In recent years, FPGA platforms have shown significant potential for accelerating artificial intelligence (AI) applications, particularly in Embedded AI. While various studies have explored adaptive AI deployment on FPGAs, there remains a gap in methodologies fully integrating software adaptability with FPGA hardware reconfigurability. This article presents a novel end-to-end co-design methodology for deploying adaptable and scalable Convolutional Neural Networks (CNNs) on FPGA platforms. The framework enhances computational performance and reduces latency by dynamically modifying hardware acceleration units by combining CNN architecture adaptability with dynamic partial reconfiguration of FPGA hardware. The proposed methodology enables automated synthesis and runtime customization of both hardware accelerators and CNN architectures, eliminating the need for iterative synthesis. This approach has been implemented and tested on a Xilinx XC7020 FPGA board for a CNN-based image classifier, achieving superior computation performance (0.68s/image) and accuracy (97%) compared to state-of-the-art alternatives.

Currently, many deep learning models are being used to classify COVID-19 and normal cases from chest X-rays. However, the available data (X-rays) for COVID-19 is limited to train a robust deep-learning model. Researchers have used data augmentation techniques to tackle this issue by increasing the numbers of samples through flipping, translation, and rotation. However, by adopting this strategy, the model compromises for the learning of high-dimensional features for a given problem. Hence, there are high chances of overfitting. In this paper, we used deep-convolutional generative adversarial networks algorithm to address this issue, which generates synthetic images for all the classes (Normal, Pneumonia, and COVID-19). To validate whether the generated images are accurate, we used the k-mean clustering technique with three clusters (Normal, Pneumonia, and COVID-19). We only selected the X-ray images classified in the correct clusters for training. In this way, we formed a synthetic dataset with three classes. The generated dataset was then fed to The EfficientNetB4 for training. The experiments achieved promising results of 95% in terms of area under the curve (AUC). To validate that our network has learned discriminated features associated with lung in the X-rays, we used the Grad-CAM technique to visualize the underlying pattern, which leads the network to its final decision.

Fog computing is a promising technology that can provide storage and computational services to future 6G networks. To support the massive Internet-of-Things (IoT) applications in 6G, fog computing will play a vital role. IoT devices and fog nodes have energy limitations and hence, energy-efficient techniques are needed for storage and computation services. We present an overview of massive IoT and 6G-enabling technologies. We discuss different energy-related challenges that arise while using fog computing in 6G-enabled massive IoT. We categorize different energy-efficient fog computing solutions for IoT and describe the recent work done in these categories. Finally, we discuss future opportunities and open challenges in designing energy-efficient techniques for fog computing in the future 6G massive IoT network.

With a high rise in the popularity of Internet of Things (IoT), mobile computing, and wearable devices, a huge amount of data is being generated. Running complex tasks such as that are machine learning-based with minimum energy consumption is a challenge. It requires complex algorithms to run locally such as on middleware fog within the proximity of the devices generating data, or globally in a cloud to analyze the acquired data and create robust and smart applications. However, it depends on the type of task execution policy applied at each level; local or global, to decide on energy and performance efficiency, since certain tasks are high in complexity. Hence, task execution will be hierarchically distributed among the IoT nodes, fog, and cloud. Given that, we present in this paper a three-tier IoT-fog-cloud model. We argue that with distributed task execution, we can achieve high scalability of IoT services, and manage the global energy consumption as well. As a proof-of-concept, we evaluate our three-tier architecture by taking into account computational tasks for various applications in IoT related to medical, multimedia, location-based, and text. We evaluate using real datasets, based on three scenarios: fog-only, cloud-only, and fog-cloud collaborative. Task execution policy (at fog/cloud) play a key role in efficiently processing a task (especially large tasks, such as in deep learning). Therefore, we take that into account and elaborate what types of policies suit what type of offloading environment (fog-only, cloud-only, or fog-cloud collaborative).

Fog computing (FC) has emerged as a complementary solution to the centralized cloud infrastructure. An FC node is available in closer proximity to users and extends cloud services to the edge of the network in a highly distributed manner. However, with an increase in streaming and delay-sensitive Internet-of-Things (IoT) applications, FC also needs to address the issue of higher latency while forwarding compute-intensive jobs to remote cloud data centers. Hence, there is a need to investigate the use of computational resources at the edge of the network. Volunteer computing (VC) offers a reduction in the cost of maintaining high-performance computing by making use of user-owned underutilized or idle resources, e.g., laptops and desktop computers closer to fog devices. We propose volunteer-supported FC (VSFC), as a computing paradigm, that explores the interplay of these two distributed computing domains to help minimize inherent communication delays of cloud computing, energy consumption, and network usage. To this effect, we have extended the iFogSim toolkit to support VSFC. Extensive simulations show that VSFC outperforms traditional FC-cloud computing by reducing delay by 47.5%, energy by 93%, and network usage by 92% under normal to heavy load conditions.

Human emotions are strongly coupled with physical and mental health of any individual. While emotions exbibit complex physiological and biological phenomenon, yet studies reveal that physiological signals can be used as an indirect measure of emotions. In unprecedented circumstances alike the coronavirus (Covid-19) outbreak, a remote Internet of Things (IoT) enabled solution, coupled with AI can interpret and communicate emotions to serve substantially in healthcare and related fields. This work proposes an integrated IoT framework that enables wireless communication of physiological signals to data processing hub where long short-term memory (LSTM)-based emotion recognition is performed. The proposed framework offers real-time communication and recognition of emotions that enables health monitoring and distance learning support amidst pandemics. In this study, the achieved results are very promising. In the proposed IoT protocols (TS-MAC and R-MAC), ultralow latency of 1 ms is achieved. R-MAC also offers improved reliability in comparison to state of the art. In addition, the proposed deep learning scheme offers high performance ([Formula: see text]-score) of 95%. The achieved results in communications and AI match the interdependency requirements of deep learning and IoT frameworks, thus ensuring the suitability of proposed work in distance learning, student engagement, healthcare, emotion support, and general wellbeing.

Rapid developments in industry 4.0, machine learning, and digital twins have introduced new latency, reliability, and processing restrictions in Industrial Internet of Things (IIoT) and mobile devices. However, using current information and communications technology (ICT), it is difficult to optimally provide services that require high computing power and low latency. To meet these requirements, mobile-edge computing is emerging as a ubiquitous computing paradigm that enables the use of network infrastructure components such as cluster heads/sink nodes in IIoT and cellular network base stations to provide local data storage and computation servers at the edge of the network. However, optimal location selection for edge servers within a network out of a very large number of possibilities, such as to balance workload and minimize access delay, is a challenging problem. In this article, the edge server placement problem is addressed within an existing network infrastructure obtained from Shanghai Telecom's base station data set that includes a significant amount of call data records and locations of actual base stations. The problem of edge server placement is formulated as a multiobjective constraint optimization problem that places edge servers strategically to balance between the workloads of edge servers and reduce access delay between the industrial control center/cellular base stations and edge servers. To search randomly through a large number of possible solutions and selecting those that are most descriptive of optimal solution can be a very time-consuming process, therefore, we apply the genetic algorithm and local search algorithms (hill climbing and simulated annealing) to find the best solution in the least number of solution space explorations. Experimental results are obtained to compare the performance of the genetic algorithm against the above-mentioned local search algorithms. The results show that the genetic algorithm can quickly search through the large solution space as compared to local search optimization algorithms to find an edge placement strategy that minimizes the cost function.

Abstract Fog computing is a promising technique to reduce the latency and power consumption issues of the Internet of Things (IoT) ecosystem by enabling storage and computational resource close to the end‐user devices with additional benefits such as improved execution time and processing. However, with an increase in IoT devices, the resource allocation and job scheduling became a complicated and cumbersome task due to limited and heterogeneous resources along with the locality restriction in such computing environment. Therefore, this paper proposes a cache‐based approach for efficient resource allocation in fog computing environment, while maintaining the quality of service. The proposed algorithm is realized using iFogSim simulator and a comprehensive comparison is presented with the traditional First Come First Served and Shortest Job First policies. The performance evaluation revealed that with the proposed scheme the execution time, latency, processing delays and power consumption decreased by 38%, 11.1%, 6%, and 17.8%, respectively, as compared to those of the traditional schemes.

Fog computing has emerged as a complementary solution to address the issues faced in cloud computing. While fog computing allows us to better handle time/delay-sensitive Internet of Everything (IoE) applications (e.g. smart grids and adversarial environment), there are a number of operational challenges. For example, the resource-constrained nature of fog-nodes and heterogeneity of IoE jobs complicate efforts to schedule tasks efficiently. Thus, to better streamline time/delay-sensitive varied IoE requests, the authors contributes by introducing a smart layer between IoE devices and fog nodes to incorporate an intelligent and adaptive learning based task scheduling technique. Specifically, our approach analyzes the various service type of IoE requests and presents an optimal strategy to allocate the most suitable available fog resource accordingly. We rigorously evaluate the performance of the proposed approach using simulation, as well as its correctness using formal verification. The evaluation findings are promising, both in terms of energy consumption and Quality of Service (QoS).

Modeling of Underwater Wireless Sensor Networks (UWSNs) with a goal of maximum network lifetime and throughput with minimum energy consumption is a quite difficult task because of limited battery power and harsh underwater environment. Balanced Energy Efficient Rectangular routing protocol (BEER) covers the maximum network area with the mobility of sinks and collects the data from sensor nodes in their transmission range using direct transmission. Sink movement maximizes the throughput and balanced the energy consumption. Simulation results verify that our scheme performs outstanding in terms of network lifetime, stability period and throughput with minimum energy consumption.

Underwater wireless sensor networks (UWSNs) arouse as a better alternative of underwater wired instruments for data gathering. Acoustic signals offer low bandwidth and UWSNs faces low reliability, high delay and high energy consumption issues. Moreover, energy holes creation decreases network performance in terms of energy and throughput. The design of routing protocols which considers these challenges can improve data gathering. In this paper, we propose forward layered multipath power control-one (FLMPC-One) and FLMPC-Two routing protocols to reduce energy utilization, achieve reliability and elude energy holes. Both FLMPC-One and FLMPC-Two are multicast routing protocols. In order to achieve reliability, both schemes direct multiple copies towards surface through different paths which posses low noises by establishing binary tree. Mostly, current forwarder takes decision of next forwarder selection and gets deceived by energy holes. Therefore, FLMPC-One and FLMPC-Two makes decision by including two and three hops neighbors, respectively to detect and elude energy hole. In this way, they conserve energy and reduce delay introduced by retransmission.

In this paper, we propose an algorithm to alleviate coverage hole problem using geographic routing strategy for wireless sensor networks (WSNs). In order to accomplish desired results, an optimal number of forwarder nodes is computed along with the selection of path that has minimum energy consumption. Moreover, at each hop residual energy of a sensor is calculated and knowledge up-to one hop neighbors of forwarder node that ensures the avoidance of energy hole problem. Simulations are conducted to validate that our claim of outperforming compared existing schemes in terms of packet delivery ratio (PDR) and energy dissipation of the network nodes.