International Journal of Sensors Wireless Communications and Control - Volume 12, Issue 1, 2022

Abstract: Feature ranking can have a severe impact on the feature selection problem. Feature ranking methods refer to the structure of features that can accept the designed data and have a positive effect on the quality of features. Moreover, accessing useful features helps in reducing cost and improving the performance of a feature ranking algorithm. There are numerous methods for ranking the features that are available in the literature. The developments of the past 20 years in the domain of knowledge research have been explored and presented in terms of relevance and various known concepts of feature ranking problems. The latest developments are mostly based on the evolutionary approaches which broadly include variations in ranking, mutual information, entropy, mutation, parent selection, genetic algorithm, etc. For a variety of algorithms based on differential evolution, it has been observed that although the suitability of the mutation operator is extremely important for feature selection yet other operators can also be considered. Therefore, the special emphasis of various algorithms is observing and reviewing the algorithms and finding new research directions: The general approach is to review a rigorous collection of articles first and then obtain the most accurate and relevant data followed by the narrow down of research questions. Research is based on the research questions. These are reviewed in four phases: designing the review, conducting the review, analyzing, and then writing the review. Threats to validity is also considered with research questions. In this paper, many feature ranking methods have been discussed to find further direction in feature ranking and differential evolution. A literature survey is performed on 93 papers to find out the performance in relevance, redundancy, correlation with differential evolution. Discussion is suitable for cascading the direction of differential evolution in integration with information-theoretic, entropy, and sparse learning. As differential evolution is multiobjective in nature so it can be incorporated with feature ranking problems. The survey is being conducted on many renowned journals and is verified with their research questions. Conclusions of the survey prove to be essential role models for multiple directions of a research entity. In this paper, a comprehensive view on the current-day understanding of the underlying mechanisms describing the impact of algorithms and review current and future research directions for use of evolutionary computations, mutual information, and entropy in the field of feature ranking is complemented by the list of promising research directions. However, there are no strict rules for the pros and cons of alternative algorithms.

Background: In India, thousands of languages or dialects are in use. Most Indian dialects are low asset dialects. A well-performing Automatic Speech Recognition (ASR) system for Indian languages is unavailable due to a lack of resources. Hindi is one of them as large vocabulary Hindi speech datasets are not freely available. We have only a few hours of transcribed Hindi speech dataset. There is a lot of time and money involved in creating a well-transcribed speech dataset. Thus, developing a realtime ASR system with a few hours of the training dataset is the most challenging task. The different techniques like data augmentation, semi-supervised training, multilingual architecture, and transfer learning, have been reported in the past to tackle the fewer speech data issues. In this paper, we examine the effect of multilingual acoustic modeling in ASR systems for the Hindi language. Objective: This article’s objective is to develop a high accuracy Hindi ASR system with a reasonable computational load and high accuracy using a few hours of training data. Methods: To achieve this goal, we used Multilingual training with Time Delay Neural Network- Bidirectional Long Short Term Memory (TDNN-BLSTM) acoustic modeling. Multilingual acoustic modeling has significantly improved the ASR system's performance for low and limited resource languages. The common practice is to train the acoustic model by merging data from similar languages. In this work, we use three Indian languages, namely Hindi, Marathi, and Bengali. Hindi with 2.5 hours of training data and Marathi with 5.5 hours of training data and Bengali with 28.5 hours of transcribed data, was used in this work to train the proposed model. Results: The Kaldi toolkit was used to perform all the experiments. The paper is investigated three main points. First, we present the monolingual ASR system using various Neural Network (NN) based acoustic models. Second, we show that Recurrent Neural Network (RNN) language modeling helps to improve the ASR performance further. Finally, we show that a multilingual ASR system significantly reduces the Word Error Rate (WER) (absolute 2% WER reduction for Hindi and 3% for the Marathi language). In all three languages, the proposed TDNN-BLSTM-A multilingual acoustic models help to get the lowest WER. Conclusion: The multilingual hybrid TDNN-BLSTM-A architecture shows a 13.67% relative improvement over the monolingual Hindi ASR system. The best WER of 8.65% was recorded for Hindi ASR. For Marathi and Bengali, the proposed TDNN-BLSTM-A acoustic model reports the best WER of 30.40% and 10.85%.

Aim: The method of utilization of IoT and other evolving techniques in medical equipment design field is discussed in the present paper. A remotely managed interface equipped in a wheelchair cum bed is embedded for elderly or physically challenged people. With the help of a camera embedded in the proposed solution, a real-time remote monitoring of the patient is achieved using an android application on the concerned person. For achieving the above mentioned purpose, the use of linear actuators has been done. This paper further aims to explore the hidden potentials of the merger of all these fields to benefit the end users. Objectives: Remote monitoring of health of the patient through a cloud-based android application. Automatic adjustment of the wheelchair into bed and vice-versa. Automatic stool passing chamber facility is available under the proposed model. Injuries during transportation of the patient from one chair to another (or chair to bed) have been limited in our designed model. Methods: The basic mechanism of proposed wheelchair has been designed using the computer-aided design software. The basic methodology adopted for development of prototype & subsequent “user review analysis” is displayed. The CAD Model of the wheelchair cum stretcher was designed using the “Solid works solid modelling techniques”. The basic structure has been designed with several modifications when compared to the conventional wheelchair design. The computer made design was then utilized for final fabrication of the prototype. The prototype was tested for endurance, load bearing capacity and customer comfort during various phases of development. The feedbacks of several subjects were recorded for future utilization in improved design & fabrication. Results: The proposed model is of utmost importance as the number of critical patients like accident cases, critical pre and post-surgery cases is increasing day by day. Sometimes these patients need intime medication during transition in ambulance while they are picked up from houses and referred to nearby big hospitals. During transition or in hospital, critical patients can be handled efficiently by a specialist doctor through his/her smart phone applications. It also optimizes the services of specialist doctors as we can find the shortage of specialists in Indian hospitals. In a nutshell, this WheelChair system can be moved anywhere due to its portability. Following are the most highlighted features: 1. Authorized relatives and Doctors can see and interact with the patient remotely at any time on his/her smart phone. 2. Authorized relatives and Doctors can see the Vital Sign of patient like BP, ECG, and Pulse etc. at any time through Smart Phone. 3. Doctor can instruct the caretaker to release the emergency drugs through Infusion Pump. 4. Doctor can plan the exceptions, drug infusion, alarm, etc. 5. System is portable and can easily be shifted to ambulance. All transmissions are wireless, so there is no hassle of wires and connectivity. Conclusion: The presented work is limited to the design and fabrication of a new model of wheelchair, which works as a stretcher and has locomotive capabilities. The key feature of the design is its versatility and adaptability to various working conditions. The feedback obtained from various subjects during the testing of wheelchair shows their confidence and a fair degree of comfort which they felt while using the wheelchair. The easy and user-friendly use of the android application helps to monitor the health of the patient. The smartphone camera helped to achieve this data. The analysis of the data can be done in the cloud-based station. The linear actuator has proved to be the low cost and highly reliable equipment to propel the wheelchair.

Background: In developing countries, the healthcare system is facing numerous challenges. One of the major challenges faced by the healthcare system is that the healthcare service providers are meager and geographically far from the densely populated area. Objective: To overcome the above challenge, the present research work proposes SHRMS (Smart Heart Rate Monitoring System) which provides the ad-hoc services to the patients who are in the transit mode in the emergency vehicle. Methods: A pulse sensor is attached to the patient’s fingertip to fetch the heart rate of the patient. The patient’s data is further transmitted to the microcontroller which in turn transmits the data to the Thing- Speakcloud service. Result: SHRMS provides the real-time monitoring of the patient and helps to provide emergency aid as per the patient’s current situation. Conclusion: This device is beneficial for developing countries where the healthcare service providers are very less and geographically scattered.

Aims: Recent advancements in sensing technology and wireless communications have accelerated the development of the Internet of Things (IoT) which promotes the usage of wearable sensors. An emerging trend is to develop self-sustainable wearable devices, thus eliminating the necessity of the user to carry bulky batteries. In this work, the development of a flexible piezoelectric energy harvester that is capable of harvesting energy from low frequency vibrations is presented. The target application of this energy harvester is for usage in smart shoes. Objectives: The objective of this research is to design, fabricate and test an energy harvester on PET substrate using Aluminum Zinc Oxide as its piezoelectric layer. Methods: The energy harvester was designed as a cantilever structure using PET/AZO/Ag layers in d33 mode which can generate large output voltages with small displacements. The electrodes were designed as an interdigitated structure in which two significant design parameters were chosen, namely the effect of gap between electrodes, g and number of inter-digital electrodes (IDE) pairs, N to the output voltage and resonant frequency. Results: The sputtered AZO on PET showed c-axis orientation at 002 peak with 2 values of 34.45° which indicates piezoelectric behavior. The silver IDE pairs were screen-printed on the AZO thin film. Functionality of the device as an energy harvester was demonstrated by testing it, using a shaker. The energy harvester was capable of generating 0.867 Vrms output voltage when actuated at 49.6 Hz vibrations. Conclusion: This indicates that the AZO thin films with printed silver electrodes can be used as flexible, d33 energy harvesters.

Background: In most of the researches, the autonomous driving problem is either solved with policy gradients or DQN. In this paper, we tried to eliminate the problem of policy gradients which is high variance and an overestimation of values in DQN. We used DDQN as it has low variance, and it solves the problem of overestimation in DQN. Aim: The main aim of this paper is to propose a framework for an autonomous driving model that takes in raw sensor information as input data and predicts actions as output from the model which could then be used for simulating the car. Objective: The main objective of this paper is to use DDQN and Discretization technique to solve the autonomous driving problem and get better results even with a continuous action space. Methods: To solve the bridge between self-driving cars and reinforcement learning we used Double Deep Q-Networks as this could help to prevent the overestimation of values by decoupling the selection from the evaluation. Also, to solve the problem of continuous action space we used the discretization technique in which variables are grouped into bins and each bin is assigned a value in such a way that the relationship between the bins is preserved. Result: The experimental results showed improved performance of the agent. The agent was tested for different conditions like curve roads and traffic, which showed the agent can drive at different conditions as well. We also illustrated how DDQN performed well over policy gradients just by adding a simple discretization technique to make the action space discrete and overcoming the issue of overestimation of q-values. Conclusion: The gym environment and reward function were designed for DDQN to work. We have also used CARLA as a virtual simulator for training purposes. Finally, we have demonstrated that our agent could perform well in different cases and conditions. As a reminder note, we can improve our agent to also work for following traffic light rules and other road safety measures.

Background: Cloud computing is a service that has undergone accelerated growth in the field of information technology in recent years. Privacy and security are challenging issues for cloud users and providers. Objective: This work aims at ensuring secured validation of user and protecting data during transmission for users in a public IoT-cloud environment. Existing security measures have failed by their single level of security, adaptability for a large amount of data and reliability, to overcome these issues and to achieve a better solution for vulnerable data. Methods: The suggested method utilizes secure transmission in the cloud using key policy attributebased encryption (KP-ABE). Initially, user authentication is verified. Then the user data is encrypted with the help of KP-ABE algorithm. Finally, data validation and privacy preservation are done by Burrows-Abadi-Needham (BAN) logic. This verified, shows that the proposed encryption is correct, secure and efficient to prevent unauthorized access and prevents data leakage with fewer chances of data/identity theft of a user as this analysis performed by KP-ABE is based on access control approach. Results: Here, the method attains a maximum of 88.35% of validation accuracy with a minimum 8.78ms encryption time, which is better when compared to the existing methods. The proposed mechanism is done by MATLAB. The performance of the implemented method is calculated based on the time of encryption and decryption, execution time and validation accuracy. Conclusion: Thus, the proposed approach attains high IoT-cloud data security and increases the speed for validation and transmission with high accuracy and is used for cyber data science processing.

Background: Cloud Computing can process and utilize efficient resources within a metered premise. This feature is a significant research problem, giving great Quality-of-Services (QoS) to the clients in cloud. Objective: The objective of this study is to confirm QoS with minimum resource utilization time and costs, cloud service providers ought to receive self-versatile resource provisioning at each level. Various guidelines and model-based methodologies have been proposed for the management of resources in cloud services. Methods: In this research article, resource allocation is done using the Salp Swarm Algorithm (SSA), which combines various VMs on a constrained Data Center with SLA and QoS factors. Results: Different existing optimization algorithms are available such as First Fit, Greedy Crow Search (GCS) and Hybrid Crow Search algorithm (TSPCS). The combination of the Travelling Salesman Problem (TSP) and Crow Search Algorithm (CSA) is more efficient than the Fist Fit, GCS, and TSPCS in terms of the parameters such as resource utilization and response time. It is clearly shown that a user’s request takes minimum time and maximum QoS when employing the SSA algorithm in cloud computing. Conclusion: The proposed mechanism is simulated on Cloudsim Simulator. The simulation results show less migration time that improve the QoS and minimizes the energy consumption in a cloud and IoT environment.