In this paper, we describe a simulation-based framework for detecting potholes using radar in addition to assessing optimal routes over roads of varying surface materials using machine learning. This framework creates B-scan radar signals that simulate the ground plane with different terrains and materials. After extracting the signal features, we train supervised learning models to classify the potholes. In addition to this, we apply a feature-based scoring approach to score terrain segments to filter and obtain the desirable driving route.

In the contemporary digital landscape, efficient indoor navigation has become imperative for large, intricate spaces like shopping malls, airports, hospitals, and universities. Traditional GPS-based methods often prove inadequate in these indoor settings due to poor reception and a lack of fine-grained location accuracy. To address these challenges, the work employs WiFi signals, pervasive in modern buildings, and harnesses their ubiquity and reliability for sophisticated indoor navigation. The team developed a Python package for collecting WiFi signal details, implementing robust preprocessing to filter out private networks, and storing the data in a database. The core of the system lies in a machine learning ensemble consisting of Gaussian Bayes classifier, decision tree, and K-nearest neighbors (KNN). This ensemble utilizes WiFi signal strengths to predict users’ real-time positions within large indoor spaces. The system aims to enhance user experiences and simplify wayfinding in complex indoor environments.

Medicinal plants have a huge significance today as it is the root resource to treat several ailments and medical disorders that do not find a satisfactory cure using allopathy. The manual and physical identification of such plants requires experience and expertise and it can be a gradual and cumbersome task, in addition to resulting in inaccurate decisions. In an attempt to automate this decision making, a data set of leaves of 10 medicinal plant species were prepared and the Gray-level Co-occurence Matrix (GLCM) features were extracted. From our earlier implementations of the several machine learning algorithms, the k-nearest neighbor (KNN) algorithm was identified as best suited for classification using MATLAB 2019a and has been adopted here. Based on the confusion matrices for various k values, the optimum k was selected and the hardware implementation was implemented for the classifier on FPGA in this work. An accuracy of 88.3% was obtained for the classifier from the confusion chart. A custom intellectual property (IP) for the design is created and its verification is done on the ZedBoard for three classes of plants.

With the advancement in deep learning, the consolidation of text classification and image processing has agitated enormous scrutiny in the past years. Captioning images are advanced research mainly in the field of computer vision. Identification of relevant items, their properties, and their connection in images is required for image captioning. This image captioning is a complex and challenging task, several developments are done in this field by researchers. But now, Technology can develop a model that can bring out the closest text or captions explaining an image because of the advancement in deep learning methodologies and the enormous amounts of data that is available. In the beginning, it was contemplated impractical that a computer could characterize an image.

With the increased usage of online social media platforms, there has been a sharp hike in toxic comments. Toxicity must be reduced. Classification of toxicity in comments has been an effective research field with various newly proposed approaches. This research and analysis provide a novel usage of the Natural Language Processing approach to classify the type of toxicity in comments. This analysis intends to interpret the type of comment and determine the various types of toxic classes such as obscene, identity hate, threat, toxic, insult, severe toxic. The input to our algorithm is comments from online platforms like toxic or non-toxic. Our model aims to predict the toxicity class. This project intends to analyze in phases. In Phase I, the objective is to evaluate the toxicity in comments by giving data through various techniques like TDIDF, spacy that helps data to perceive how every word in a comment is classified into a particular category of toxic class. Here, Algorithm will take comments from test data and predict the type of toxicity for test data like a toxic, threat, and so on. In Phase II, Data is analyzed to organize the comments into toxic and non-toxic categories. This promotes us to perceive the particular comment is toxic or not.

An analytical polynomial expression, for accurate and computationally efficient frequency estimation of a single real sinusoid under Additive White Gaussian Noise (AWGN), is derived and proposed in this paper. The method, which can be easily adapted for real time frequency estimation, is based on transforming the frequency estimation problem as the solution of a fourth order (quartic) expressed as powers of a trigonometric function containing the unknown frequency. The coefficients of the quartic polynomial can be found using the complex magnitudes of three Discrete Fourier Transform (DFT) bins, centered at the maximum magnitude value of the DFT coefficients. Simulated results illustrate that, the performance of the proposed estimator has a mean squared-error (MSE) performance which is very close to the Cramer Rao Lower Bound (CRLB) for high signal-to-noise ratio (SNR) region, as well as close to previously published estimators in the low SNR region.

With the ongoing pandemics alike COVID-19 patronage such as stock markets, textiles become subsided. Stock market prognostication is the appearance of seeking to circumscribe the anticipated marketability of stocks and different financial apparatuses patronized on an exchange. Prophesying how the retail valuation will execute is one of the ultimate back-breaking contrivances to the predicament. Retail prediction is significant for quantitative interpreters and investment organizations. Retail valuation prophecy is predominant for merit expenditure in the retail market. Analyzing the valuation interrelationship of duplet assets for the anticipated period of time is essential in portfolio optimization. The recommended explication is catholic as it comprises preprocessing of the retail advertise dataset, application of various exploratory analysis procedures, collaboration beside custom-built algorithms for retail valuation bias prognosis. In this case, Facebook Prophet and Arima models are used in forecasting the retail valuation of future stocks that are used to analyze future values of stock markets and how it varied from previous stock markets. With the circumstantial architecture and consideration of conjecture premises and data pre-processing techniques, this effort commits to retail estimate analysis.

Spectrum spreading is a communication technique by which a signal produced by a bandwidth given is intentionally spread over the frequency domain, resulting in a broader signal bandwidth. This helps to give channels immunity by not allowing interference or disruption of any sort, thus ensuring security in communication systems. The Inefficient spectrum use contributes even to the invention of new methods which enable users not allowed to utilize the radio spectrum each time a hole in the spectrum or free space is available. The technology which pinpoints the presence of a licensed radio spectrum user using complex spectrum access techniques is called cognitive radio. The paper’s key contribution is the study and comparison of various spreading techniques implemented in a signal for transmission and further implements the technique for energy detection of spectrum sensing.

In today's world as vehicles are increasing on the roads, reducing road accidents is a major concern. Neglecting the rules while driving is a major cause for accidents. In this scenario, this paper proposes a suitable method to reduce the road accidents to a great extend due to over speed. Unlike the existing methods, which controls the speed of the vehicle at a constant rate, this work proposes an integrated approach to effectively control the speed of the vehicle at varying rate based on the maximum permissible speed limit of the protected or restricted zones such as hospitals, schools and crowded areas. A passive RFID integrated with GPS locator in the vehicle provides a promising solution to control the speed of the vehicle based on the allowable speed limit of the particular location. A prototype resembling an electronically controlled road vehicle with customized speed limits is used for our work. RFID reader inside the vehicle captures the unique encoded signal from the RFID tag placed at the speed limit zone and automatically controls the speed of the vehicle, when the vehicle approaches the proximity of RFID tag. The work also simulates speed control through GPS data which overcomes the limitation of RFID, and also proposes a solution to enhance the security of RFID tags.

This work focuses on the wireless channel characterization for agriculture applications using a simulation based approach. Wireless channel modeling is a challenging problem for near ground communications primarily due to significant attenuation owing to the proximity to the surface of the earth. In this paper, we estimate the wireless channel coefficients and compute the RMSE to ascertain the accuracy. Using the corrected channel parameters, we characterize the BER of the wireless channel in AWGN for a BPSK communication system using a simulation based approach. We use Software Defined Radio (SDR) with GNU Radio, which is an open-source software development tool for generation of baseband signals, defining modulation formats and RF system parameters. Our results show that the corrected channel parameters from our system yield simulated BER values that are very close to the theoretical values.

Ever-expanding utilization of wireless applications is applying pressure on the restricted,lacking, and expensive authorized range of frequencies. As a matter of fact, on account of allotment of fixed range, there is a high risk of several ranges of frequencies going underutilized. Spectrum sensing can be utilized for the proficient and successful utilization of the radio range. It recognizes the unused spectrum channels in cognitive radio systems. In this paper the vacancy of wireless channels is determined using peak detection method and effective utilization and allocation of the frequency bands is implemented using CSMA/CA with Backoff algorithm calculation.

The inefficient utilization of the radio spectrum leads to the invention of new techniques that help those users with no access or license to use the spectrum, whenever spectrum hole or free space is available. Cognitive radio is such a technology which finds the existence of a licensed user in the radio spectrum using dynamic spectrum access techniques. This paper discusses the comparison of Energy Detection and Matched Filter based spectrum sensing techniques in Matlab and Verilog. Implementation and analysis of Energy Detection and Matched Filter detection techniques using Verilog has also been performed and the results have been presented.

Ayurvedic medicinal plants are very important since it is one of the key sources of medicine. It can cure various diseases such as Cardiac disorders, Respiratory diseases, Fertility issues etc. So a precise identification of medicinal plant is crucial for proper treatment. Manual recognition can be imprecise and also will be time consuming. In order to stay away from these issues an automatic recognition for medicinal plants are preferred. The features are extracted from images of plant leaves and then classified. The features considered are the shape, textural and colour features. Then some machine learning classification techniques such as KNN and SVM are used for classification and a comparison is made among their performances. All the simulations are done in MATLAB R2019a.

In this era, social media has become a platform for education, e-commerce, job seeking and many more. This work intends to create a digital portfolio which helps freshers to upgrade their skillsets, create their resume according to the company profile and as per the company requirements, interacting with people of similar interests and giving a chance to showcase their skills on the current platform. Students will be rated according to their performances and will be awarded badges. This platform will not only help students to work on themselves but also give confidence to them while entering into the corporate world.

Differential privacy is a method adopted to check for any privacy breach that occurs during communication for the exchange of confidential information. Here, in this work, differential privacy is being implemented in the context of vehicular ad hoc networks (VANETs). In this paper, we implement the concept of differential privacy using the Diffie–Hellman key exchange algorithm and the advanced encryption standards (AES) algorithms that are very powerful in terms of their performance. Algorithms like Laplace and Gaussian algorithms, which are currently the most commonly implemented algorithms, have been used for verification. The algorithms were analyzed by considering a situation where an initial location and final location have been defined and these have been encrypted using the mentioned algorithms and the privacy has been preserved.

This paper discusses the implementation of a novel technique of encoding data bits using the concept of bit stuffing in addition to the conventional methods of source coding. This technique can be applied to any of the existing methods of source encoding under controlled conditions. In particular, the method is very efficient when the encoded bits have more number of ones or zeros than a predefined threshold, at any point of time and in any part of the stream. Usually, bit stuffing is a common method used for data compression in data communication layers to reduce the bandwidth. In this paper, we have attempted to incorporate bit stuffing in various encoding schemes and have compared the improvement in performance with and without bit stuffing. The software used for simulation is MATLAB. The primary motivation of this work is to determine the maximum amount of bandwidth savings that can be achieved due to bit stuffing for a random series of alphabets.

The way reviews are written can affect the consumer's perception of information helpfulness.Two people may convey the same information in two different ways and hence changing the way they are perceived by others. The major influences are in message content and descriptive features. The proposed paper exhibits a study that uses reviews from Amazon India. The intention behind the research is to identify the change in the helpfulness of review with the variation in message content and the descriptive features. This could further help predict the helpfulness of review if the descriptive and message content is known. The methodology in the paper incorporates a regression model which includes the review rating, review length, review valence, number of First-person singular pronouns, number of First-person plural pronouns, number of second-person plural pronouns, number of third-person plural pronouns and affect. These are the descriptive features and the message content considered. From the study, it was clear that review helpfulness is influenced by the message content and descriptive features. This may not immediately contribute to short-term sales performance but will increase customer satisfaction and lead to long-term firm value.

To enhance the quality of the Intelligent Transportation system an effective vehicle communication and protection is needed. This can be achieved by developing a smart automatic tracking device utilizing GPS, GSM module and reduction in the power consumption of this system is achieved with the help of LDR. When the LDR sensors detect the visual light from the headlights of a vehicle it enables the GPS. A GSM modem is used to send the coordinates of the vehicle provided by the GPS to the end-users phone using the mobile network. A low power micro controller interfaced with LDR sensors can be used as a prototype model for efficient power consumption of vehicle tracking and will make the system cost-effective. This tracking system can be built with the help of hardware components such as U-Blox NEO-7M GPS receiver module, SIM900A GSM module, Arduino Uno and LDR. In this paper, we showed the wireless modelling of the GSM module and also evaluated the effective power of the common vehicle tracking system and the intended system and the results showed that the aimed system was less power consuming. Thus, the system proposed is proved to be efficient.

This paper mainly focuses on implementing a new approach rather than using the conventional method of generating an intermediate frequency in a mixer. The ordinary mixer is compared with a switching mixer through several techniques in order to strengthen the point that switching mixers have a better performance. Theoretically, we know that switching mixers are more efficient but here we also give it a practical justification through this paper. Software such as MATLAB and Proteus has been used for the same. The main objective of this research work is to determine the maximum amount of noise that can be removed to obtain a good reconstruction of the input signal.

With the ongoing COVID-19 pandemic, businesses and organizations have acclimated to unconventional and different working ways and patterns, like working from home, working with limited employees at office premises. With the new normal here to stay for the recent future, employees have also adapted to different working environments and customs, which has also resulted in psychological stress and lethargy for many, as they adapt to the new normal and adjust their personal and professional lives. In this work, data visualization techniques and machine learning algorithms have been used to predict employees stress levels. Based on data, we can develop a model that will assist to predict if an employee is likely to be under stress or not. Here, the XGB classifier is used for the prediction process and the results are presented showing that the method facilitates getting a more reliable model performance. After performing interpretation utilizing XGB classifier it is determined that working hours, workload, age, and, role ambiguity have a significant and negative influence on employee performance. The additional factors do not hold much significance when associated to the above discussed. Therefore, It is concluded that concluded that increasing working hours, role ambiguity, the workload would diminish employee representation in all perspectives.