In this article, we propose and implement i-AVR, an Internet of Things (IoT)-based critical-aware system for point-of-care recommendation during ambulatory in-transits. The delay due to ambulances stuck in traffic congestion, disruptive roadways, and far-away hospitals restrain the smooth ambulance services. Therefore, in order to assist the time-critical scenario of a hospital-bound patient, we consider a guidance system to address the necessity. Moreover, these patients require continuous vitals monitoring, which may vary with the progress of time, to reduce the response time upon reaching the destination. The implemented i-AVR comprises two units: 1) a portable healthcare unit and 2) an android navigation unit. The healthcare unit aims to compute the criticality index of the en-route patient and recommend the nearest healthcare center while the navigation unit recommends the convenient route in case of any anomaly in vitals. We show the effectiveness of i-AVR regarding network performance while highlighting the response time of the system. We observe the system response time for computation in orders of seconds and interunit communication in milliseconds. Eventually, this analysis indicates the effectiveness of i-AVR in providing quick decisions during the time-critical situations. Our implementation provides essential intervention toward IoT-based healthcare technologies.

In this work, we proposeShadows, a virtual blockchain (VC) for achieving parallel consensus and efficient management of data in industries by utilizing BC. Typically, industrial processes involve heterogeneous activities which require real-time consensus, managed execution, isolation, data sharing, accelerated computation, and efficient utilization of various computational resources such as CPU, RAM, and storage. Achieving these in real-time using a single conventional blockchain (BC) leads to the exertion of computational power. To achieve resource-efficient real-time consensus, we virtualize the nodes of the BC network and create different BC for various activities. Further, to virtualize BC and provide better access to data, we propose smart contracts liable for providing a unified view of a single BC, dynamically creating BCs, allocating resources to these, and making communication between the same. Through lab-scale experiments, we demonstrate thatShadowsis capable of utilizing the resources efficiently and achieving real-time consensus. In particular,Shadowsuses 18% CPU and 92% memory while reducing consensus time by 56%, compared to a single conventional BC.Shadowsalso accesses the data efficiently by utilizing smart contracts and dynamically balances the load by migrating the virtual nodes. Further,Shadowsreduces the number of migrations to make the balance system by 67%.

The social IoMT-based activity-monitoring system comprises several devices with different datasets. It faces challenges like a collection of a global activity dataset which comprises a myriad of activities. In this article, we propose a federated Siamese network-based data-independent group activity segregator—Skipper—which aims to identify anomalies in an activity-monitoring social IoMT system. The novelty of this work is that Skipper does not require any dataset before its deployment, which removes the need for any prior training of the model for activity monitoring. As a proof of concept, we select activities pertaining to school environments to identify low-performing students in a classroom, who would require teachers’ close attention to ensure balanced growth and proper health. Skipper monitors the students independently for their motion signatures through a wearable device that consists of an accelerometer. A federated Siamese network calculates indices that signify the degree of similarity among the students’ activities. Skipper identifies the students who do not perform the same activity. With real-world implementations, we observe that Skipper requires network rates of 10 Kb/s, making it suitable for low bandwidth networks while we achieve just 20% CPU and 10 MB memory utilization on constrained edge devices. Further, with an increasing number of students up to 100, the time delay for final results is limited to 80 s. Hence, Skipper is a fast, easy, and accurate solution for recognizing outliers in IoMT social systems.

Storing data from Industrial-Internet-of-Things (IIoT) sensors in blockchain (BC) for monitoring the applications leads to management issues like bloating. The crux of this work is generating traces (the part of industrial data) using an ARIMA model and storing only the metadata over the network, resulting in reduced delay and managed data. We determine the size of the traces for storing on the store and generate (S&G) blocks (blocks that store traces along with their metadata) by considering principal parameters, such as training time, block size, and error. In general, S&G consists of three phases: 1) categorizing the data into groups based on their sampling rates, 2) storing the trace of data and metadata into the blocks, and 3) retrieving the entire data. We demonstrate the feasibility of S&G with errors and regret in the range of 0.07–0.10 and 0.20–0.25, respectively, using the appropriate ARIMA model.

In this work, we propose Timed Loop Gears (TLG), as a distributed method for enablingfragmented learningin Resource-Constrained networked IoT edge devices. TLG identifies atomic operations (gears), such as feed-forward and back-propagation, necessary for training Machine Learning (ML) models. Each of these gears executes on a Fog Node (FN) exclusively for each data point at a time rather than the whole dataset in its entirety. Additionally, the networked Edge Devices (EDs) offload the training data to the fog layer using the Message Queuing Telemetry Transport (MQTT) protocol such that the participating FNs subscribe to incoming training data and store them based on topics, simplifying data sharing. TLG enables the FN to then transfer the partially learned weights to the next suitable FN for further training. This looping of weights is repeated across FNs until the training is complete. Through extensive analysis, we observe that, compared to existing distributed ML training approaches, for$n$devices, TLG reduces the probability of disruption due to device failure by$n^{2}$times. Implementation results of our fragmented learning method demonstrate that, although TLG negligibly increases the memory consumption of the IoT devices by$0.8\%$, it reduces CPU usage by almost$90\%$. The proposed method proves beneficial for developing and hosting ML models, even on constrained IoT devices, in contrast to existing lightweight ML methods.

This work presents an efficient data-centric client selection approach, named DICE, to enable federated learning (FL) over distributed edge networks. Prior research focused on assessing the computation and communication ability of the client devices for selection in FL. On-device data quality, in terms of data volume and heterogeneity, across these distributed devices is largely overlooked. The obvious outcome is the selection of an improper subset of clients with poor-quallity data, which inevitably results in an inefficient trained model. With an aim to address this problem, in this work, we design DICE which prioritizes the data quality of the client devices in the selection phase, in addition to their computation and communication abilities, to improve the accuracy of FL. Additionally, in DICE, we introduce the assistance of vicinal edge devices to account for the lack of computation or communication abilities in certain devices without violating the privacy-preserving guarantees of FL. Towards this aim, we propose a scheme to decide the optimal edge device, in terms of latency and workload, to be selected as the helper device. The experimental results show that DICE improves convergence speed for a given level of model accuracy. Further, the simulation results show that DICE reduces delay by at least 16%, energy consumption by at least 17%, and packet loss by at least 55% compared to the existing benchmarks while prioritizing the on-device data quality across clients.

In this work, we propose an environment-aware electrical safety assistant using smartphones in pervasive domains like industry, homes, and healthcare. Conventional methods involve using eye shields, gloves, finger guards, and safety toe shoes. We depend on IoT-based solutions and propose Magdroid, an autonomous and standalone smartphone application that detects any electrical anomaly around the user and alerts all the users in the network. Magdroid supports edge computing and provides in-app inferences without any dependency on remote servers. It extracts the in-built magnetometer readings to detect any electrical anomaly in the environment. Since the readings vary with the environments, Magdroid first senses and then uses a cascaded deep learning technique to predict the electrical anomaly around the smartphone. We use two Convolution Neural Network (CNN) architectures and cascade the inference of one with the input of another to generate efficient results for detecting electrical anomalies that are particular to that environment. The first model achieves a test accuracy of 98.97% for the prediction of the environment and the cascaded CNN achieves a test accuracy of 81.88% with 7.82% and 37.27% loss, respectively. Additionally, Magdroid is a low resource-consuming application that utilizes 8% CPU and 126.MB memory of the smartphone.

In this work, we present a solution toward facilitating dynamic encryption schemes—Constrained Encryption as a Service (CEaaS)—in fog-enabled IoT environments. CEaaS is a two-level fuzzy inference system (FIS) in the fog layer which offers customized encryption algorithm decisions to the IoT user devices based on the current configuration, data size, and network state. Fog nodes use the two-tier FIS system to determine the category of the requesting IoT device at the first level and then the encryption scheme at the second level. Existing research on encryption focuses on developing new lightweight algorithms as a global solution for all devices without considering the heterogeneity and corresponding communication links. The device and network configurations collectively add operational delays, which elevates time and security threats. Under such circumstances, a solution that considers both the conditions (varying) for determining the appropriate encryption scheme and the key is important. Through extensive implementation and deployment of heterogeneous fog nodes, we observe that CEaaS is feasible for both powerful and resource-constrained IoT user devices with CPU and memory usage as low as 0.24% and 0.9%, respectively. CEaaS also incurs delays in the range of 0.7 s and energy consumption of 0.07 Joules while securing data transmission. With the feasibility of CEaaS, the dynamic encryption schemes ensure secure communications irrespective of the device types in a fog-enabled IoT environment.

In this article, we propose a dynamic-consensus-based blockchain system—A-Blocks—for efficiently managing the data produced by the sensors in an Industrial Internet of Things (IIoT) environment. Typically, industries deal with a heterogeneous set of data from a diverse range of sensors. Conventional blockchain adoptions are a popular choice in such scenarios for data security while satisfying both transparency and immutability. However, stringent consensus algorithms are inadequate for managing heterogeneous data, especially due to its implicit constraints. For instance, while PoW provides inevitable security and is highly distributive, it is not scalable and requires more energy. In contrast, PoS is energy efficient but has reduced scalability and PBFT is suitable for faster processing. A-Blocks exploits the features of the available consensus algorithms and dynamically selects the best one in real time. It operates in two phases: 1) categorizing the data into groups based on their traits and then 2) selecting the appropriate consensus algorithm. Extensive experimental results using open industrial data sets demonstrate the effectiveness of A-Blocks with 8% CPU and 78% memory consumptions on resource-constrained devices. Furthermore, compared to the existing methods, although A-Blocks increases energy consumption by 11%, it also reduces mining time by 7%.

We propose the adaptation of vibration-based condition monitoring systems and techniques, popularly used in industrial condition-based maintenance, for identifying the possibility of compromising the privacy of personal computing systems. This work exploits the automated fan-based heat dissipation features and read/write operations of disk-based storage, commonly present in personal computers, to read computing task-specific vibration signatures on the computer’s cabinet/case. These vibration signatures are then used to identify the broad classes of tasks being executed on a separate computer without ever needing to log into the monitored machine. This work builds upon the premise that heterogeneous tasks have distinct computing requirements, which translates to variations in the amount of heat generated by the computer’s processor, eventually leading to variations in the computer’s heat control fan speed. The variations in the fan’s speed and the frequency of read/write operations to disk-based storage create unique vibration signatures, which maps uniquely to the computer’s processing operations, leading to a breach of privacy of the computer. Our work’s preliminary results suggest that computer-based tasks can be mapped from their vibration signatures with an accuracy of at least$70\%$. We additionally study the task identification granularity of such an approach.

IoT deployments that have limited memories lack sustained computation power and have limited connectivity to the Internet due to intermittent last-mile connectivity, particularly in rural and remote locations. For maintaining congestion-free operations, most of the collected data from these networks are discarded, instead of being transmitted remotely for further processing. In this article, we propose the paradigm Timed Loop Storage to distribute the data and use the underutilized bandwidth of local network links for sequentially queuing packets of computational data that are being operated on in parts in one of the IoT nodes. While the sequenced packets are executed sequentially on the target IoT device, the remaining packets, which are currently not being operated on, distribute and keep looping over the network links until they are required for processing. A time-synchronized packet deflection mechanism on each node handles data transfer and looping of individual packets. In our implementation, although we observe that the proposed approach requires data rates of 6 Mbps, it incurs only 45 Kb usage of primary storage systems even for sizeable data, ensuring scalability of the connected IoT devices' temporary storage capabilities, thereby making it useful for real-life applications.

In this paper, we present a method for dynamically selecting leaders in a master-slave communication model in a swarm of micro-Unmanned Aerial Vehicles (UAVs). With the growing size of the UAV swarm in complex missions, it becomes a challenge to control them for efficient execution of missions. In a traditional centralized communication model where all UAVs in the swarm are controlled directly through ground control, channel capacity limits the number of UAVs in the swarm which restricts the scalability. In the context of low-power miniature drones, we limit the communication of the ground Base Station (gBS) with only one UAV (leader) which controls the rest of the UAVs (followers). Towards this, we propose a greedy heuristic method for selecting the UAV leader that requires minimal time to communicate with the gBS in real-time. The proposed master-slave model enhances the scalability of the swarm by improving the utilization of channel resources. Simulation results demonstrate that the proposed dynamic leader selection enhances the lifetime of the entire network with a multifold decrease in energy consumption, compared to the state-of-the-art. Additionally, the lifetime of the network also decreases on operating with a single UAV leader. We also observe reductions in delays by almost 60% and an increase in data rate by 50%.

In this article, we present a Q-learning-enabled safe navigation system-S-Nav-that recommends routes in a road network by minimizing traveling through categorically demarcated COVID-19 hotspots. S-Nav takes the source and destination as inputs from the commuters and recommends a safe path for traveling. The S-Nav system dodges hotspots and ensures minimal passage through them in unavoidable situations. This feature of S-Nav reduces the commuter's risk of getting exposed to these contaminated zones and contracting the virus. To achieve this, we formulate the reward function for the reinforcement learning model by imposing zone-based penalties and demonstrate that S-Nav achieves convergence under all conditions. To ensure real-time results, we propose an Internet of Things (IoT)-based architecture by incorporating the cloud and fog computing paradigms. While the cloud is responsible for training on large road networks, the geographically aware fog nodes take the results from the cloud and retrain them based on smaller road networks. Through extensive implementation and experiments, we observe that S-Nav recommends reliable paths in near real time. In contrast to state-of-the-art techniques, S-Nav limits passage through red/orange zones to almost 2% and close to 100% through green zones. However, we observe 18% additional travel distances compared to precarious shortest paths.

The Internet-of-Things (IoT) environments have hard real-time tasks that need execution within fixed deadlines. As IoT devices consist of a myriad of sensors, each task is composed of multiple interdependent subtasks. Toward this, the cloud and fog computing platforms have the potential of facilitating these IoT sensor nodes (SNs) in accommodating complex operations with minimum delay. To further reduce operational latencies, we breakdown the high-level tasks into smaller subtasks and form a directed acyclic task graph (DATG). Initially, the SNs offload their tasks to a nearby fog node (FN) based on a greedy choice. The greedy formulation helps in selecting the FN in linear time while avoiding combinatorial optimizations at the SN, which saves time as well as energy. IoT environments are highly dynamic, which mandates the need for adaptive solutions. At the chosen FN, depending on the dependencies on the DATGs, its corresponding deadlines, and the varying conditions of the other FNs, we propose an ϵ-greedy nonstationary multiarmed bandit-based scheme (D2CIT) for online task allocation among them. The online learning D2CIT scheme allows the FN to autonomously select a set of FNs for distributing the subtasks among themselves and executes the subtasks in parallel with minimum latency, energy, and resource usage. Simulation results show that D2CIT offers a reduction in latency by 17% compared to traditional fog computing schemes. Additionally, upon comparison with existing online learning-based task offloading solutions in fog environments, D2CIT offers an improved speedup of 59% due to the induced parallelism.

The atmospheric buoyancy and intangible nature of fatal communicable viruses lead to rapid transmissions among individuals, resulting in global pandemics. Strategic lockdowns and mandatory social distancing are immediate solutions in such scenarios. However, this leads to operational disruptions in education, manufacturing, economy, transportation, governance, and community. Although technological assistance is beneficial in overcoming such issues, the current Internet of Things (IoT) infrastructure has limitations. In this article, we provide a comprehensive review of the possible IoT-based solutions that have the capacity of combating the COVID-19-like viruses. We highlight the societal impacts due to pandemics and identify the specific lacunae in current IoT solutions. We also provide comprehensive detail on how to overcome the challenges along with directions toward the possible technological trends for future research. Compared to existing reviews, our work offers a holistic view of the cause, effects, and the possible solutions that are existing, along with already existing solutions that can be customized to serve the special needs during the pandemic.

In this article, we propose a fog-enabled federated learning framework-FogFL-to facilitate distributed learning for delay-sensitive applications in resource-constrained IoT environments. While federated learning (FL) is a popular distributed learning approach, it suffers from communication overheads and high computational requirements. Moreover, global aggregation in FL relies on a centralized server, prone to malicious attacks, resulting in inefficient training models. We address these issues by introducing geospatially placed fog nodes into the FL framework as local aggregators. These fog nodes are responsible for defined demographics, which help share location-based information for applications with similar environments. Furthermore, we formulate a greedy heuristic approach for selecting an optimal fog node for assuming a global aggregator's role at each round of communication between the edge and cloud, thereby reducing the dependence on the execution at the centralized server. Fog nodes in the FogFL framework reduce communication latency and energy consumption of resource-constrained edge devices without affecting the global model's convergence rate, thereby increasing the system's reliability. Extensive deployment and experimental results corroborate that, in addition to a decrease in global aggregation rounds, FogFL reduces energy consumption and communication latency by 92% and 85%, respectively, as compared to state of the art.

In this article, we propose a lightweight blockchain-inspired framework-Magnum-as a magazine of transfer learning models in blocks. We propose the storage of these blocks on proximal fog nodes to simplify access to pretrained base models by industrial plants to tune them before deployment. We design Magnum for B5G-enabled scenarios to reduce the block transfer time. We formulate a demand-centric distribution scheme to further reduce the search and access time by adopting a nonlinear program model and solving it using the branch-and-bound method. Through extensive experiments and comparison with state-of-the-art solutions, we show that Magnum retains the accuracy of the models and present its feasibility with a maximum CPU and memory usage of 80% and 6%, respectively. Additionally, while Magnum requires a maximum of 10 s for writing models as large as 17 Mb on the blocks, it requires 16 ms for fetching the same.

In this paper, we develop and analyze a smart digital stethoscope - SkopEdge - to provide reliable remote e-health monitoring with a minimum delay while enhancing overall network performance. SkopEdge initially records the heart sounds from individuals and then senses the quality of the network. Depending on the network traffic, SkopEdge converts the audio clip into an appropriate format before transferring it to remote locations for estimating the number of heartbeats and storage. Towards this, we formulate the link quality along with SkopEdge's current configuration as a Markov Decision Process (MDP) with actions as conversion format selection. The remote server then returns the result, which SkopEdge displays on its screen. Real-time implementations show that SkopEdge works efficiently in all network conditions. Further, audio conversions usually degrade the quality of sound, but our proposed system does not change its primary components. Although SkopEdge exhibits an increase in energy consumption by 79% while converting to lower-quality formats, it also reduces the energy consumption by 99% while transmitting the same, which subsequently results in energy savings. Further, we provide an analysis of the estimated heartbeats in an audio clip by SkopEdge.