In recent decades, deep-learning dependent fruit disease detection and classification techniques have evinced outstanding results in technologically advanced horticulture investigation. Due to the comparatively limited image processing capabilities of edge computing devices, implementing deep learning methods in actual field scenarios is currently difficult. The use of intelligent machines in contemporary horticulture is being hampered by these restrictions, which are emerging as a new barrier. In this research, we present an efficient model for citrus fruit disease prediction. The proposed model utilizes the fusion of deep learning models CNN and LSTM with edge computing. The proposed model employs an enhanced feature-extraction mechanism, with a down-sampling approach, and then a feature-fusion subsystem to ensure significant recognition on edge computing devices with retaining citrus fruit disease detection accuracy. This research utilizes the online Kaggle and plan village dataset which contains 2950 citrus fruit images with disease categories black spots, cankers, scabs, Melanosis, and greening. The proposed model and existing model are tested with two features with pruning and without pruning and compared based on various performance measuring parameters, i.e., precision, recall, f-measure, and support. In the first phase experimental analysis is performed using Magnitude Based Pruning and in the second phase Magnitude Based Pruning with Post Quantization. The proposed CNN-LSTM model achieves an accuracy rate of 97.18% with Magnitude-Based Pruning and 98.25% with Magnitude-Based Pruning with Post Quantization, which is better as compared to the existing CNN method.

The COVID-19 Pandemic affected all sectors worldwide including education sector changing the students learning environment at a large scale. In the education sector, governments had to close schools and universities around the world and turn into online mode of delivery. The sudden transformation to online learning demands educators and institutions to develop more innovative teaching methods in supporting students during this growing crisis. It is pertinent that this is the era of large-scale online learning, where education is delivered remotely utilizing various digital platforms. To this end, this move presented many challenges to students, educators, families, administrators, and government regulations. In this paper, we conduct an oriented and well-rounded survey that captures all aspects of distance learning verticals as perceived by educators in order to study online delivery aspects, including teaching models, communication media, and teaching best practices in distance learning. The paper will answer subtle and comprehensive questions related to student satisfaction and success, technical needs and challenges, teaching models, educator mobility, etc. across different universities in the UAE. In particular, the paper tries to identify best model to use given many key issues related to the adoption of e-learning during COVID-19. Overall, the survey results showed almost all participants agreed that online learning during the COVID-19 is a very good alternative solution to Face-to-Face approach. However, teaching models do vary in promoting students’ participation and students’ engagement. All participants agree that absence of the proper infrastructure at home will hinder the operation especially in courses with hands-on components. Finally, the educators questioned the credibility of online assessments despite the enforced exam integrity tools and methods. The paper concluded with many recommendations for continuous development of the distance learning systems during and after COVID -19.

In Body Area Networks (BANs), quality of service is needed to provide reliable data communication over prioritized data streams coming from energy constrained sensors attached to, or possibly implanted in, patients. In this work, we focus on BAN for real-time data streaming applications, where the real-time nature of data streams is of critical importance for providing a useful and efficient sensorial feedback for the user while system lifetime should be maximized. Thus, bandwidth, throughput and energy efficiency of the communication protocol must be carefully optimized. In this paper, we present an Energy-Balanced Rate Assignment and Routing Protocol (EBRAR). EBRAR selects routes based on the residual energy, thus, instead of continuously routing data through an optimized (energy efficient) fixed path, the data is transported more intelligently and the burden of forwarding the data is more equally spread among the nodes. Another unique property of EBRAR is its ability to provide adaptive resource allocation. Our experimental results show that the proposed protocol performs well in terms of balancing energy consumption across the BAN and thus guarantees longer network lifetime.

A key design issue in wireless networks is represented by the irregular and dynamic radio coverage at each node. This is especially true for wireless sensor networks, which usually employ low quality radio modules to reduce the cost. It results in irregularity in radio coverage and variations in packet reception in different directions. Due to its likely impact on the upper layer protocols, many services, such as localization, routing and others, needs to be resilient to the irregular and dynamic radio propagation, and to include mechanisms to deal with these problems. As such, accurate models of radio propagation patterns are important for protocol design and evaluation. In this paper, measurements of radio propagation patterns have been carried out using the motes themselves. With empirical data obtained from the Mica2 platforms we were able to observe and further quantify such phenomena. The results demonstrate that the radio pattern is largely random; however, radio signal attenuation varies along different direction, and more importantly, is time-varying while stationary.

Sensor network which operates on battery are used to gather data in a variety of environments. The data collected by each node is communicated through the network to the sink, which uses all reported data to determine characteristics of the environment or detect an event. Prolonging sensor's operable lifetime is a main design challenge of these networks. A good energy saving technique in this direction is to schedule nodes sleep interval with the communication radio turned off. In this paper, we propose a distributed topology control algorithm, termed ECTC, which uses a clustering approach. It is built on the notion that when a region of a shared channel wireless sensor network has a sufficient density of nodes, significant energy saving is obtained by allowing redundant nodes to sleep. Using the two-hop neighborhood information, certain nodes sequentially select a subset of nodes to be active among all nodes in the neighborhood, to ensure connectivity. Moreover, to ensure fairness, the role of active nodes is rotated periodically to ensure energy-balanced operations. Results from stochastic geometry are used to derive solutions for the values of parameters of our algorithm that minimize the total energy spent in the network when all sensor nodes report data through the cluster heads to the sink.