ubiquitous networks, and nodes with static hardware and functional goals. We present a novel paradigm in the design of WSNs. Our goal is to achieve a resilient architecture that decouples operational mandates from the nodes. We present wirelessly interfaced components, which introduce functionality physically decoupled from the sensing nodes; boosting resilience, dynamicity and resource utilization. This approach dissects the study of nodal capacity to its “connected” components. It also enables re-introducing only the components required to suffice for network operation. More importantly, critical resources in the network will be shared within their neighborhoods. Thus network lifetime will relate to functional cliques of dynamic nodes. We present our paradigms with insights into application and design novelty.

Recent research endeavours are capitalizing on state of the art technologies to build a scalable Internet of Things (IoT). Envisioned as a technology to integrate the best of Wireless Sensor Networks and RFID systems, there is much promise for a global network of objects that are identifiable, track-able, and harmoniously informing. However, the realization of an IoT framework is hindered by many factors, the most pressing of which is attributed to the integration of these heterogeneous nodes and devices. A considerable subset of these nodes undergoes movement and dynamically enters and leaves the network backbone/topology. Routing packets and inter-nodal communication has received little attention; mainly due to the sheer reliance on the Internet as a backbone. However, spatially correlated entities in the IoT, and those which most often interact, would pose a significant overhead of communication if all intermediate packets need to be routed over distant backhauls. In remedy, we present a Pruned Adaptive IoT Routing (PAIR) protocol that selectively establishes routes of communication between IoT nodes. Since nodes in the IoT belong to different owners, we also introduce a pricing model to cater for the exchange of monetary costs by intermediate nodes to utilize their relaying resources. We also establish a cap on inter-nodal routing to dynamically utilize the Internet backbone if the source to destination distance surpasses a preset (case optimized) threshold. The PAIR routing protocol is elaborated upon, building upon the detailed system model presented in this paper. We finally present a use case to demonstrate the utility and practicality of PAIR in the heterogeneous IoT as it scales.

In a technology where multiple networks are often deployed in concurrency, significant resource underutilization is witnessed in Wireless Sensor Networks (WSNs). As the manufacturing and deployment costs drop, multiple networks are introduced in overlapping vicinities to satisfy new functional requirements. Mostly with dedicated goals and deterministic operation schemes, practitioners seldom investigate the usability of visible resources already deployed in the region of interest, their utilization and the accommodation for transient resources that “pass-by” with a set of functional capacities. This paper presents a framework for classifying resources that contribute to the set of functional capacities of WSNs deployed in a given region, and the mapping of functional requirements set by multiple applications on these resources. We present a utility function to cater for successful utilization of transient resources; highlighting their importance in WSN longevity as well as dynamicity. An optimal formulation is presented for this mapping, with tunable rounds that cater for the temporal behavior of the network and its constituting resources. A use case further explains this paradigm.

Today's abundance of sensors and their wireless/wired networks, coupled with a growing plethora of applications, necessitate a dynamic approach to the assignment of tasks to a network. The current practice in WSN design is almost always application specific, due to functional and resource tradeoffs that have justified much of the tailored research done so far. Identifying this as a major bottleneck in WSN advancement, this paper presents a new paradigm which decouples applications from WSN architectures and protocols. This paradigm views the network as an abundance of connected resources (hence functionalities) to match requirements of applications (old and new) based on utilization and feasibility factors. We present an elaborate abstraction of network resources, with detailed description of its governing utility attributes. Then we describe the view of applications as an aggregation of functional requirements based on a given set of resources. The intermediate mapping between applications and resources is then solved by a reduced linear optimization formulation, to realize the system as a whole. The paradigm is further explained via a multiple-application scenario and its representation and operation under our paradigm.

Tag collisions impose a significant hindrance to reading rates of Radio Frequency Identification systems. The parallel singulation approach, being a major milestone, clusters tags and autonomously interrogates each cluster in parallel. This technique reduces the number of tags being interrogated at a given time, reducing collisions, and achieves higher reading rates. However, such an approach faces two limitations as the number of clusters increase. The exponential increase in tag responses may hinder tag functionality due to energy spent on communication. Moreover, energy inefficiency is incurred at cluster-heads to process significantly more tag responses. These issues overshadow the promising benefits of employing parallel singulation. In this paper, we remedy such hindrances by proposing energy efficient enhancements to the parallel singulation technique. The essence of these enhancements lies in minimizing an important measure of communication overhead, referred to as tags traffic rate, which indicates the efficiency of interrogation cycles in communicating with all tags without incurring unnecessary overhead. Analyses carried out via simulation demonstrate significant improvements by the proposed schemes in reducing energy consumption of cluster-heads, without posing constraints on tag operations nor incurring significant degradation of reading rates.

Decentralized protocols offer high adaptability to topology changes prominent in Wireless Sensor Networks (WSN). Protocols resilient to topology changes stemming from nodes dying, being added, relocating or duty cycling, improve network performance in terms of lifetime and percent of events sensed and reported. Topology dependant protocols, such as cluster-based, face many hindrances especially in terms of scalability, dynamicity, and adapting to varying traffic rates. Accordingly, a novel approach is introduced in sensing, whereby a single node is elected to report a sensed event, in a decentralized manner, thereby avoiding redundant reports by other nodes which exhaust network resources. Election is based on the node with the highest likelihood of successfully reporting the event. This protocol is coupled with a localized multi-hop routing protocol, to route that report back to the sink, by electing the most reliable next-hop neighbor to relay the report. Simulation results demonstrate the increase in network lifetime, detection/reporting efficiency, and resilience to varying node density.

Prolonging network lifetime while efficiently detecting and reporting events are arguably the most important objectives of wireless sensor networks (WSNs). Different WSN protocols aim to improve such measures, yet partially focus on certain aspects (eg. reliability and time latency) and sacrifice others (eg. power efficiency) in application specific approaches. We present DMULD (decentralized multi-level duty cycling), a cross-layer design paradigm aiming at raising performance measures of general WSNs. It integrates tailored multi-level sleep states having varying levels of performance (hence energy consumption) with novel sensing, medium access control (MAC) and routing protocols. Nodes carry on tasks in a decentralized manner with efficient load balancing. DMULD is a dynamic model which is adaptable to application specific requirements, through fine tuning its parameters. DMULD was thoroughly simulated, examining the effects of varying its parameters on network lifetime and efficiency. It achieved over double the lifetime of multi-hop CSMA/CA.