Table of contents

4^th National Conference on Communication Systems (NCOCS-2022) December 23^rd, 2022

ABOUT THE CONFERENCE

National Conference on Communication Systems (NCOCS-2022) is fourth in the series of conferences intended to be held every year at NITPY, aimed to disseminate knowledge on the latest technological developments and cutting-edge research in wireless communication, signal processing, antenna and RF design, optical communication, internet of things and sensor networks & electronics among academicians, researchers, scientists, industry personnel, and entrepreneurs across India. This conference will provide the premier interdisciplinary forum relevant with Electronics and Communication Engineering to discuss the recent technologies, innovations, recent research findings, challenges, solutions, and perspectives of the future directions of Communication Engineering through keynote lectures and paper presentations. Young researchers are encouraged to participate and utilize the opportunity to network with others who seek to develop their research abilities and experience.

OBJECTIVE & SCOPE OF THE CONFERENCE

The objective of the NCOCS 2022 is to provide a platform to perceive, share, and exchange the recent trends related to various aspects of Communication Engineering. The main aim of this conference is to bring together researchers, young faculties and practitioners working in multi-disciplinary fields to discuss and deliberate on challenges, opportunities and strategies involved for next-generation communication systems. The conference will provide a forum to academic researchers and practitioners to discuss critical issues concerning latest technologies towards 6G and mmWave communications.

Conference Chairperson

Dr. M. Surendar

Assistant Professor, Department of ECE,

National Institute of Technology Puducherry, Karaikal – 609 609, India

Conference Website: https://sites.google.com/view/ncocs-2022/home?authuser=0

List of Conference Committee is available in this Pdf.

All papers published in this volume have been reviewed through processes administered by the Editors. Reviews were conducted by expert referees to the professional and scientific standards expected of a proceedings journal published by IOP Publishing.

• Type of peer review: Single Anonymous

• Conference submission management system: Morressier

• Number of submissions received: 85

• Number of submissions sent for review: 72

• Number of submissions accepted: 45

• Acceptance Rate (Submissions Accepted / Submissions Received × 100): 52.9

• Average number of reviews per paper: 1

• Total number of reviewers involved: 19

• Contact person for queries:

Name: Dr. M. SURENDAR

Email: surendar.m@nitpy.ac.in

Affiliation: Assistant Professor, National Institute of Technology Puducherry

Rate splitting multiple access (RSMA) and intelligent reflecting surface (IRS) are the promising candidates and front runners for 5G and beyond wireless communication. Recent studies have proved RSMA-IRS outperforms non-orthogonal multiple access (NOMA) and enhances the quality of service (QoS). The performance of the downlink RSMA-IRS system is investigated in this paper using perfect and imperfect channel state information (CSI) conditions over the Rayleigh fading channel. Furthermore, the numerical analysis for the sum rate capacity is carriedout for the users under perfect and imperfect conditions. To assess the system's performance, two distinct scenarios are explored. Initially, the system total capacity of each user's common and private parts is determined without considering the channel estimate error (CEE). The total capacity for each user's common part is 10 b/s, for private1 is 8.9 b/s, and for private2 is 6.5 b/s. After introducing CEE, the sum rate for the common part is 8.5 b/s, and the total capacity for private1 and private2 is 7.5 b/s and 6 b/s, respectively. The simulation results show that the perfect CSI, achieves enhanced sum rate than channel with CEE. however, to tolerate the imperfect scenario (CEE) optimum power allocation is determined for attaining the QoS.

This article examines coverage probability for a multi-user hybrid satellite terrestrial relay network (HSTRN) with opportunistic user scheme. Better quality information are gathered via HSTRN-based networks for users on the ground. Simultaneous satellite signals are amplified by relays and relayed to terrestrial users with the highest instantaneous signal-to-noise ratio (SNR) on an individual channel. Analytically approximate coverage probability is derived for fixed and variable gain relaying. The multi-user relay network is analyzed under Nakagami-m fading, and Rician shadowed fading. We perform simulation and analytical validation to validate our study and compare the performance of fixed and variable gain relaying.

A novel multicarrier system based on chaotic sequences with permutation and index modulation to improve spectral efficiency is presented in this paper. In this design, the quasi-orthogonality property of the permuted versions of chaotic sequences is used for getting multiple reference signals, thereby reducing by the transmission overhead of reference chaotic signals over additional subcarriers. This helps in improving the spectral efficiency of the proposed system. Also, index modulation is applied to the permuted chaotic signals in all subcarriers to provide better efficiency in terms of spectrum utilization as well as energy consumption. Hence, the information bits are conveyed by both M-ary modulation symbols and the indices of the selected permutated signals. In addition, the analytical expressions for bit error rate (BER) of the proposed system over additive white gaussian noise (AWGN) is derived and simulation results are used to validate the analytical values. The performance of the proposed system is compared with the multicarrier chaotic system available in the literature, thereby proving that the proposed system outperforms the existing system with improved spectral efficiency and energy efficiency and also provides improvement in BER performance of about 2dB over AWGN channel with hardware complexity trade-off.

In this article, the user resource allocation and user scheduling in a cell-free user-centric clustering network and coherent transmissions is investigated. However, the setup of the user-centric clustering is not sufficient for multi-user transmissions, and user scheduling becomes difficult. To overcome the aforementioned problem and to provide user fairness, an optimum resource allocation algorithm with perfect user scheduling, beam forming, and radio frequency transmitter (RFT) clustering is proposed. Thereby, the cell edge user achieves the optimum throughput and spectrum efficiency. The simulation result shows that the proposed method outperforms the conventional method. In addition, the pilot-reusing factor is used to obtain loss due to imperfect channel state information and pilot train overhead.

The growth of IoT and data-driven applications has caused network evolution with a demand for high internet speeds capable of handling data-intensive operations and high-performance computing in real-time. Despite massive strides in the past decade to achieve low-latency ultra-reliable network connections, by 2030, 5G technology will be unable to keep pace with emerging applications like unmanned autonomous vehicles, brain-computer interactive devices, extended reality (XR), ubiquitous AI, Internet of Everything and global network coverage. Subsequently, there will arise a need to transcend from 5G to 6G. 6G is being touted as the next giant leap forward to support ultra-fast intelligent networks resulting in speculations about the various use cases which will adopt 6G. We discuss the enabling technologies and recent advancements in the development of 6G. We talk about the most advanced and promising technologies that will revolutionise 6G communication. Furthermore, we give an objective analysis of these technologies and possible hindrances in their adoption.

One of the vital entity of any communication systems is Channel coding, which leads to the design of high performance codes for future wireless systems having low complexity encoder and decoder design. These systems will have a requirement of operating in highly reliable conditions, maximum throughput, and low and high code rates and to work with short and long information messages. Polar codes are one of the promising error correcting channel codes that can be used in these situations to obtain maximum throughput and coding gain in a communication systems because of their capacity approaching performance and finds interests in Satellite communication and 4G/5G services. These codes uses the concept of channel polarization to be constructed. The work proposed in this paper focusses on the Bit Error Rate evaluation and analysis of the Polar codes using traditional approach and Deep learning approach. The feedforward deep learning networks using different activation functions were used for the Deep learning approach. The Successive Cancellation algorithm and its variant using List decoding were used as traditional decoding methodology. The results obtained using Deep learning approach were satisfactory and was matching as per the traditional decoding.

This paper offers a method for developing an efficient indoor navigation system with the consideration of the shortest path between source and destination. The challenge for the indoor navigation system is to provide personal navigation information and the optimal route. Applications of indoor navigation systems need consideration of the Shortest Path problem. The shortest pathways can be used to find solutions to the current problems using Dijkstra's algorithm. Based on the issue with the indoor navigation system, the shortest way and the best path are calculated. This is crucial to navigation systems since it can aid in making wise decisions and time-saving choices. The primary goal is to obtain the implementation at an affordable price. These applications and services are made available indoors, where the GPS does not function. The goal of indoor navigation is to direct users inside buildings. Dijkstra's algorithm for locating objects and for moving along the shortest path in an indoor setting are examined in this work. Experimental results of indoor navigation systems were carried out on my organization's indoor environment and verified the applicability of the presented Indoor Navigation System. The techniques provided include map digitization, locating a user, and choosing the shortest route. This is accomplished through a mobile application created for the Android operating system, and indoor navigation is carried out by using Dijkstra's algorithm. The proposed method is implemented in our college academic block, and the experimental results show that our navigation method is feasible and effective. To verify the reliability of the algorithm, the proposed application fulfils the criteria of an indoor navigation system to produce the optimal route between two points when applied to a map of our college's indoor terrain.

Speech recognition is a method where an audio signal is translated into text, words, or commands and also tells how the speech is recognized. Recently, many deep learning models have been adopted for automatic speech recognition and proved more effective than traditional machine learning methods like Artificial Neural Networks(ANN). This work examines the efficient learning architectures of features by different deep neural networks. In this paper, five neural network models, namely, CNN, LSTM, Bi-LSTM, GRU, and CONV-LSTM, for the comparative study. We trained the networks using Audio MNIST dataset for three different iterations and evaluated them based on performance metrics. Experimentally, CNN and Conv-LSTM network model consistently offers the best performance based on MFCC Features.

Digital cameras and mobile image capture of documents are two examples of new developments in the fields of optical character recognition and text recognition. Scans of text or text photographic images and even natural photography results can be distorted to the point where OCR digitization is inaccurate. It offers a unique non-parametric unattended approach to correct unwanted document image distortions to achieve optimal OCR accuracy. It applies a highly effective stack of document image enhancement algorithms to restore perfect images distorted by unknown sources of distortion. First, it provides a means of modifying local brightness and contrast in order to better handle different illumination levels and atypical light transmission patterns in the image. Then apply a nifty grayscale conversion method to your photo to give it a new look. Third, it uses unsharp masking techniques to further enhance important details in grayscale images. Finally, we use the best global binarization technique to prepare the final document image for OCR recognition. The proposed technique has the potential to significantly improve the text recognition rate and accuracy of optical character recognition.

This paper presents the state-of-the-art literature on Multiple Input Multiple Output (MIMO) antennas for 5G systems. The degree of freedom offered by MIMO as a result of increasing transmitter and receiver elements has raised serious concern towards the impact due to mutual coupling between the adjacent elements in the MIMO antenna. Hence, this paper summarizes the various measures for suppressing mutual coupling between MIMO antenna elements. It also reports the methods for enhancing the bandwidth at the sub 6 GHz band by the deployment of different radiating structures and their influence over the crucial MIMO parameters. A study on the ways and means of handling the propagation losses at mm-wave frequencies by proper choice of PCB for fabrication is also discussed.

In this paper, a compact 2 X 2 MIMO antenna is proposed for 5G wireless applications. The proposed 2 X 2 MIMO antenna is designed using four compact circular monopole antennas which are optimized to operate from 3.6 to 57.7 GHz. The antenna structure consists of four orthogonal antenna elements of size 21 mm × 29 mm, which are spaced 1mm apart from each other. The antenna and the defective ground plane are placed on the same side of the substrate. It is demonstrated that MIMO system with 4- antennas produce very good isolation (>-20 dB) without using any additional isolating or decoupling techniques and antenna efficiency is also high. The antenna is simulated using CST Microwave Studio. The simulation results of the proposed MIMO antenna gives good radiation efficiency, Envelope Correlation Coefficient (ECC), improved bandwidth of 54.1 GHz, and optimum radiation characteristics. The proposed compact 2 X 2 MIMO antenna is suitable to operate under 5G wireless applications from 3.6 to 57.7 GHz. The applications of the proposed antenna also include broadcasting satellites, Modern radars, a 5GHz Wi-Fi channel, TV broadcasting satellites and microwave devices

MIMO antenna are widely used in many areas for the great reception of signals. In transport industry for the development of automation and for the advancement they required efficient and high gain antenna transceivers. In this paper, single-element dual band antenna is used for 5G Vehicle to Everything (5G-V2X) application but the design will be enhanced to six-element for full coverage, on analysing single element. Three antennas of same structure as specified in this paper are arranged in a single substrate as a shape of triangle of combining all three, and similar structure is made in another substrate such that both are separated by a strip like structure and arranging both substrates that both triangularly arranged substrate forming a hexagonal structure from the top view. These will increase the coverage over 360° surrounds these will increase the gain over 9dB at mm-Wave range.

In this paper, two compact T-Slot Microstrip antennas have been proposed, designed, and presented using design equations. Each antenna design has the same dimensions and characteristics. High-frequency structure simulator has been used for the proposed design and output has been verified. FR-4 Substrate which is having 1.6mm thickness is used to mount the prototype. The patch operates for 25.75 GHz resonant frequencies and T-Shaped slot has been designed for K-Band applications. In this work, 50 x 40 mm² measurements is used. The antenna designed is having perfect radiation patterns with good VSWR, Return loss, and Radiation efficient achieved.

In this paper, we report a liquid-based tunable Cylindrical Dielectric Resonator Antenna (CDRA), whose resonance is varied by loading the cavity installed in the CDRA with different volumes of liquid. The tunability of the CDRA is investigated by loading the cavity with two different types of liquids, one with lower permittivity (i.e.) oil and the other with higher permittivity (i.e.) water. CDRA loaded with oil is found to be tunable over a small frequency range of 32MHz (5.76GHz to 5.728GHz), whereas the change in water volume in the cavity provides a wide tuning range of 452MHz (5.76GHz to 5.308GHz). Further, the antenna gain is found to be almost constant, around 5.1 dB, for oil-based CDRA. In a water-filled cavity, the gain is observed to reduce by 1dB as the volume is increased from 0 to 100%.

This research article discusses the investigations on the design of an Ultra-Wideband (UWB) antenna and its performance analysis from 20 to 30 GHz frequency range which is suitable for Short Range (SR) RADAR applications. This is designed by incorporating a rectangular slot on rectangular patch and reducing the ground plane size. The antenna characteristics namely return loss, current distribution, E-plane and H-plane far field patterns, gain over frequency, group delay in the operating frequency and radiation efficiency are investigated through simulations. Brief discussion regarding FCC guidelines related to the use of 24 GHz frequency band is provided. The antenna occupies the size of 10 X 10 X 1.6 mm³. Necessary parametric analyses were performed to confirm the finalized dimensions as optimum. The impedance bandwidth characteristics was measured using VNA (Vector Network Analyzer) and the obtained –10 dB bandwidth agrees with the simulated result.

This research proposes a unique design that makes use of a defective microstrip topology to decrease electromagnetic coupling between parallel high speed interconnects. The performance of this new microstrip construction with defects is evaluated using near-end crosstalk and far-end crosstalk. Serpentine shaped defected microstrip is introduced in one parallel of high speed interconnects and its performance is also compared the existing structure. In this research, the proposed defected microstrip structure (DMS) is simulated and compared with a conventional microstrip structure using Ansoft HFSS software which employs the Finite Element Method (FEM). Simulation results indicate that the DMS design effectively reduces crosstalk in comparison to the previous structure. It reduces more than 5dB near end crosstalk and more than 3dB far end crosstalk compared to conventional model.

Wireless communication continues to lead the world. The term "wireless" significantly influences everything from telephone to satellite communication. Recent years have seen the need for appropriately built Low Noise Amplifiers (LNAs) in receiver front end because of the modem wireless communication in broadcast and microwave radio. In this paper a high performance LNA for the radio frequency range is presented. To achieve high gain a three stage LNA with each stage as Cascode stage is considered here. The simulation shows that the presented LNA has a gain of 26.12 dB and a Noise Figure (NF) of 2.6 dB. The achieved output and input returns losses are -8.4 dB and -11.6 dB respectively.

Quantum Cascade Lasers (QCLs) as Terahertz (THz) frequency sources offer a potentially viable solution for new applications in mid and far-infrared frequency bands. This research work exhaustively investigates the temperature dependence on the impedance of temperature dependent Quantum Cascade Lasers (QCLs) operating at 116μm, for the first time. In the 90-stage QCL considered for the work, the cold finger temperature is varied from 15K to 45K. When the device is biased at 0.6A current along with a cold finger temperature of 45K, the magnitude of intrinsic impedance was found to be 23.91mΩ, at a frequency of 4GHz. As the cold finger temperature is increased from 15K to 45K, the impedance response of the device becomes flat and stays constant. At 45K with an injected current of 1.5A, maximum impedance of 3.1mΩ is obtained. The resonant frequency characteristics of the device increase with increase in injected current and cold finger temperature. Also, it is observed that the magnitude of intrinsic impedance decreases with increase in injected current. The impact of cold finger temperature on the intrinsic impedance characteristics are detailed for prospective Radio over Fiber (RoF) applications.

Developing technology including the Internet of Things, digital services, smart microgrids, and machine-to-machine systems encourage the implementation of self-configuring, automated systems in massive wireless sensor networks (WSNs). Effective Power utilization is critical in order to keep a network operational for the maximum possible time. Sensor nodes, which are small, battery-powered devices, are used in WSNs. As a result, one of the significant research issues in improving the lifespan of a WSN is resource management. A modified k-means (Mk-means) algorithm for clustering has been proposed to select Cluster Head (CH) to minimize energy usage of the nodes. Several optimization methods have been suggested in this area to extend the WSN lifetime. Following that, the Firefly Algorithm (FFA) is employed to generate the optimal routing through the CHs to a Base Station (BS), where multiple fitness functions such as residual energy, distance, and routing traffic are taken into account to optimize the FFA. As a result, information transmission among intermediate CHs in a hierarchy cluster-based design aids in lowering node energy usage. The proposed MKMFA (modified k-means Firefly Algorithm) technique's performance is evaluated with K-means AODV (Ad-hoc On-demand Distance Vector) and IPC-KMAN(Improved Performance Clustering Using Modified K-Means Algorithm in Mobile Adhoc Networks) using network lifetime, throughput, Load Balancing Factor (LBF), and Packet Delivery Ratio (PDR). This work is based on recent advances in WSNs, which include application fields, design parameters, and lifetime prediction designs.

A robot is a machine that can automatically do a task or a series of tasks based on its programming and environment. They are artificially built machines or devices that can perform activities with utmost accuracy and precision minimizing time constraints. Service robots are technologically advanced machines deployed to service and maintain certain activities. Research findings convey the essential fact that serving robots are now being deployed worldwide. Social robotics is one such field that heavily involves an interaction between humans and an artificially built machine. These man-built machines interact with humans and can also understand social terms and words. Modernization has bought changes in design and mechanisms due to this ever-lasting growth in technology and innovation. Therefore, food industries are also dynamically adapting to the new changes in the field of automation to reduce human workload and increase the quality of service. Deployment of a robot in the food industries which help to aid deaf and mute people who face social constraints is an ever-growing challenge faced by engineers for the last few decades. Moreover, a contactless form of speedy service system which accomplishes its task with at most precision and reduced complexity is a feat yet to be perfected. Preservation of personal hygiene, a better quality of service, and reduced labour costs is achieved.

In current decades, employing clinical imagery analysis to automatically segregate exudates from color fundus pictures has proven to be a difficult endeavor. This paper compares the efficacy of several picture delineation techniques using a Raspberry Pi chip. By employing various techniques to regular publically accessible samples, the optimum delineation methodology is selected, while efficacy is measured using characteristics such as resemblance factors, implementation duration, sensitivity, as well as specificity. The source hue ocular pictures are initially obtained using publicly available resources. Gaussian distortion, impulse distortion, and speckle distortion could all be present in such pictures. As a result, a pre-processing approach is used to the source pictures in effort to reduce the distortion and boost brightness. After that, several delineation methods such as a thresholding technique, mean-shift algorithm, watershed algorithm, distance transform, K-means clustering, Fuzzy C-Means grouping approach and Active Contour Model are used to segment the normal and abnormal region in color fundus images. The Fuzzy C-Means grouping approach yields higher delineation precision yet requires longer execution time, according to the findings.

The main idea of the paper is to get individuals to monitor their body mass index (BMI) and temperature. The BMI is a technique used to test people for potential weight issues and to track the weight status of people. Body mass index is calculated by using two main parameters they are height and weight of the person. We develop a commercial device, which calculates the BMI of the user with their respective height and weight. When the user stands in front of the device, The device scans the user's face and classifies the user's gender as correct BMI of user is dependent on gender, Later the height is calculated using ultrasonic sensor and weight is also measured. The device also shows the temperature and this temperature is measured by MLX 90614 sensor Now the device calculates the BMI and displays the user's BMI to the user. The device also displays some message to the user according to their BMI, such as whether the person is over weighted or low weighted.

Robots are playing key role in today's world. Robots are actively taking part in the industries, hospitals, defense applications as well. This paper describes about how hand gesture movements can control the robotic arm. Hand gesture data is recorded by the leap motion controller. That captured gestures are then processed and sent to the microcontroller. In this paper we are intended to design a robotic arm that is completely useful to the people with disabilities. We are using new kind of sensor called Leap motion controller, which captures the hand gestures in 3D. we develop a robotic arm that is useful for the paralyzed people, industrial applications, medical fields by just simply using the hands, we can guide the leap motion controller so that the robotic arm can perform its operations.

The Smart Grid model uses energy from renewable sources and utilities to manage energy in a way that ensures uninterrupted power supply for users. In order to operate independently from the main power grid, a DC micro grid is a small power system that generates and uses its own DC electricity. Solar panels, fuel cells, and wind turbines are the most typical power sources for DC micro grids, with batteries used to store any extra energy. The independence of the power grid is one feature of a DC micro grid that promotes increased lifetime. Building owners have more latitude to pursue their sustainability goals since they have control over the production, delivery, and consumption of power. Super-twisting provides a dependable method and an effective instrument for the control of uncertain nonlinear systems by addressing the basic flaws of conventional sliding mode control, notably large control effort and chattering. Fractional order controllers provide greater design freedom than conventional integer order controllers. The purpose of this work was to create a metastable-smart grid(MSSG) user end model that could dynamically manage and optimize energy produced by renewable energy sources and the utility to ensure that customers always have access to electricity. Controlling these many power sources using MSSG and delivering energy in accordance with each user's unique power plan is the responsibility of the central control unit of the distribution grid. The supply is only switched to solar by the central control unit if a user's power consumption falls below a predetermined threshold. Giving consumers with less purchasing power more clout as a result.

Traffic due to automobiles is a major problem in many urban places in India as well as in other countries. Malfunctioning of the traffic lights and various other dysfunctionalities has led to traffic congestion which hurts the economy, environment, and overall quality of life. Hence it is very important to manage the traffic congestion problem effectively. In this paper, a prototype is proposed for a traffic management system using IR sensors, Arduino, serial to the parallel shift register, and LED displays. The density of the traffic is measured by placing the IR sensors at the 4 lane junction after a certain distance. The data collected from sensors is used to dynamically change the sequence of green lights as well as to dynamically change the green light delays. The proposed road traffic management system is implemented in the proteus software. For high traffic zone, a higher green light delay is given whereas for low density zone, the duration of the green light delay is reduced. The system is also designed in such a way as to give preference to the lanes which have no vehicles/very few vehicles by providing the least green light delays.

CT (Computed Tomography) and MRI (Magnetic Resonance Imaging) are both used to diagnose internal damage and deformities of the organs. These are 3D data so doctors diagnose these scans by splitting them into many 2D slices and analyse each slices for detecting disease. This is a hard and time-consuming task to accomplish. Experts are enough trained to analyse this data for interpreting insights. But when they try to explain vulnerable situations to their patients, these scans are not very readable and understandable for them. Thus, it scares them a lot when they are asked to take these scans for diagnosis. Thus, an approach is given to viewing these scans in AR (augmented reality), making them available to view via their smartphones. The idea is to build an application that can take CT and MRI scans as input from ".stl" files and help to view them in AR with plane detection for mesh to be rendered and placed properly. This also reduces the task of storing these scans in bulk. Instead, one can just send data to their patients and store it as records on their phones. Whenever necessary, one can visualise it on the spot.

The fast and precise water quality prediction in smart aquaculture can assist the farmers in taking corrective actions before the ecological environment deteriorates. However, the empirical methods take high computation time and provide low accuracy in water quality prediction due to the non-linear and dynamic nature of water quality parameters. This research work proposes long short-term memory (LSTM) and gated recurrent unit (GRU) deep learning recurrent neural network (DL-RNN) models for aquaculture water quality prediction (A-WQP). This work also presents an extensive study about the impact of hyper-parameters (h_p) on the performance of the proposed DL-RNN models using two different water quality datasets. The experimental evaluations show that for the optimal sets of hyper-parameters, the proposed DL models offer superior prediction accuracy and computation efficiency.

The lack of favourable atmospheric conditions leads to the loss of many crops each year. In India alone, over 11 billion dollars are lost. By combining IoT and machine learning technologies, this team has created a system that integrates agriculture's three primary operations: crop selection, autonomous watering, and fertiliser suggestion. The following crops—Apple, Rice, Maize, Grape, Banana, Orange, Cotton, and Coffee—were considered in the study. Three systems are covered in the paper: The crop recommendation system employs machine learning to examine factors including nitrogen (N), phosphorous (P), potassium (K), pH, and weather before recommending a crop. The crop type and the current levels of soil nutrients are the two main determinants on which the fertiliser recommendation method bases its recommendation. When employing an automatic irrigation system, the crop is irrigated automatically while taking current soil moisture levels and weather forecasts into consideration. This paper attempted to implement the mentioned systems. The paper discusses the successes of the crop recommendation system, the automatic watering system, and the fertiliser recommendation system. In this paper, we report the results of simulations of the mentioned systems.

Dynamic Scene deblurring is a difficult low-level vision problem induced by camera shaking, object motion, and other causes. Since the blur movements of moving objects and the backdrop in dynamic scenarios are different, segmentation of the motion blur is necessary for deblurring each blur motion precisely. In contrast to this limiting assumption, we tackle the issue of deblurring broad dynamic images with many moving objects and camera motion in the current study. A multilayer convolutional neural network (CNN) based video image deblurring technique is suggested as a solution to the issue that inter-frame information and spatiotemporal are easily lost during restoration of a digital video/image that has become blurred. We use the multiscale CNN-built adaptive Laplacian regularisation term to the problem of restoring images from videos. First, we provide a new restoration model by mixing several regularises, in particular by combining NLM regularise with the multiscale CNN, to extract redundant information from video images' self-similarity. We utilised a basic gradient descent technique to solve the model for restoring images from videos. The experimental findings demonstrate the effective deblurring effect and some noise resilience of our approach.

Image pseudo colorization is the process of adding RGB colours to grayscale images to make them more appealing. Deep learning technology has made progress in the field of automatic colouring. In general, we divide automatic colouring methods into three groups based on where the colour information comes from: colouring based on what you already know and on reference pictures. The colouring method can meet the needs of most users, but there are some drawbacks. For example, users can't colour different reference graphs for the different things in a picture. In order to solve this problem by recognising several objects and background regions in a picture and combine the final colouring results, the proposed method uses the deep learning approach that regional mixed colours be used more and the method be mastered by using deep learning. Qualitative results (visual perception) validate the effectiveness of pseudocolorisation which split into foreground colour based on a reference picture and background colour based on prior knowledge. Quantitative results such as Structural Similarity (SSIM), Peak Signal to Noise Ratio (PSNR), Image Matching Error and Entropy validates the effectiveness of strong edge strength, visually appealing quality and retention of maximum information without disturbing quality of image.

In recent years, object detection has emerged as a crucial component of many popular consumer applications, including video surveillance and security systems, mobile text recognition, etc. The potential for autonomous vehicles (AVs) to increase driver satisfaction and decrease fatalities and injuries in traffic accidents has attracted a lot of interest in recent years. Object detection is critical to autonomous driving infrastructure. Autonomous automobiles need precise environmental interpretation to drive safely. Locating and identifying these things in real time is a significant challenge, but deep learning-based object detectors play a crucial role in this endeavour. In this paper, a prototype of the autonomous vehicle controlled by a microcontroller for fire object detection is proposed. Localization of the fire object in a picture using the deep learning model is performed using the live video feed from the camera installed on the prototype remote-operated car. The gadget may also pick up on specifics and alert the user. Experimental findings showed that the suggested prototype with deep learning architecture recognised and alerted devastating fires with high speed and accuracy in diverse weather conditions—sunny or overcast, day or night.

Rainfall prediction has a major effect on human civilization and is one of the most difficult, unpredictable activities. Precise and accurate predictions will help to rising human and financial risks pro-actively. This work presents a current supervised learning models of machine learning to focused on the Rainfall Prediction. Rainfall is also a significant issue in the planet because it impacts any single aspects that relies on the human being. Unpredictable and reliable estimation of rainfall is a challenging job today. In this work, gives a maximum outcome and a stronger forecast for rainfall using logistic regression and support Vector Machine (SVM) classifier for better prediction.

White blood cells (WBCs) are cells that is key factor of the immune systems which is help to our body fight off contagions and other diseases. In order to enhance the diagnosis of various diseases in medical field by using image processing techniques from the blood cells. In that, Leukemia is associated with one type of cancer of the blood and bone marrow. It is look like spongy tissue inside the bones where blood cells are made. In this paper, a fully connected. Convolution neural network is used to segmented and classification of blood cell microscope WBC images for healthy and unhealthy conditions. The performance of the classifier was analyzed. The accuracy sensitivity specificity and pression are 96.84%, 96.26%,97.35% and 96.39% respectively.

Person identification which has been widely used in computer vision and one of the difficult applications employed in a variety of fields such autonomous vehicles, robots, security tracking, aiding visually impaired people, etc. As deep learning quickly developed, numerous algorithms strengthened the link between video analysis and visual comprehension. The goal of all these algorithms, regardless of how their network architectures operate, is to find many people inside a complicated image. The freedom of movement in an unknown environment is restricted by the absence of vision impairment, thus it is crucial to use modern technologies and teach them to assist blind people whenever necessary. We provide a system that will identify all potential daily multiple people and then prompt a voice to inform the user about both nearby and distant people.

This paper presents a software-engineered approach using a classification algorithm for the classification of liver disease. The ILPD dataset is used for the proposed work. Different attributes of liver patient records such as direct bilirubin, age, sex, total bilirubin, alphos, albumin, sgpt, globulin ratio, sgot are used to classify liver disease. The proposed Convolution Neural Network classification technique shows an accuracy of 67% and a precision of 71%. Various classification algorithms such as CNN, RNN, ANN, and logistic regression are executed on the liver patient dataset and their accuracy is determined.

Medical imaging like MRI and CT scan images are crucial for accurately diagnosing human brain disease. The traditional method for tumour analysis relies on the radiologist or physician visually inspecting the specimen, which can result in some incorrect classifications when a large number of MRI pictures need to be processed. An automated intelligent classification system is suggested that requires picture categorization in order to reduce human mistake rates. One of the illnesses that kills the majority of individuals worldwide is the brain tumour. If the tumour is accurately anticipated at an early stage, the likelihood that someone would survive can be increased. The human brain is studied using the magnetic resonance imaging (MRI) method to identify illnesses. In this project, Support Vector Machines (SVM)-based classification approaches are suggested and implemented to classify brain images; DWT will extract features from MRI images. The primary goal of this research is to provide a superior result, which is higher accuracy and reduced error rates for SVM-based MRI brain tumour prediction.