The design complexities of trending UAVs necessitates Control Law formulation utiliz-ing intelligent techniques that are robust and modal-free. In this research, a controlarchitecture to maximize UAV glide range for an experimental UAV is presented. Tohandle control complexities, while keeping them computationally acceptable, a distinctRL technique namely Modied Model Free Dynamic Programming (MMDP) is proposed.The methodology has been specically modied to congure the problem in continuousstate and control space domains. The ecacy of the results and performance charac-teristics, demonstrated the ability of the presented algorithm to dynamically adapt tothe changing environment, thereby making it suitable for UAV applications. Nonlinearsimulations performed under dierent environmental conditions demonstrated the eec-tiveness of the proposed methodology over the conventional classical approaches.

A novel coronavirus (2019-nCoV) or SARS-CoV-2 mainly causes an infection in the human respiratory system, first sited at Wuhan City, China in December 2019. The World Health Organization (WHO) announced it as a pandemic on March 12th, 2020. The origin and the transmission modes are not clear yet but assumed to be transmitted from wild animals like bats, pangolins, snakes or birds, etc. Studies revealed that SARS-CoV-2 and bat coronavirus has about 96.2% genome similarities. It is a single-strand RNA virus with a diameter of 80-120nm. The quinine-based drugs (chloroquine and hydroxychloroquine) have possible inhibition effect against most coronaviruses e.g. SARS-CoV-1. Thus, these drugs have been used for COVID-19 treatment. Herein, the action of these drugs on COVID-19 in the human body, and the age, dose, and the pH-dependent mechanism with the SARS-CoV-2 duplication cycle have been discussed.

The paper contains an implemented approach to stress recognition based on registered sound parameters LZE, LZeq, LZF and LZS at different speech levels in working environment. Said approach combines apparatus k – nearest neighbors (k-NN) with metric distances Euclidean, Cityblock, Minkowski and Chebychev, Decision tree (DT) method with CART training algorithm and artificial neural networks (ANN) with reverted error propagation in Levenberg-Marquadt (LM) and Scaled Conjugate Gradient (SCG) training. A matching has been done between parameters of models according k-NN method as maximum accuracy of 93.99 % with minimum parameter k = 3 has been registered for Cityblock distance. There has been established a fourth optimum level of nodes pruning from the structure for multiple choice of the classification group using DT method with achieved accuracy of 99.05 %. In investigation of LM algorithm during training of networks using purelin, tansig and logsig transfer functions, there has been achieved identical accuracy of 99.99 %. With regard to a minimal indication of criterion Mean Squared Error (MSE) 0.0064 with 11 hidden neurons, there has been selected ANN architecture with tansig output activation. On the grounds of the presented approach there has been synthesized a model for correct recognition, 100.00 % of speech obtained with the help of artificial intelligence (AI) during SCG training.

Respiratory infections are one of the major reasons for mortality and morbidity throughout the whole world. Every year millions of people suffer and die from this infection. The current study involves with hundreds of patients comprising both males and females who underwent medical treatment in Cumilla city. The relevance of the infection with the age, sex and timing of the year were studied side by side with the determination of the responsible bacterial pathogen and their antibiotic sensitivity patterns. The study found six distinct bacteria causing respiratory infections (Pseudomonas spp., Klebsiella spp., Acinetobacter spp., Enterobacter spp. Escherichia coli, Serratia spp.) among which Klebsiella spp. and Pseudomonas spp. were the most prevalent. Hotter temperatures also showed most of the infections than colder months according to the study and the distinct pathogens. While studying age group, people from 41-60 years range were mostly infected with respiratory disease regardless of the gender. Out of 100 samples, 21 isolates were susceptible to all of the fourteen antibiotics used in the study and 4 isolates were completely resistant towards all of the antibiotics. Amongst other isolates, most of them showed multidrug resistance. Proper use of antibiotics and development of new drug is extremely important for saving the lives of millions of people from infection with multidrug resistant bacteria.

This research presents a scalable and novel property/land registration framework. It is based on blockchain technology where all registry offices willing to participate in this system must join the proposed blockchain based framework in order to improve the registration process. Transparent and immutable nature of blockchain technology makes it reliable for the intrusive and fraudulent activities in land registration. The framework uses the Inter Planetary file system (IPFS) as a peer-to-peer (P2P) swarm network to seamlessly integrate different registry offices of a state in the country. It allows the sharing and storing of data in an efficient, decentralized, and transparent manner. This research proposes an efficient consensus mechanism based on the trust value that minimizes the computational overhead of broadcasting the new block as compared to existing proof of work approach. Hence, this approach requires less time to add the block into the blockchain, as well as message exchange per consensus is much lesser instead of Proof of Work (PoW) approach.

This study deals with the identification of source function from final time state observation in a two-dimensional hyperbolic problem. The solution to direct problem is obtained by weak solution approach and finite element method. In the part of inverse problem, the trust region method and Levenberg-Marquardt method which are nonlinear least squares optimization methods are used for identification of source function. The findings are presented with numerical examples.