Search

Solar power systems have been growing globally to replace fossil fuel-based energy and reduce greenhouse gases (GHG). In addition to panel efficiency deterioration and contamination, the produced power of photovoltaic (PV) systems is intermittent due to the dependency on weather conditions, causing reliability and resiliency issues. Monitoring system parameters can help in predicting faults in time for corrective action to be taken or preventive maintenance to be applied. However, classical monitoring approaches have two main problems: neither local nor centralized monitoring support distributed PV power systems nor provide remote access capability. Therefore, this paper presents an appraisal of a remote monitoring system of PV power generation stations by utilizing the Internet of Things (IoT) and a state-of-the-art tool for virtual supervision. The proposed system allows real-time measurements of all PV system parameters, including surrounding weather conditions, which are then available at the remote control center to check and track the PV power system. The proposed technique is composed of a set of cost-effective devices and algorithms, including a PV power conditioning unit (PCU); a sensor board for measuring the variables that influence PV energy production such as irradiance and temperature, using a communication module based on Wi-Fi for data transmission; and a maximum power point tracking (MPPT) controller for enhancing the efficiency of the PV system. For validating the proposed system, different common scenarios of PV panel conditions including different shading circumstances were considered. The results show that accurate, real-time monitoring with remote access capabilities can provide timely information for predicting and diagnosing the system condition to ensure continued stable power generation and management.

In the last few years, the deep learning (DL) computing paradigm has been deemed the Gold Standard in the machine learning (ML) community. Moreover, it has gradually become the most widely used computational approach in the field of ML, thus achieving outstanding results on several complex cognitive tasks, matching or even beating those provided by human performance. One of the benefits of DL is the ability to learn massive amounts of data. The DL field has grown fast in the last few years and it has been extensively used to successfully address a wide range of traditional applications. More importantly, DL has outperformed well-known ML techniques in many domains, e.g., cybersecurity, natural language processing, bioinformatics, robotics and control, and medical information processing, among many others. Despite it has been contributed several works reviewing the State-of-the-Art on DL, all of them only tackled one aspect of the DL, which leads to an overall lack of knowledge about it. Therefore, in this contribution, we propose using a more holistic approach in order to provide a more suitable starting point from which to develop a full understanding of DL. Specifically, this review attempts to provide a more comprehensive survey of the most important aspects of DL and including those enhancements recently added to the field. In particular, this paper outlines the importance of DL, presents the types of DL techniques and networks. It then presents convolutional neural networks (CNNs) which the most utilized DL network type and describes the development of CNNs architectures together with their main features, e.g., starting with the AlexNet network and closing with the High-Resolution network (HR.Net). Finally, we further present the challenges and suggested solutions to help researchers understand the existing research gaps. It is followed by a list of the major DL applications. Computational tools including FPGA, GPU, and CPU are summarized along with a description of their influence on DL. The paper ends with the evolution matrix, benchmark datasets, and summary and conclusion.

Recent research has introduced the notion of an Intelligent Reflective Surface (IRS) so as to boost the source–destination communication environment. IRS refers to a plane that consists of a number of passive elements that have a smaller size than the wavelength of the incident signal. Those elements use reflection to beam-form and to forward the incident radio signals toward the intended destination without power consumption. This article examines the performance of an IRS-based non-orthogonal multiple access (NOMA) system, and adopts the idea of using multiple IRS planes. It additionally investigates improving the performance of an IRS-NOMA combination through proposing various solutions that include optimizing the overall number of the IRS elements as a first step. After that, a closed form expression is derived for the considered IRS-NOMA system to determine the optimal number of the reflective elements. Secondly, the paper utilizes a new approach termed as multiple-IRS-based NOMA. This encompasses using multiple reflective planes (multi-IRS) along with IRS-NOMA, to boost the received signal characteristics, especially in situations where it is there a weak or no direct link between the source and the destination. An energy efficiency-spectral efficiency (EE-SE) trade-off is likewise presented, so as to have a complete view of the IRS-NOMA performance, along with the proposed solutions. The obtained simulation results depicts that IRS-NOMA is better than the existing orthogonal techniques. In addition, the results confirm that exploiting multiple IRSs offers a considerable gain in performance, as it enhances the conditions of the propagation environment. In specific, the considered case of a two IRS surfaces-assisted NOMA system offers higher performance levels than the case of a single IRS surface-assisted NOMA system.

In multi-hop wireless communication, a sensor node should consume its energy efficiently for relaying of data packets. However, most IoT-devices are equipped with limited battery power and computing resources for wireless communications, and thus energy optimization becomes one of the major concerns in wireless sensors routing design. The wireless technologies usually use unlicensed frequency bands of 2.4 GHz to transmit the data. Due to the broadcasting medium, the wireless transmission interferes with the reception of surrounding radios. As a result, data transmission failure increases resulting in low-communication quality. Therefore, one of the best solutions to this problem is to select the hop distance node that has a few neighbor nodes to disseminate packets until it reaches the ultimate receiver. The proposed routing selects the node that has few neighboring nodes and thus less interference. In another word, the scheme finds a better load balancing, and thus minimizes the probability of overload on a sensor node. It also introduces a new clustering algorithm around a single base station to shorten the transmission distances. This approach periodically selects the cluster heads (CHs) according to its location based distance from the final destination. The extensive simulation studies reveal that the proposed algorithm finds the best routing node and clustering formation to forward the traffic and thereby minimizes the interference ratio. In addition, the proposed protocol achieves low-energy consumption and longer network lifetime than other popular protocols.

The past decade has shown considerable growth in the field of deep learning techniques, changing the context of several areas of research. In medicine, deep learning techniques have achieved encouraging results with additional precision in the processing of various image datasets, such as brain MRI, chest X-ray, and retinal imaging. For example, many human organs can be scanned quickly and at a lower cost using X-ray machines, which are widely available in hospitals and clinics. It is common practice for expert radiologists to interpret various radiographic images manually. Training a deep learning network with these images provides medical staff with valuable help in diagnosing COVID-19 patients. Such a scenario helps the developing countries, due to the availability of X-ray machines, but there is a large shortage in the availability of the experts. This study aims to compare the effectiveness of four widely used deep learning models in terms of accuracy and processing time to determine the best. These models include ResNet-50, AlexNet, GoogLeNet, and VGG16. The findings indicate that the processing time is proportional to the accuracy. ResNet-50 had the highest diagnostic accuracy but was the slowest in processing time. Both CPU (serial) and GPU (parallel) platforms are used in this comparative study. When the models used a CPU platform, the processing time was in the range of 1–1.5 s, but it dramatically decreased to the range of 7.1–20.7 ms when running on a parallel platform (GPU). Hence, the GPU platform is very successful and was deemed best suited for real-time diagnostic applications.

Breast cancer is a significant factor in female mortality. An early cancer diagnosis leads to a reduction in the breast cancer death rate. With the help of a computer-aided diagnosis system, the efficiency increased, and the cost was reduced for the cancer diagnosis. Traditional breast cancer classification techniques are based on handcrafted features techniques, and their performance relies upon the chosen features. They also are very sensitive to different sizes and complex shapes. However, histopathological breast cancer images are very complex in shape. Currently, deep learning models have become an alternative solution for diagnosis, and have overcome the drawbacks of classical classification techniques. Although deep learning has performed well in various tasks of computer vision and pattern recognition, it still has some challenges. One of the main challenges is the lack of training data. To address this challenge and optimize the performance, we have utilized a transfer learning technique which is where the deep learning models train on a task, and then fine-tune the models for another task. We have employed transfer learning in two ways: Training our proposed model first on the same domain dataset, then on the target dataset, and training our model on a different domain dataset, then on the target dataset. We have empirically proven that the same domain transfer learning optimized the performance. Our hybrid model of parallel convolutional layers and residual links is utilized to classify hematoxylin&ndash, eosin-stained breast biopsy images into four classes: invasive carcinoma, in-situ carcinoma, benign tumor and normal tissue. To reduce the effect of overfitting, we have augmented the images with different image processing techniques. The proposed model achieved state-of-the-art performance, and it outperformed the latest methods by achieving a patch-wise classification accuracy of 90.5%, and an image-wise classification accuracy of 97.4% on the validation set. Moreover, we have achieved an image-wise classification accuracy of 96.1% on the test set of the microscopy ICIAR-2018 dataset.

The mission-critical internet of things (IoT) has recently evolved to cover many applications such as e-health, power distribution systems, industrial and smart homes and cities. Such applications are anticipated to work every time without any fail which requires a constant data transmission. Hybrid communication systems are often used to provide reliable transmission, improve the performance of communication systems and provide better mobility to the end users. Contrary to the previous works on this area which discuss the implementation of relays between different communication links, the authors of this paper study the performance of cooperative visible-light communication (VLC) and radiofrequency (RF) communication systems with the presence of AF relay. The performance of the system under consideration is analyzed in terms of ergodic capacity. A mathematical method is exploited for this hybrid system to formulate the overall capacity of the proposed system by using the characteristics of the VLC and RF channels. The analytical results of the derived expression are verified by Monte Carlo simulations throughout this letter. The authors also investigate the effect of the several system parameters on its performance. The results revealed that the length of each link has a negative impact on the system performance. It was also found that the capacity of the integrated VLC/RF system is significantly improved by increasing the AF relay gain.

In this paper, multi-classifier of K-Nearest Neighbor and Support Vector Machine (SVM) classifiers with multi-domain features are employed, as a proposed methodology for recognizing the normality status of the heart sound recordings (so-called Phonocardiogram - PCG). The PhysioNet/CinC Challenge 2016 offers the dataset used in this paper. Heart sounds are complex signals and required trained clinicians for diagnosis, which motivated us to develop an algorithm for automatic classification of heart sounds into two classes normal and abnormal. Entropy, high-order statistics, Cyclo-stationarity, cepstrum, the frequency spectrum of records, energy, state amplitude, the frequency spectrum of states, and time interval, are the nine-domain features employed. These domain features are extracted to a total of 527 features. These features have been used to train the KNearest Neighbor and Support Vector Machine (SVM) classifiers. Fine-KNN classifier outperformed types of SVM classifiers by achieving the accuracy of 93.5% while Cubic-SVM classifier achieved 90.9% which is the highest accuracy of all SVMs. The Fine-KNN classifier and the proposed features are both efficient and significant for PCG recognition.

Sickle cell anemia is an abnormal red blood cell which leads to blood vessel obstruction joined by painful episodes and even death. It is also called abnormal hemoglobin. Hemoglobin is responsible for passing oxygen through the blood vessel for all over the body. Normal red blood cells are in a circular shape and they are compact and flexible, enabling them to move freely through small capillaries. On the other hand, abnormal red blood cells are in sickle shape and they are stiff and angular causing them to become stuck in small capillaries. Due to that, it will be a reason for pain to patients and lead to low oxygen and dehydration. The manual assessment, classification, and counting of biological cells require for an immense spending of time and it may lead to wrong classification and counting since red blood cells are millions in one smear. Also, cells classification is challenging due to heterogeneous and complex shapes, overlapped cells and a variety of colors. We overcome these drawbacks by introducing a new robust and effective deep Convolutional Neural Network to classify Red Blood Cells (RBCs) in three classes namely: normal (‘N’) abnormal (sickle cells anemia type (‘S’)) and miscellaneous (‘M’). In order to improve the results further, we have used our model as features extractor then we applied an error-correcting output codes (ECOC) classifier for the classification task. Our model with ECOC showed outstanding performance and high accuracy of 92.06%.

Internet of Things (IoT) is becoming more and more pervasive in everyday life and connecting an array of diverse physical objects. It is fast growing and receiving a tremendous amount of research focus. Billions of objects communicate each other with or without human intervention to achieve smart applications. Most of the connected devices are constrained nodes to its ecosystem which have limited memories, CPU capabilities and power sources. Therefore, for implementing autonomous smart systems, efficient energy consumption is imperative. This chapter introduces a novel scheduling algorithm called Long Hop (LH) first to optimize energy usage on a wireless sensor network (WSN) that enables IoT systems. The selected algorithm proposes an optimized solution to the energy efficient for scalable IoT networks. LH technique assigns high priority for packets coming with more hops and longer distances to be served first at the cluster head (CH) nodes. Since these packets require more links and nodes (thus increased energy and bandwidth usage) to reach the ultimate destination if not prioritized. The proposed technique reduces the overall energy usage and minimizes the total number of packets re-transmission and the effective data transmission distances. Thus, this improves the overall system performance and elongates the network lifetime.

Recently, Internet of Things (IoT) have witnessed significant attention due to their potential of bringing massive number of interconnected heterogeneous sensor nodes to collect big data for harnessing knowledge thus enabling real time monitoring and control. These wireless sensors relay their packets on to the base station via multi-hop transmissions and is therefore dependent on the intermediate nodes. However, unbalanced data forward among intermediate nodes results in high-energy consumption and packet drops. Since, the network lifetime and link reliability are the most import factors for multi-hop environment this paper proposes a link quality-oriented route (LQOR) protocol for scalable IoT networks. The main goal of this scheme is to balance the load for any sensor by selecting the next hop such that it has less number of neighboring nodes. Specifically, an architecture for a typical IoT network is presented that uses LQOR and various metrics for the network, such as network lifetime, energy consumption, etc. are analyzed. The comparative performance evaluation corroborates the positive impact of the proposed algorithm on the entire path as compared to other techniques in the literature. We show that the proposed algorithm achieves less packets loss and minimizes the number of data re-transmissions. Thus, improving the overall system performance and elongating the network lifetime.

Internet of things (IoT) is considered to revolutionize the way internet works and bring together the concepts such as machine to machine (M2M) communication, big data, artificial intelligence, etc. to work under a same umbrella such that cyber space and human (physical systems) are more intertwined and thus ubiquitous giving rise to cyber physical systems. This will involve billions of connections and smart products communicating with each other mostly without human intervention to achieve smart objectives. The idea of IoT has enticed significant research attentions since the massive connectivity bring varieties of challenges and obstacles including heterogeneity, scalability, security, big data, energy requirements, etc. The paper looks into providing a concise review of the concepts on IoT and applications describing the main components and features. Furthermore, open issues and challenges that need addressing by the research community and some potential solutions are discussed.

Recently, energy efficiency in sensor enabled wire-less network domain has witnessed significant attention from both academia and industries. It is an enabling technological advancement towards green computing in Internet of Things (IoT) eventually supporting sensor generated big data processing for smart cities. Related literature on energy efficiency in sensor enabled wireless network environments focuses on one aspects either energy oriented path selection or energy oriented message scheduling. The definition of path also varies in literature without considering links towards energy efficiency. In this context, this paper proposes an energy oriented path selection and message scheduling framework for sensor enabled wireless network environments. The technical novelty focuses on effective cooperation between path selection and message scheduling considering links on path, location of message sender, and number of processor in sensor towards energy efficiency. Specifically, a path selection strategy is developed based on shortest path and less number of links on path (SPLL). The location of message sender, and number of processor in specific sensor are utilized for developing a longer hops (LH) message scheduling approach. A system model is presented based on M/M/1 queuing analysis to showcase the effective cooperation of SPLL and LH towards energy efficiency. Simulation oriented comparative performance evaluation attest the energy efficiency of the proposed framework as compared to the state-of-the-art techniques considering number of energy oriented metrics.

Due to the scarce energy supplies of the internet of things (IoT) devices, it is necessary to design the architecture and operation of the network to optimize the power usage. Scheduling algorithm is an essential part of WSNs and IoT networks. Such algorithms allow to classify the queues and decide which process to run. Therefore, scheduling message at cluster head nodes has been suggested for this study. Long hop (LH) first algorithm is new unified scheduling technique that schedules high priority for data comes from far distances and accesses higher number of devices to be routed first to the target. The performance experience shows the proposed method compares with two existing studies, a popular scheduling algorithm is called first-come-first-serve (FCFS) and nearest job next (NJN) algorithms. The results and simulation have been shown that the proposed study has consumed less power and maximize throughput. It saves up to 56% power consumption and 62.50% in network throughput. In addition, it minimizes the packets delay and loss, transmission distance and thus extend the lifetime of the network.

Internet of Things (IoT) is becoming more and more pervasive in everyday life and connecting an array of diverse physical objects. It is fast growing and receiving a tremendous amount of research focus. Billions of objects communicate each other with or without human intervention to achieve smart applications. Most of the connected devices are constrained nodes to its ecosystem which have limited memories, CPU capabilities and power sources. Therefore, for implementing autonomous smart systems, efficient energy consumption is imperative. This paper introduces a novel scheduling algorithm called Long Hop (LH) first to optimize energy usage on a Wireless Sensor Network (WSN) that enables IoT system. LH algorithm schedules high priority for packets coming with more hops and longer distances to be served first at the cluster head (CH) nodes of the WSN. Since these packets require more links and nodes (thus increased energy and bandwidth usage) to reach the ultimate destination if not prioritized, the proposed algorithm reduces the overall energy usage and minimizes the total number of packets re-transmission and the effective data transmission distances. This improves the overall system performance and elongates the network lifetime.

Internet of Things (IoT) is a rapidly growing technology with a wide range of applications in various fields. It has unified a plethora of devices and infrastructure under the same umbrella and is considered by many technologies leaders as the network of the future (NoF). IoT connects heterogeneous devices that provide sensing, control, actuation, and monitoring activities for smarter environments. Smart IoT devices or objects are characterized with a unique identifier to transfer data over the network without human intervention. IoT is expected to further extend the boundaries of the autonomous world with advanced connectivity of physical entities, systems, services. However, there are some serious obstacles and challenges, which must be resolved before the full potential of IoT is realized. The focus of this study is to discuss the key drivers of change, challenges that may slow the adoption of IoT and the potential of the internet of things technology.

The present study was conducted to find out the water requirements and most suitable irrigation frequencies for cowpea (Vigna unguiculata L.) var grown under drip irrigation. The treatments were based on the IW:CPE ratio at different empirical pan factors 0.6 , 0.8, 1.0, 1.,1.4 , and 1.6 Ef (where Ef = IW/CPE). It was observed that the irrigation interval was variable values decreased by increasing Ef value and with the progress of the growing season. The 1.2 and 1.0 IW: CPE treatments with approximately 4 days irrigation interval were achieved the best results. The total amount of applied water during Cowpea growing season was varied between 247.7 and 266.5mm with 254.8mm as a mean. Irrigation treatment with Ef1.2 was superior over the rest of other treatments in fresh seed yield (5.13 ton.hec.-1), crop water productivity (2.14 kg.m-3), biological yield (6.88 ton.hec.-1) , fresh pod yield (7.33 ton.hec.-1), weight of 100 seed (31.28gm), number of seed/pod (9.34) and netting percentage (37.1). The lowest values of the most parameters used in this study were obtained by Ef 0.6 irrigation treatment.