The research domain of UWSN (Under Water Sensor Networks) is gaining wide attentionthese days. The sensors are deployed on the sea surface or seabed for monitoring, data collection orsurveillance purpose. Unlike terrestrial networks, the UWSNs use acoustic signals for data communica-tion. The speed and bandwidth of acoustic signals is low compared to radio signals which are used inconventional terrestrial WSNs (Wireless Sensor Networks). As a result, protocol design in UWSN mustbe addressed separately than terrestrial networks. In this work, we have reviewed TDMA (Time Divi-sion Multiple Access) protocols for channel access. As TDMA is a collision-free method, it is preferredover other methods like random access methods or handshaking based medium access protocols. Someopen issues are also proposed as follows: (i) It is found that very less or no work is done in the area ofjoint scheduling and topology formation for UWSNs. (ii) The problem of balancing schedule length ofsink-rooted trees in multiple sinks UWSNs is also unattempted.
The Covid-19 app based on Chest X-rays & CT scans is an integration of Machine learning and android. Our group developed an android application that uses an ML model and predicts if the image captured or uploaded by the user of a Chest X-rays or a CT scan has Covid19. The main tools used for the development of this project were android studio & Firebase for android app development, Jupyter notebook & google Colab as a code editor along with google drive to fetch the data. The main Libraries used for training ML models were Tensorflow, Keras and the method used was Transfer Learning with the help of InceptionV3 pre-trained model on the Imagenet dataset In coping and fighting against COVID-19, the most critical step is to effectively screen and diagnose infected patients. RT-PCR is considered to be a reliable test for detection of coronavirus, but the problem with RT-PCR test and Elisa tests is that they take a lot of time to generate results, Since Covid is highly contagious and spreads through Human-to-Human Interactions it’s Crucial to detect it as early as possible to stop the transmission. Corona is an Infectious disease that affects the Lungs similar to Pneumonia, Deep Learning and Machine Learning models have produced significant results in the past for pneumonia detection in Lungs in this project we tried the same approach where we trained the Deep learning model on more than 13000 images of CT scans & 9000 images for X-rays and used the model to build an android application for users, the application is user friendly and easy to use. Since radiology scanning takes far less time to generate results as compared to the RT-PCR & ELISA tests, we believe our app will be convenient and it'll help people in the early detection of Covid.
The approach adopted to handle hard problems is known as metaheuristics. The problem is considered as hard if discovering the optimal solution for it may not be always possible within the stipulated time. Discovering a single solution to a problem is easy and can be accomplished extremely fast, but finding the best possible solution to the same problem is very long. Optimization algorithms are intended to bridge this gap. The research paper aims at solving the problems for finding the optimal solution for two popular metaheuristic algorithms, GWO (Grey Wolf Optimization) and FF (Firefly) algorithms. Both the metaheuristics algorithms, GWO and FF algorithms are studied and implemented. The two technical features comprised of metaheuristic algorithms are exploration and exploitation. The optimal solution has been evaluated alongside Makespan and Utilization Rate for both GWO and FF algorithms. The lower value of the Makespan and higher Utilization Rate is always desirable. Both the algorithms have been modified via using mathematical functions to enhance the readings concerning performance evaluation parameters. The GWO is been modified via developing a hybrid version comprising GWO and PSO (Particle Swarm Optimization) algorithms denoted as the Hybrid Modified GWOPSO algorithm. The FF algorithm too has been modified and is denoted as a Modified FF algorithm. The conducted modifications have been measured via different performance evaluation parameters. Finally, the fault tolerance factor is considered and the modified versions Hybrid Modified GWOPSO and Modified FF are hybridized to develop a new hybrid algorithm Hybrid GWOFF (Hybrid Grey Wolf Firefly) algorithm and its performance have been evaluated with and without fault tolerance.
Agriculture and allied industries play an important role in the development of our nation. In India more than 55% of people make a living from farming. Crop yields are an essential aspect of every farmer’s day. It depends on many factors like soil quality, seeds, planting practices, humidity, fertilizers and pesticides. Besides all factors, diagnosing soil quality is a fundamental and essential task in farming, as it provides background knowledge of the soil and its physical, chemical and biological prominence. Hence, soil analytics is inevitable that gives information about the present nutrient availability or the need of the nutrients for effective cultivation. It helps to interpret the physico-chemical properties of soil nutrients and to classify the nutrient content as very low, low, medium, high, or very high based on pH values. Thus, predictive analytics based on the soil parameters offer precise and sensible solutions for soil fertility problems and enable suitable decision on crop cultivation. This study attempts to exploit the benchmark classification algorithms from data mining to classify soil samples of Tiruppur district using pH levels. The prediction of pH levels is important to know the nutrients availability in the soil. Classification algorithms like Logistic Regression (LR), Bernoulli Naive Bayes (BNB), Decision Tree (DT), Extra Tree (ET), Random Forest (RF) and K-Nearest Neighbor (KNN) are used to evaluate and predict the pH values. After the comprehensive evaluation, this study determined that the performance of the DT and RF model for pH prediction is high compared to the other algorithms in terms of accuracy. Further, the classifiers performance has improved by possessing feature scaling techniques like normalization and standardization. Results showed that the prediction accuracy of KNN and BNB with feature scaling outperforms the other algorithms.
