The next-generation wireless technology modernizes the telecommunication infrastructure by incorporating cloud-based services into its existing framework. Cloud radio access network (C-RAN) was introduced in the 4G era and became popular in recent trends due to the flexibility of sharing resources and decoupling of the data plane and control plane. While this decoupling improves radio resource allocation and processing efficiency, it also introduces challenges in optimizing routes between user equipment (UE), remote radio heads (RRHs), and baseband units (BBUs) under diverse QoS requirements. In this work, the route optimization problem for 5G C-RAN architecture is formulated by developing an optimization function for a dense urban network that consists of 500 UEs connected to 100 RRHs and further linked to 20 BBUs via fronthaul connections. To enhance path selection and routing optimization, several nature-inspired algorithms, that is, ACO, WOA, PSO, GWO, CSO, BOA, and SMO are applied, as these methods are effective for solving complex optimization problems. The performance of these algorithms is analyzed in terms of best fitness value, network efficiency, number of satisfied users, number of blocked users, and number of dropped users under high traffic density. To reduce the impact of randomness in the search process, the experimental results are statistically validated using ANOVA and Tukey's post hoc tests. A detailed comparison of nature-inspired algorithms utilized in this work has been carried out to evaluate their performance, which further confirms the practical suitability and advantages of ACO in 5G C-RAN systems.

For the purpose of creating reliable prediction models for breast cancer screening, precise and effective feature selection is critical. The multiverse optimization (MVO) algorithm is employed in this paper to demonstrate a unique feature selection approach for handling breast cancer datasets. The MVO algorithm is inspired by the notion of parallel universes and how they interact to discover optimal feature subsets by exploring a variety of solution spaces. In an effort to avoid local optima and achieve more powerful feature selection, MVO balances exploration and exploitation by simulating several universes, each governed by unique rules of physics. Our test findings on popular breast cancer datasets reveal that, in comparison with conventional techniques, the proposed MVO-based feature selection significantly boosts classification accuracy and decreases computing complexity. This work establishes a new standard for feature selection methods in machine learning and emphasizes the potential of multiverse-inspired algorithms for enhancement of medical diagnostics.

The widespread availability of miniaturized wearable fitness trackers has enabled the monitoring of various essential health parameters. Utilizing wearable technology for precise emotion recognition during human and computer interactions can facilitate authentic, emotionally aware contextual communication. In this paper, an emotion recognition system is proposed for the first time to conduct an experimental analysis of both discrete and dimensional models. An ensemble deep learning architecture is considered that consists of Long Short-Term Memory and Gated Recurrent Unit models to capture dynamic temporal dependencies within emotional data sequences effectively. The publicly available wearable devices EMOGNITION database is utilized to facilitate result reproducibility and comparison. The database includes physiological signals recorded using the Samsung Galaxy Watch, Empatica E4 wristband, and MUSE 2 Electroencephalogram (EEG) headband devices for a comprehensive understanding of emotions. A detailed comparison of all three dedicated wearable devices has been carried out to identify nine discrete emotions, exploring three different bio-signal combinations. The Samsung Galaxy and MUSE 2 devices achieve an average classification accuracy of 99.14% and 99.41%, respectively. The performance of the Samsung Galaxy device is examined for the 2D Valence-Arousal effective dimensional model. Results reveal average classification accuracy of 97.81% and 72.94% for Valence and Arousal dimensions, respectively. The acquired results demonstrate promising outcomes in emotion recognition when compared with the state-of-the-art methods.

Identification of parking slots using machine learning models is not only required for autonomous vehicles but also helpful to human beings. It assists the drivers in parking their vehicles, which helps to reduce the chances of accidents, increase the utilization of parking space, and reduce time to park the vehicle and many more benefits. Several computer vision algorithms are used to detect parking spaces. These algorithms require a lot of image data to detect objects and distance, independent of environmental factors such as light. In this paper, the author evaluated the performance of 3D Convolutional Neural Network (3D CNN) model for the parking slot detection from dashcams. The 3D CNN is also combined with the Long Short-Term Memory (LSTM) model to evaluate performance. 3D CNN achieved an accuracy of 90.93

Multiple radio access technologies (RATs) encompass various communication mechanisms that connect user equipment to the core network. The 3rd Generation Partnership Project (3GPP) ensures devices can connect to the core network using different RATs (i.e., gNB-NR, Wi-Fi, etc.). The decision to choose the appropriate RAT for individual connections causes a delay in the data transmission process. The standard 5G as defined in 3GPP specification utilizes a common interface to access multiple RATs, whereas the 5G Flow network not only utilizes a unified access network interface to access the multiple RATs, but also one common central controller that controls the connection to avoid delay and throughput losses. To achieve this, the 5G Flow network uses a software-defined network controller and an open flow switch, based on the open flow protocols. This enables centralized control of the RAT as well as decoupling of communication between UE and the core network. On the other hand, the user association also plays a vital role in the RAT selection process which balances the traffic load on the multiple RATs. This paper presents a user association cum RAT selection algorithm which is based on a utility function that takes into account both the user priority and SNR value. The proposed algorithm makes efficient load balancing in multiple RATs and performs well under heavy traffic load especially in eMBB environment. The performance of the proposed approach is compared with 5G Standard and 5G Flow user association approaches under varying circumstances and service priorities. The experimental evaluations are validated using throughput, delay, and spectral efficiency metrics. The results demonstrated a significant gain in the performance of the proposed user association algorithm as compared to the reference study.

All over the world numerous women get affected with breast cancer, which could be controlled and reduced with early and appropriate intervention. This research endeavors to use ML models for prognostic assessment of breast cancer through the analysis of three distinct datasets: WDBC dataset, SEER dataset, and an additional Breast Cancer Prediction dataset. Each dataset offers a unique perspective on the disease, encompassing a wide range of features from cell nucleus characteristics to patient demographic and clinical data. In this paper, five ML models such as RF, NB, and K-NN have been simulated using three datasets. In WDBC dataset, highest accuracy of 0.982 was achieved by LR. In SEER dataset, random forest achieved the highest accuracy of 0.982 and in breast cancer prediction dataset, LR and NB achieved the highest accuracy of 0.942.

The control of robotic systems is a challenging task due to their non-linear dynamics, non-linear actuators, flexibility-induced deformations, environmental factors, singularities, sensor errors, etc. Problems like load variation, uncertainties, friction, coupling, nonlinearities, etc. further complicate the robotic control. Further, precise control of robotic arm manipulators is highly essential to ensure safety, accuracy and reliability in sensitive applications. Therefore, a Fuzzy logic integrated Proportional Derivative controller is designed for 2-link planar rigid robotic arm manipulator. The simulation of two degrees of freedom link planar robotic arm is carried out in MATLAB. The designed controller is compared with traditional PID and PD controller in terms of error-based parameters. The results reveal that the Fuzzy Logic integrated PD control scheme efficiently evaluates the required joint torques, based on fuzzy rules and outputs of PD controllers. It is also more challenging to control link 1 as compared to link 2 because link 2 acts as an additional weight on link 1. Hence the proposed controller outperforms PID and PD controllers with deviations from the desired trajectory.

ABSTRACTTraditional MPPT techniques have inherent limitations such as slow convergence, incompetent toward rapid and extreme atmospheric conditions, and oscillations at maximum power peak (MPP). This paper presents a novel adaptive maximum power point tracking controller to mitigate the drawbacks of traditional MPPT techniques. In the devised control strategy, proportional and integral controller gains are made adaptive with respect to instantaneous error, which in turn reduces the power losses. Further, non‐integer integral and derivative coefficients of PID controller have been employed to enhance the accuracy. Here, Lambert‐W function is used to convert implicit equation of solar PV model into explicit equation to calculate the output power and its derivative. Parameters of the proposed controller are optimized using dragonfly optimization algorithm. The proposed scheme is analyzed on a string of KC130GT solar module consisting of a 4S1P configuration. The performance of the proposed controller is examined in terms of convergence rate and power tracking competency under six different irradiance profiles, that is, (i) constant irradiance, (ii) slow varying irradiance, (iii) fast varying irradiance, (iv) sinusoidal irradiance (v) extreme varying irradiance, and (vi) partial shading condition. Moreover, the performance of the proposed controller is investigated over other state‐of‐art MPPT controllers based on efficiency analysis, RMSE value, and time response analysis. The result reveals that the proposed technique offers the highest tracking efficiency of 99.38% under constant irradiance, whereas PID and AN‐PID recode the average efficiency of 99.05% and 98.79%, respectively. It can be noted that ANFOPID recorded the highest tracking efficiencies for other irradiance patterns in comparison with other designed controllers. Moreover, AN‐FOPID consistently exhibits the lowest RMSE values and efficient time response performance, indicating its superior performance compared to the PID and AN‐PID techniques.

Fifth-generation mobile technology provides ultra-reliable low-latency communication (URLLC) for mission-critical applications, massive machine-type communication (mMTC) for IoT connectivity, and enhanced mobile broadband (eMBB) services for high-speed data transfer. To accomplish this goal, beamforming plays a crucial role in radio resource allocation by directing a focused beam toward the user. Some of the beamforming mechanisms calculate the optimal beam pair index based on the channel state information (CSI) and signal to interference and noise ratio (SINR). To identify the optimal beam pair index in the beam selection procedure, it is essential to employ a unique technique that minimizes overhead during beam sweeping and selection while maintaining low complexity. In this paper, we used neural networks for beam selection based on the Global Positioning System (GPS) coordinates of the receiver. Neural networks take GPS coordinate of the receiver and optimal beam pair index as input to train the model. In the output of the neural network, the K optimal beam pair indices are chosen based on the average reference signal received power (RSRP). Neural networks contribute to the high accuracy and average RSRP beam selection as compared to the benchmarked algorithm.

5G NR enables three types of use case scenarios viz. enhance mobile broadband (eMBB), ultra-reliable low latency communication (URLLC), and massive machine type communication (mMTC). The eMBB is suitable for applications that demand higher throughput. Whereas URLLC is suitable for mission-critical applications with stringent requirement of low latency and reliability. The mMTC on the other hand is suitable for machine-to-machine (M2M) communications and massive IoT infrastructures. To meet all these requirements, 5G NR combines eMBB with URLLC services under a unified 5G air interface framework. Most of the coexistence mechanisms between eMBB and URLLC were presented by the researchers with a goal to enhance eMBB throughput with stringent latency and reliability requirements of URLLC services. Formulation of optimal resource scheduling and allocation were found to be the key problems of eMBB and URLLC traffic. This paper investigated the 5G state-of-the-art focused on coexistence mechanisms between eMBB and URLLC traffic for resource scheduling. We followed the PRISMA statement and classified the works to five major classes viz. multiplexing-, QoS-, Machine learning-, Network slicing-, and C-RAN architecture-based approaches. Each work was carefully examined against their methodology and performance metrics. In addition, several key issues, challenges, and future directions were also highlighted to provide detailed insights for researchers working in the field of 5G.

5G New Radio (NR) operates in two frequency ranges viz. Frequency Range 1 (sub-6 GHz band) and Frequency Range 2 (millimeter wave communication). It utilizes Time Division Duplex (TDD) communication, allowing the same frequency band for uplink and downlink transmission using sounding reference signals. These signals help estimate channel conditions for scheduling downlink traffic. Multi-User Multiple Input Multiple Output (MU-MIMO) technology enables simultaneous data transmission to multiple users using multiple antennas. This study examines MU-MIMO performance in TDD by analyzing variables like radio conditions, modulation schemes, code rates, precoding algorithms, and sub-carrier spacing. Simulation results show that Zero Forcing (ZF) precoding outperforms Block Diagonalization (BD) and Regularized Zero Forcing (RZF) in good radio conditions. However, in bad radio conditions, RZF performs better than ZF. Choosing higher code rates in 16 and 64-QAM and lower code rates in 256-QAM from the MCS table improves throughput. This indicates that MCS and user selection are crucial for higher throughput in TDD-based MU-MIMO for 5G NR.

There has been an exponential increase in discussions about bias in Artificial Intelligence (AI) systems. Bias in AI has typically been defined as a divergence from standard statistical patterns in the output of an AI model, which could be due to a biased dataset or biased assumptions. While the bias in artificially taught models is attributed able to bias in the dataset provided by humans, there is still room for advancement in terms of bias mitigation in AI models. The failure to detect bias in datasets or models stems from the "black box" problem or a lack of understanding of algorithmic outcomes. This paper provides a comprehensive review of the analysis of the approaches provided by researchers and scholars to mitigate AI bias and investigate the several methods of employing a responsible AI model for decision-making processes. We clarify what bias means to different people, as well as provide the actual definition of bias in AI systems. In addition, the paper discussed the causes of bias in AI systems thereby permitting researchers to focus their efforts on minimising the causes and mitigating bias. Finally, we recommend the best direction for future research to ensure the discovery of the most accurate method for reducing bias in algorithms. We hope that this study will help researchers to think from different perspectives while developing unbiased systems.

Fifth generation new radio (5G NR) is bringing revolutions in almost every field such as healthcare, industrial automation, smart farming, and defense. These fields utilize three major use case scenarios of 5G NR viz. eMBB, URLLC, and mMTC. 5G NR deployment relies on two major frequency bands such as sub-6 GHz and high-frequency bands (millimeter waves). Among all sub-6 GHz bands, the n78 band (3300–3800 MHz) is gaining popularity due to its unique characteristics viz. higher capacity and coverage, wide bandwidth, and higher data rate. Despite these unique characteristics, in the 5G standalone architecture, users expect high-speed video/voice transmission without delay and good quality of services (QoS) requirements. QoS is one of the important paradigms that indicates the quality of services experienced by end users instead of services claimed by service providers. In this work, we have designed a QoS scenario that utilizes the n78 band with an “urban macro mixed office scenario with low loss” path loss model. As per types of services are concerned, the scenario utilizes six different traffic templates (i.e., FTP, Voice, Video, COAP, CBR, and Email). The path loss model operates on non-line-of-sight scenarios with Rayleigh fading and eigen beamforming using 64 × 32 antenna elements. The performance of the proposed scenario is examined through different metrics i.e., throughput, packet drop ratio, and delay. Here, priorities to different traffic templates are assigned based on GBR and non-GBR resource types. In addition, the proportional fair scheduling algorithm is employed to schedule the resources. It was observed that video services achieved a higher throughput of 25.93 Mbps due to their prioritization in the QoS model. Conversely, FTP services experienced higher delay, measured at 64 ms, and a packet delivery ratio of 5.94

Emotion recognition is vital for augmenting human–computer interactions by integrating emotional contextual information for enhanced communication. Hence, the study presents an intelligent emotion detection system developed utilizing hybrid stacked gated recurrent units (GRU)-recurrent neural network (RNN) deep learning architecture. Integration of GRU with RNN allows the system to make use of both models’ capabilities, making it better at capturing complex emotional patterns and temporal correlations. The EEG signals are investigated in time, frequency, and time–frequency domains, meticulously curated to capture intricate multi-domain patterns. Then, the SMOTE-Tomek method ensures a uniform class distribution, while the PCA technique optimizes features by minimizing data redundancy. A comprehensive experimentation including the well-established emotion datasets: DEAP and AMIGOS, assesses the efficacy of the hybrid stacked GRU and RNN architecture in contrast to 1D convolution neural network, RNN and GRU models. Moreover, the “Hyperopt” technique fine-tunes the model’s hyperparameter, improving the average accuracy by about 3.73

This paper introduces a research endeavour that utilizes a fractional-order control methodology based on Particle Swarm Optimization (PSO) to address the intricacies of the non-linear Twin Rotor Multi-input Multi-Output System (TRMS). The primary purpose of this research is to achieve propeller trajectory tracking and system stabilization. Various control schemes, including the Linear Quadratic Regulator with a Proportional Derivative (PD) controller, the Proportional Integral Derivative (PID) controller, and the Non-Linear Fractional Order PID controller, are assessed for their performance on the interconnected system. The pitch and yaw angles are independently regulated by controllers specifically designed for the tail and main rotors. The parameters of these controllers are optimized using the PSO technique. The results indicate that the most effective controller among those tested is the enhanced PID controller, specifically the PSO-based Fractional Order PID controller.

Modern Programmable Logic Controllers (PLC) programming tools support object-oriented programming and pointers. In this paper, we show how it is possible to link any number of motor controllers like a chain and obtain a unified interface with redundancy but without using a master. The main principle is twofold, encapsulate existing and well-proven function blocks within other function blocks with added support for pointers and actions (methods). Then add one manager function block for each function block and link these manager function blocks to each other to create a chain. The first manager has no predecessor, and the last has no successor. Status information will flow upward to the first manager, and commands will flow toward the last manager.

This paper aims to analyze the capability of 2DOF PID controller while dealing with multiple and conflicting objectives problem. A basic PID control system fails to track the reference input and reject the disturbance simultaneously. This problem may be better dealt with the use of a two-degree of freedom controller which combines the effect of two controllers in separate loops to meet the two criteria. The problem considered in this study is the drug dose control in chemotherapy. In this study, multi-objective GA and multi-objective swarm optimization algorithms are implemented to obtain the optimum amount of the drug delivery for chemotherapeutic treatment. The designed controller is compared with the conventional PID controller. Various analyses like step response analysis, bode analysis, parameter perturbation analysis and disturbance analysis are carried out to justify the performance of the designed controller.

5G NR has been deployed in most of the countries and covers three major use case scenarios, viz. enhanced mobile broadband (eMBB), ultra-reliable low latency communication (URLLC), and massive machine type of communication (mMTC). To enable the 5G services on the existing 4G infrastructure, the non-standalone architecture (NSA) of 5G NR is equally popular. For NSA deployment, many countries are utilizing the 3.5 GHz band on time division duplex (TDD) transmission. For both uplink and downlink transmissions, TDD is utilizing same frequency bands and scheduling of resources in such environments is quite challenging. Here, the performance of 5G NR is analysed on 20 MHz and 100 MHz channel bandwidth in terms of resource utilization by comparing the performance of different scheduling algorithms, i.e. Professional Fair, Round Robin, and Best CQI on different MIMO (64 × 64, 32 × 32, 16 × 16, 8 × 8) set-up and block error rates. The performance of Best CQI algorithm is found to be better than other scheduling algorithms under considerations.

In agriculture, weed infestation presents various challenges to quality farming. They (weed) compete for resources and hinder a crop’s growth. To combat weed infestation on cropland, farmers use expensive fertilizer which uses toxic chemicals that can have lowered crop yield and quality. Fertilizer application is labor-intensive if done manually, automation is prohibitively expensive and resource-intensive. With the rise in popularity of deep neural networks especially CNN for image classification, it is an excellent candidate to automate weed classification in real-time and potentially automate weed removal using robots/UAVs cost-effectively. The paper adopts dual training approach, one from scratch and the other using transfer learning on 17,509 weed images with nine different classes of weed on five CNN architecture. Various preprocessing techniques are employed to further increase the accuracy of the models. ResNet50V2 achieved the highest classification accuracy of 93%.