A Multi-Agent Reinforcement Learning-Based Optimized Routing for QoS in IoT

The Routing Protocol for Low power and lossy networks (RPL) is used as a routing protocol in IoT applications. In an endeavor to bring out an optimized approach for providing Quality of Service (QoS) routing for heavy volume IoT data transmissions this paper proposes a machine learning-based routing algorithm with a multi-agent environment. The overall routing process is divided into two phases: route discovery phase and route maintenance phase. The route discovery or path finding phase is performed using rank calculation and Q-routing. Q-routing is performed with Q-Learning reinforcement machine learning approach, for selecting the next hop node. The proposed routing protocol first creates a Destination Oriented Directed Acyclic Graph (DODAG) using Q-Learning. The second phase is route maintenance. In this paper, we also propose an approach for route maintenance that considerably reduces control overheads as shown by the simulation and has shown less delay in routing convergence.

Decision Making in Real Estate: Portfolio Approach

An investment policy is suggested about assets on real estate markets. Such analysis recommends investments in non-financial assets and optimization of the results from such decisions. The formalization of the investment policy is based on the portfolio theory for asset allocation. Two main criteria are applied for the decision making: return and risk. The decision support is based on Mean-Variance portfolio model. A dynamical and adaptive investment policy is derived for active portfolio management. Sliding procedure in time with definition and solution of a set of portfolio problems is applied. The decision defines the relative value of the investment to which real estates are to be allocated. The regional real estate markets of six Bulgarian towns, which identify the regions with potential for investments, are compared. The added value of the paper results in development of algorithm for a quantitative analysis of real estate markets, based on portfolio theory.

Linear Regression Trust Management System for IoT Systems

This paper aims at creating a new Trust Management System (TMS) for a system of nodes. Various systems already exist which only use a simple function to calculate the trust value of a node. In the age of artificial intelligence the need for learning ability in an Internet of Things (IoT) system arises. Malicious nodes are a recurring issue and there still has not been a fully effective way to detect them beforehand. In IoT systems, a malicious node is detected after a transaction has occurred with the node. To this end, this paper explores how Artificial Intelligence (AI), and specifically Linear Regression (LR), could be utilised to predict a malicious node in order to minimise the damage in the IoT ecosystem. Moreover, the paper compares Linear regression over other AI-based TMS, showing the efficiency and efficacy of the method to predict and identify a malicious node.

COVID-19 Confirmed Cases Prediction in China Based on Barnacles Mating Optimizer-Least Squares Support Vector Machines

The Covid19 has significantly changed the global landscape in every aspect including economy, social life, and many others. After almost two years of living with the pandemic, new challenges are faced by the research community. It may take some time before the world can be declared as totally safe from the virus. Therefore, prediction of Covid19 confirmed cases is vital for the sake of proper prevention and precaution steps. In this study, a hybrid Barnacles Mating Optimizer with Least Square Support Vector Machines (BMO-LSSVM) is proposed for prediction of Covid19 confirmed cases. The employed data are the Covid19 cases in China which are defined in daily periodicity. The BMO was utilized to obtain optimal values of LSSVM hyper-parameters. Later, with the optimized values of the hyper-parameters, the prediction task will be executed by LSSVM. Through the experiments, the study recommends the superiority of BMO-LSSVM over the other identified hybrid algorithms.

A Three-Tier Authentication Scheme for Kerberized Hadoop Environment

Apache Hadoop answers the quest of handling Bigdata for most organizations. It offers distributed storage and data analysis via Hadoop Distributed File System (HDFS) and Map-Reduce frameworks. Hadoop depends on third-party security providers like Kerberos for its security requirements. Kerberos by itself comes with many security loopholes like Single point of Failure (SoF), Dictionary Attacks, Time Synchronization and Insider Attacks. This paper suggests a solution that aims to eradicate the security issues in the Hadoop Cluster with a focus on Dictionary Attacks and Single Point of Failure. The scheme roots on Secure Remote Password Protocol, Blockchain Technology and Threshold Cryptography. Practical Byzantine Fault Tolerance mechanism (PBFT) is deployed at the blockchain as the consensus mechanism. The proposed scheme outperforms many of the existing schemes in terms of computational overhead and storage requirements without compromising the security level offered by the system. Riverbed Modeller (AE) Simulation results strengthen the aforesaid claims.

An Efficient Method to Enhance IP Telephony Performance in IPV6 Networks

IP telephony have played an essential role during the COVID 19 pandemic lockdown. One of the issues that lower the service level of the IP telephony solutions is the inefficient bandwidth exploitation. This paper proposes a Smallerize/Zeroize (SmlZr) method to enhance bandwidth exploitation. The SmlZr method is explicitly designed for the P2P IP telephony calls over IPv6 networks. The essence concept of the proposed method is to use the unnecessary fields in the header to keep the voice media of the packet. Doing so leads to smallerize or zeroize the packet payload and, thus, enhance the bandwidth exploitation. The SmlZr method has outperformed the RTP method for all the comparison parameters. For instance, the SmlZr method shrinks the bandwidth by 25% compared to the RTP protocol. Bandwidth saving is helpful for P2P IP telephony calls because it alleviates the traffic load. Thus, improve the call capacity boosts the call clarity.

Access Control Models

Access control is a part of the security of information technologies. Access control regulates the access requests to system resources. The access control logic is formalized in models. Many access control models exist. They vary in their design, components, policies and areas of application. With the developing of information technologies, more complex access control models have been created. This paper is concerned with overview and analysis for a number of access control models. First, an overview of access control models is presented. Second, they are analyzed and compared by a number of parameters: storing the identity of the user, delegation of trust, fine-grained policies, flexibility, object-versioning, scalability, using time in policies, structure, trustworthiness, workflow control, areas of application etc. Some of these parameters describe the access control models, while other parameters are important characteristics and components of these models. The results of the comparative analysis are presented in tables. Prospects of development of new models are specified.

Evaluation of Computational Approaches of Short Weierstrass Elliptic Curves for Cryptography

The survey presents the evolution of Short Weierstrass elliptic curves after their introduction in cryptography. Subsequently, this evolution resulted in the establishment of present elliptic curve computational standards. We discuss the chronology of attacks on Elliptic Curve Discrete Logarithm Problem (ECDLP) and investigate their countermeasures to highlight the evolved selection criteria of cryptographically safe elliptic curves. Further, two popular deterministic and random approaches for selection of Short Weierstrass elliptic curve for cryptography are evaluated from computational, security and trust perspectives and a trend in existent computational standards is demonstrated. Finally, standard and non-standard elliptic curves are analysed to add a new insight into their usability. There is no such survey conducted in past to the best of our knowledge.

A Fuzzy Approach to Multi-Objective Solid Transportation Problem with Mixed Constraints Using Hyperbolic Membership Function

Multi-objective Solid Transportation Problem (MSTP) is known as a special class of vector-minimization (or maximization) problems and has three parameters: source, destination, and conveyance. The objectives such as transportation cost, transportation time, transportation safety level, and objectives in terms of environmental and social issues are generally in conflict with each other. In this paper, we present a fuzzy approach to bring these conflicting objectives together as high as possible. Instead of using the linear membership function, which is frequently used in the literature for ease of use, we use the hyperbolic membership function in our approach. Also, while most of the papers in the literature deal with the standard equality constrained form of MSTP, the mixed constrained form is addressed in this paper. Finally, a numerical example from the literature is used to illustrate the construction of the hyperbolic membership function and how well it represents the objective functions’ degree of satisfaction.

Neural Networks in Engineering Design: Robust Practical Stability Analysis

In recent years, we are witnessing artificial intelligence being deployed on embedded platforms in our everyday life, including engineering design practice problems starting from early stage design ideas to the final decision. One of the most challenging problems is related to the design and implementation of neural networks in engineering design tasks. The successful design and practical applications of neural network models depend on their qualitative properties. Elaborating efficient stability is known to be of a high importance. Also, different stability notions are applied for differently behaving models. In addition, uncertainties are ubiquitous in neural network systems, and may result in performance degradation, hazards or system damage. Driven by practical needs and theoretical challenges, the rigorous handling of uncertainties in the neural network design stage is an essential research topic. In this research, the concept of robust practical stability is introduced for generalized discrete neural network models under uncertainties applied in engineering design. A robust practical stability analysis is offered using the Lyapunov function method. Since practical stability concept is more appropriate for engineering applications, the obtained results can be of a practical significance to numerous engineering design problems of diverse interest.