International Journal of Sensors Wireless Communications and Control - Volume 13, Issue 3, 2023

Background: The World Health Organization (WHO) reported that Air pollution (AP) is prone to the highest environmental risk and has caused numerous deaths. Polluted air has many constituents where Particulate Matter (PM) is majorly reported as a global concern. Currently, the most crucial challenges faced by the globe are the identification and treatment of augmenting AP. The air pollution level was indicated by the Air Quality Index (AQI). It is affected by the concentrations of several pollutants in the air. Many pollutants in the air are harmful to human health. Thus, an efficient prediction system is required. Many security problems and lower classification accuracy are faced by them even though several prediction systems have been formed. A secure air quality prediction system (AQPS) centered upon the energy efficiency of smart sensing is proposed in this paper to overcome these issues. From disparate sensor nodes, the input data is initially amassed in the proposed work. The gathered data is stored in the temporary server. Next, the air-polluted data of the temporary server is offered to the AQPS, wherein preprocessing of the input data along with classification is executed. Methods: Utilizing the Improved Spotted Hyena Optimization-based Deep Convolution Neural Network (ISHO-DCNN) algorithm, the classification is executed. Utilizing the Repetitive Data Coding Based Huffman Encoding (RDC-HE) method, the polluted data attained from the classified output is compressed and encrypted by employing the American Standard Code for Information Interchange based Elliptical Curve Cryptography (ASCII-ECC) method. Results: Afterward, the encrypted and compressed data is saved in the Cloud Server (CS). Finally, for notifying about the AP, the decrypted and decompressed data is offered to the Base Stations (BS). Conclusion: The proposed work is more effective when analogized to the prevailing methods as denoted by the experimental outcomes. Higher accuracy of 97.14% and precision of 91.44% were obtained by the proposed model. Further, lower Encryption Time (ET) and Decryption Time (DT) of 0.565584 sec and 0.005137 sec were obtained by the model.

Background: Currently, Artificial Intelligence (AI) and the Internet of Things (IoT) have transformed the field of agriculture with the innovative idea of automation and intelligence. The agriculture field completely relies on the uncertainty parameter of soil, atmosphere, and water. Technological advancement in IoT and AI assist in resolving this uncertainty factor and recommend the best crops to the farmers so that they can also enhance the productivity of the crops and meet the world's large food demand smartly. Objective: In this paper, we have suggested an IoT and AI-based model which trained with 2200 records of the dataset and seven attributes in Python. The model suggests 22 different crops to farmers after collecting samples through different sensor data. We used soil, temperature, humidity, pH, and rainfall sensors. Soil sensors were used to measure the amount of N, P, and K in soil. Methods: Various supervised machine learning algorithms such as KNN, Decision Tree, Naïve Bayes and Logistic Regression classifiers have applied to build the proposed model. The model is continuously monitoring the field via various sensor data as a sample data for the prediction of best crops to be grown for farmers. Results: In this research, we investigated the contribution of supervised machine learning classifiers like KNN, Decision Tree, Naïve Bayes and Logistic Regression classifiers. The maximum accuracy has been observed as 99.39% of the Naïve Bayes classifier. Conclusion: In this paper an AI and IoT based model is used to recommend/predict the best crop based on environmental factors. The proposed model will collect the real time sensor data to predict the crops and plants smartly.

Aims and Background: Artificial intelligence (AI) is expanding in the market daily to assist humans in a variety of ways. However, as these models are expensive, there is still a gap in the availability of AI products to the common public with high component dependency. Objectives and Methodology: To address the issue of additional component dependency on AI products, we propose a model that can use available Smartphone resources to perceive real-world huddles and assist ordinary people with their daily needs. The proposed AI model is to predict the user’s indoor position (Node) at the computer science and engineering block of CMR Institute of Technology (CMRIT) by using Smartphone sensors and wireless signals. We used SVR to predict the regular walk steps needed between two Nodes and Pedestrian Dead Reckoning (PDR) to predict the walk steps needed while the signal was lost in the indoor environment. Results: The Support vector regression (SVR) models make the locations to be available within the specified building boundaries for proper guidance. The PDR approach supports the user while signal loss between two Received Signal Strength Indicators (RSSI). The Pedestrian dead reckoning - Support Vector Regression (PD-SVR) results are showing 98% accuracy in NODE predictions with routing tables. The indoor positioning is 100% accurate with dynamic crowd-sourcing Node preparation. Conclusion: The results are compared with other indoor navigation models K-nearest neighbor (KNN) and DF-SVM are given 95% accurate NODE estimation with minimal need for network components.

Aims and Backgrounds: The Internet of Things has evolved over the years to a greater extent, where objects communicate with each other over a network. Heterogenous communication between the nodes leads to a large amount of information sharing, and sensitive information could be shared over the network. It is important to maintain privacy and security during information sharing to protect devices from communicating with malicious nodes. Objectives and Methodology: The concept of trust was introduced to prevent nodes from communicating with malicious nodes. A trust computation model for the IoT based on machine learning concepts was designed, which evaluates trust based on the Trust Marks. There are three trust marks, out of which two are evaluated. The three trust marks are knowledge, experience, and reputation. Knowledge trust marks are evaluated separately based on their trust property mathematical formulations, and then based on these properties, machine learning-based algorithms are applied to train the model to classify the objects as trustworthy and untrustworthy. Results: The effectiveness of the Knowledge Trust Mark is measured by a simulation and confusion matrix. The accuracy of the trained model is shown by the accuracy score. The trust computational model for IoT using machine learning shows higher accuracy in classifying the objects as trustworthy and untrustworthy. Conclusion: The experience trust mark is evaluated based on its properties, and the behaviour of the experience is shown over time graphically.

Background: The compute-and-forward strategy is one of the outstanding methods which is used for interference management in wireless relay networks where decoding linear combinations of code words is required. Recently, many efforts have been made for decoding integer and noninteger combinations. The difference between the methods is the manner of handling different conditions of networks, such as equal or unequal power constraints and equal or unequal channel gains. Objectives: In this work, we present a modified n-step asymmetric successive compute-and-forward strategy for the communication network where we have both unequal power constraints and unequal channel gains conditions. Methods: In the proposed method, we scale channel gains and coefficients with the square root of power constraints. In this way, despite previous methods, without the need for scaling factors in our formulation, it is still able to solve the problem of general Gaussian relay networks with unequal power constraints and unequal channel gains. We also use scaling factors in our method in order to have the ability to divide the rates between users fairly. Results and Conclusion: We evaluate the ability of the modified strategy for the uplink communication of the two-way relay channel, where one relay can help communication between the two users. At the relay, we decode the linear combinations of the messages of the two users and obtain 1/2 bit/sec/Hz per user capacity gap from the cut-set bound. Through some theoretical and simulation results, we show that by appropriately adjusting parameters, different points and areas of rate regions are achievable.

Introduction: Heterogeneous networks (HetNet) represent a promising technology that satisfies the needs of mobile users. However, several problems have influenced the performance of wireless communication, such as the maximization of energy efficiency and the problem of interferences due to the uncontrolled association of the user equipment (UE). Methods: Solving the problem of maximizing energy efficiency has captured the attention of several researchers. In this work, we propose an effective user association based on K-nearest Neighbors (KNN) approach considering a large dataset. The major novelty of this work is that the supervised learning perspective is applied to a dataset regrouped from an optimal user association, where the most valuable parameters are considered. Result: Additionally, it allows for mitigating the problem of interferences using individual user association. Simulation results have proven the efficiency of the proposed methodology. Conclusion: The suggested results have outperformed the two works in terms of accuracy, where the proposed method presents a better accuracy of 95%.