International Journal of Sensors Wireless Communications and Control - Volume 13, Issue 2, 2023

Background: QoS parameters are volatile in nature and possess high nonlinearity, thus making the IoT-based service and recommendation process challenging. Methods: An efficient and accurate forecasting model is lacking in this area and needs to be explored. Though an artificial neural network is a prominent option for capturing such nonlinearities, its efficiency is limited by the structural complexity and iterative learning method. The random vector functional link network (RVFLN) significantly reduces the time complexity by randomly assigning input weights and biases without further modification. Only output layer weights are calculated iteratively by gradient methods or non-iteratively by least square methods. It is an efficient algorithm with low time complexity and can handle complex domain problems without compromising accuracy. Motivated by these characteristics, this article develops an RVFLN-based model for forecasting QoS parameter sequences. Results: Two real-world IoT-enabled web service dataset series are used in developing and evaluating the effectiveness of RVFLN-based forecasts in terms of three performance metrics. Conclusion: Experimental results, comparative studies, and statistical tests are conducted to establish the superiority of the proposed approach over four other similar forecasting techniques.

Introduction: The SARS-CoV-2 virus causes COVID-19, a highly contagious disease. Meetings between COVID-19 patients, their families, and medical professionals are no longer safe. To meet their patients, doctors and patients' families must take extreme precautions. Even with these stringent safety precautions, there is a chance that he or she will be affected by COVID-19. In this context, remote patient monitoring via IoT devices can be a highly effective system for today's healthcare system with no safety concerns. Methods: This paper describes an IoT-based system for remote monitoring of COVID-19 patients that uses measured values of the patient's heart rate, body temperature, and oxygen saturation, the most critical measures required for critical care. This device can monitor the observed body temperature, heart rate, and oxygen saturation level in real time and can be easily synchronized with a ThingSpeak IoT cloud platform channel for instant access through a smartphone. When the sensor value exceeds the system's safe threshold, the system will send an email alert to the system user. Some people may notice a decrease in oxygen saturation without any symptoms or respiratory problems. This system can be very useful for early COVID-19 identification in this case. The proposed IoT-based technique is based on an Arduino Uno system and has been tested and validated by a large number of human test participants. As an example, five sample results are shown in this paper. Results: The system yielded promising results. When compared to other commercially available devices, the system's results were found to be accurate, with a maximum error rate of less than 5%, which is quite acceptable. The system's data can be saved in the ThingSpeak cloud server for further analysis. This system requires a unique email and password verification to maintain system security and user data privacy. This patient monitoring system has grown in popularity during this COVID-19 pandemic due to its uniqueness and diverse medical applications. Many people's lives are impacted daily when illnesses are not identified in a timely and accurate manner, denying us the opportunity to provide medical care. To deal with such scenarios, this system will help to monitor a COVID-19 patient's specific parameters, predict the patient's status on a regular basis, and send an email alert to the system user if something abnormal occurs. Conclusion: As a result, this IoT-based smart healthcare solution could help save lives during the current COVID-19 pandemic. This technology is easy to use and reduces the need for human intervention.

Introduction: Wireless communication systems provide an indispensable act in real-life scenarios and permit an extensive range of services based on the users' location. The forthcoming implementation of versatile localization networks and the formation of subsequent generation Wireless Sensor Network (WSN) will permit numerous applications. Materials and Methods: In this perspective, localization algorithms have converted into an essential tool to afford compact implementation for the location-based system to increase accuracy and reduce computational time, proposing a Machine Learning and Cost-Effective Localization (MLCEL) algorithm. MLCEL algorithm is assessed with considered localization algorithms called Support Vector Machine for Regression (SVR), Artificial Neural Network (ANN), and K Nearest Neighbor (KNN). Numerous outcomes show that the MLCEL algorithm performs better than state art algorithms. The simulation is implemented in MATLAB version 8.1 for a network size of 100 nodes. Sensor nodes are positioned in a network area of 100 ×100 m². Conclusion and Results Discussion: The results are assessed on different parameters, and MLCEL achieves better results in localization error 13% 16%, cumulative probability 19%-21%, root mean square error 14%-18%, distance error 17%-20%, and computational time 22%-24% than SVR, ANN, and KNN.

Most people consider traffic congestion to be a major issue since it increases noise, pollution, and time wastage. Traffic congestion is caused by dynamic traffic flow, which is a serious concern. The current normal traffic light system is not enough to handle the traffic congestion problems since it functions with a fixed-time length strategy. Methods: Despite the massive amount of traffic surveillance videos and images collected in daily monitoring, deep learning techniques for traffic intelligence management and control have been underutilized. Hence, in this paper, we propose a novel traffic congestion prediction system using a deep learning approach. Initially, the traffic data from the sensors is obtained and pre-processed using normalization. The features are extracted using Multi-Linear Discriminant Analysis (M-LDA). We propose Tri-stage Attention-based Convolutional Neural Network- Recurrent Neural Network (TACNN- RNN) for predicting traffic congestion. Results: To evaluate the effectiveness of the proposed model, the Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE) were used as the evaluation metrics. Conclusion: The experimental trial could extend its successful application to the traffic surveillance system and has the potential to enhance an intelligent transport system in the future.

Objective: Wireless networks and devices are consuming a significant amount of energy as wireless communication is rapidly expanding, radio frequency (RF) energy harvesting has been envisioned as a feasible technology for powering low-power wireless systems. Aims: This paper investigates a bit error rate (BER) of the non-orthogonal multiple access with cooperative relay-assisted power splitting (CR-NOMA-PS) based energy harvesting (EH). Methods: We consider that the relay works in the decode and forward (DF) mode. For more practical scenarios, we consider the imperfect successive interference cancellation (SIC) and channel state information (CSI) are available. Results: We obtain the end-to-end (e2e) BER expressions for the CR-NOMA-PS with imperfect CSI. Under different scenarios of PS, we evaluate and discuss the BER performance of the users with imperfect SIC and CSI. We validate the derivation of the BER expressions by simulation results. Conclusion: The results indicated that the high values of the PS factor reduce the users' performance. Furthermore, in the high signal-to-noise ratio (SNR), the CSI error degrade BER performance and produced an error floor.

Background: The filter bank multicarrier (FBMC) with offset quadrature amplitude modulation (OQAM) is a promising future generation of wireless systems. However, like multicarrier modulations (MCM), FBMC/OQAM has a high peak-to-average power ratio (PAPR), which allows the FBMC/OQAM signal to pass through the nonlinear region of the high-power amplifier (HPA) in the time domain and causes in-band and out of band (OOB) distortion. Methods: In this paper, a new method to overcome this problem called multi-antennas PAPR (MAP) reduction is proposed. It consists of using I antennas in transmission and reception to transmit I FBMC/OQAM sub-signals with low PAPR. The complementary cumulative distribution function (CCDF), the bit error rate (BER), and the energy efficiency are used to evaluate the method's performance. Results: The simulation results showed that the new technique can reduce the PAPR of the original signal by more than half, achieve BER comparable to that of the original signal without HPA, and when the input back-off (IBO) equals 3dB, the error vector magnitude (EVM) result can be reduced from 19% to 7%. Conclusion: The PAPR, BER, and EVM of MAP technique are much better than the original system.

Background: Kernel Fisher discriminant analysis (KFDA) is a nonlinear discrimination technique for improving automatic modulation classification (AMC) accuracy. Our study showed that the higher-order cumulants (HOCs) of the Long-term evolution (LTE) modulation types are nonlinearly separable, so the KFDA technique is a good solution for its modulation classification problem. Still, research papers showed that the KFDA suffers from high time and space computational complexity. Some studies concentrated on reducing the KFDA time complexity while preserving the AMC performance accuracy by finding faster calculation techniques, but unfortunately, they couldn't reduce the space complexity. Objective: This study aims to reduce the time and space computational complexity of the KFDA algorithm while preserving the AMC performance accuracy. Methods: Two new time and space complexity reduction algorithms have been proposed. The first algorithm is the most discriminative dataset points (MDDP) algorithm, while the second is the k-nearest neighbors-based clustering (KNN-C) algorithm. Results: The simulation results show that these algorithms could reduce the time and space complexities, but their complexity reduction is a function of signal-to-noise ratio (SNR) values. On the other hand, the KNN-C-based KFDA algorithm has less complexity than the MDDP-based KFDA algorithm. Conclusion: The time and space computation complexity of the KFDA could be effectively reduced using MDDP and KNN-C algorithms; as a result, its calculation became much faster and had less storage size.