Advances in VLSI, Signal Processing, Power Electronics, IoT, Communication and Embedded Systems

The automotive industry is seen to be making a transformation from manual to semi-autonomous to fully autonomous vehicles. Advanced Driver Assistance System (ADAS) forms a major building block for the next generation of highly Autonomous Vehicles. ADAS is among the fastest-growing segments in automotive electronics and includes large number of applications like Antilock braking system, automatic speed control, Parking assistance system, lane keeping assist etc. Parking assistant system supports several functionalities like Parking line detection, object detection, pedestrian classification, parking support functions etc. These functions are mainly implemented using Surround View Camera System (SVCS) which provides real-time view of the area surrounding the car. Parking Shield System (PSS) is one such parking support function which works with the help of SVCS. PSS is used as an aid to identify the damage caused to the parked vehicle. This work proposes a prototype development for PSS using Raspberry Pi. Acceleration sensor is used to detect the movement of the vehicle and once the vehicle movement is detected, LEDs glow (indicating opening of flaps and mirrors). Immediately, USB Camera connected to Raspberry Pi captures the image and is displayed in VNC Viewer window.

In our technologically advanced period, where urbanization is fast rising, waste accumulation is also increasing steadily. Making cities smart has become the need of the hour in order to tackle this problem. Though literature shows several methods for the implementation of smart cities, the problem still remains. One general issue while designing a system for waste management is the segregation of different types of waste. Manual segregation is still being followed in most of the cities. This paper proposes a smart waste management system that can differentiate and store biodegradable and non-biodegradable waste separately using a set of sensors such as ultrasonic, IR, moisture and inductive sensors. The proposed system also does segregation of the non-biodegradable waste as plastic, wood, and metal.

The Residue Number System (RNS) is a non-weighted number system. Because of its inherent parallelism, it has been extensively studied and used in Digital Signal Processing (DSP) systems. A key arithmetic operation in residue-based real-time computing system is modulo multiplication. For small moduli, ROM-based structures are better at realizing multipliers. Implementations with arithmetic components are more for medium and large moduli due to the exponential growth of ROM sizes. The new modular multiplier introduced in this paper is capable of easily handling medium and large moduli. The multiplier unit is proposed in this paper using shift and add, followed by the modulo operation. The implementation results show that our proposed design outperforms existing architectures in terms of area and power consumption when using the TSMC-180 nm CMOS Technology. When compared to existing works, our proposed multiplier saves 86% to 93% of area and 65% to 83% of power. The proposed multiplier improves the existing DCT architecture by 18% in terms of area.

In order to improve the quality of devices before they are delivered to customers, VLSI testing processes have been created to detect damaged devices using automated test equipment. To characterize a chip and identify manufacturing issues that must be resolved before full production, a variety of devices must be tested. This testing methodology for services includes a time-consuming and error-prone planning process. Through the use of a queuing strategy, the performance level of the process used in these testing approaches is investigated and examined. In order to utilize the concept of queuing, the process of VLSI testing procedures is transformed into a mathematical problem. The supplemental variable approach is then used to resolve this mathematical queuing problem. A simulation process is used to create performance measurements for this type of queuing system of VLSI testing techniques using the MATLAB and R tools. Results are as anticipated. A specific discussion of how to prevent or minimize disruptions so that the system can run effectively is included in the analysis section.

It is vital to update tools and methodologies in pace with technological innovation in order to handle the problems posed by the changing verification environment. The authors propose a verification of AHB2 APB bridge protocol using Universal Verification Methodology (UVM). It is the intention of this study to discuss the benefits of using the Universal Verification Methodology for AHB2APB verification. This work focuses on creating a standard AHB2APB Bridge protocol architecture that is efficient and on enhancing the authentication environment utilizing a System Verilog UVM implementation. An automated authentication platform developed by UVM will produce an AHB2APB Bridge debugging test for any given DUT. UVM-based performance verification adds randomized test scenarios to ensure high-performance by covering all potential cases. On the other hand, Verilog cannot be used to examine the active cover model in a typical verification location. System Verilog is used for coding, while Questasim is the simulation tool of choice. The code coverage for RTL design is acquired, and 98.81 percent of the code coverage and 100 percent of the functional coverage are recovered.

SOC's have multiple blocks integrated on a single chip, to interact between these blocks, on-chip bus architecture is required. ARM's AMBA (Advanced Microcontroller Bus Architecture) is one of the extensively used on-chip architecture used in the VLSI industry. AHB (Advanced High-Performance Bus) is an AMBA Bus protocol with high-performance, supports multiple masters and multiple slaves and has wider bandwidth; as a result of these features, AHB is generally used in SOC's. Previously, AHB design was verified using either Verilog or system Verilog, where the time spent on verification was more than the time spent on design. A more effective verification methodology, like UVM (Universal Verification Methodology), must be adopted to reduce the average time spent in verification and increase efficiency. An AHB protocol with one master and three slaves is designed in ModelSim using Verilog and verified in QuestaSim using UVM, and a coverage report is generated. QuestaSim and ModelSim are the EDA tools developed by Mentor Graphics for design and verification purposes.

People with impairments struggle harder in both daily life and academic settings. Students who are physically or visually challenged find it challenging to navigate an increasingly complicated world. Understanding the requirements of students with disabilities and the ways in which technology might be used to assist them is crucial. They can overcome obstacles and achieve better outcomes in both education and daily life with the aid of assistive technologies, which comprise both software and hardware. In this paper, a solution for physically disabled persons, a voice-activated plotter, which recognizes the speech and writes down the spoken content on the paper at a speed closer to average human writing speed is developed. This work on a speed of 10 WPM, which is closer to the Average human speed i.e. 13 WPM is archived.

Classification is one of the most commonly employed techniques in remote sensing studies. The thematic maps generated by the RS data classification are employed in a diversity of socio-economic applications. With the advent of space technology, new diverse and advanced RS data are generated at a very high rate. In this study, we implement two decision tree based classification techniques; classification and regression trees (CART) and iterative dichotomizer (ID3), in classifying heterogenous multispectral RS data. The study uses two Landsat-8 study areas; the North Canara District boundary and Kumta Taluk boundary, in Karnataka India. We selected seven level-1 and level-2 LULC classes from Anderson’s (Anderson, A land use and land cover classification system for use with remote sensor data, vol. 964. US Government Printing Office, 1976) classification system for each study area. Class separability is measured between each class pair using the Euclidean distance metric and severely overlapping classes are identified on each data. The paper also discusses different types of decision trees and their attribute selection measures. The results obtained indicate that CART and ID3 are excellent choices for separating severely overlapping spectral class pairs.

Persons with Sensorineural Hearing Loss (SNHL) experience a decrease in speech perception because of masking in a noisy environment. Spectral splitting of speech using time varying comb filters with complementary magnitude responses for binaural dichotic presentation is helpful in decreasing the effect of frequency masking of spectral components. Thus, there is an improvement in speech perception. FIR time varying comb filters with complementary magnitude responses of order 512 are designed using a frequency sampling method with 22 one-third octave bands ranging from 0 to 11 kHz. and the continuous shift in magnitude responses with time shifts selected below just a noticeable difference (JND), so that gap detection ability improves without offsetting the advantage of the spectral splitting technique. Filter performance is assessed by objective tests using PESQ (Perceptual Evaluation of Speech Quality), spectrographic analysis, and by subjective tests using Mean Opinion Score (MOS) for quality. Subjective test is investigated on normal hearing subjects by adding white noise at different SNRs. Test materials used for the evaluation are VC (Vowel Consonant) nonsense syllables and vowels. The results showed an improvement in the perception of processed speech in a noisy environment.

Coastal erosions usually cause drastic disasters in the ecosystems and the human lives in coastal zones. Change over time and identifying the location of the shoreline is the most important aspect of managing coastal areas and this requires frequent monitoring of the shoreline using satellite imageries over time. The effectiveness of filed-based study and image processing techniques for computing soil erosion or accretion rate are considered and the proposed method gives the precise results for soil erosion studies. In the proposed method, first principal component analysis (PCA) is applied, then image processing techniques are applied and finally, the model is developed using Machine Learning algorithms. Firstly PCA has been used for separating land and water bodies in satellite imageries, then image processing technique is used to compute soil erosion or accretion rate and finally, the model is developed using Machine Learning Algorithm. The proposed method is applied to satellite imageries, between 2014 to 2019, of the Mangaluru coastal region to attain erosion and accretion rate. The results showed that the erosion rate is comparatively high in Talapady, Someshwara, and Ullal Regions in the year 2014–2015. Also, for the years 2016–2018, there has been an erosion in these regions. Other regions of Mangaluru Coast are subjected to both erosion and accretion but there are no significant changes. As compared to field-based study and image processing technique-based soil erosion or accretion rate computation, the proposed method predicts the soil erosion or accretion for future years also. The model developed in the proposed study can process satellite imageries of several years and predicts soil erosion or accretion for future years. If the erosion rate is high then the government can take necessary action plans such as stopping sand mining, industrial and development activities and, planting more trees near the seashore.

Signal Processing is a primary and important task in most of the applications involving signal acquisition and digital processing. Signal conditioning can be thought of as a process of removing unwanted information and preserving the required feature of the signal. Denoising can be similarly correlated with signal conditioning in this regard. The presented work is gives an IIR filter architecture that is optimized for area and speed metric for an end application of ECG signal processing. The proposed IIR filter uses Canonic Signed Digits (CSD) for representing scaled version of filter coefficients and then the IIR filter is designed. It is observed that by using CSD-based IIR filter architecture gives an efficient approach to denoise the ECG signals, demonstrates the area reduction by 90%, and improves the speed of operation by 60% when compared with conventional design. The Matlab and Xilinx Vivado platform is used to validate the designed IIR filter and further the post-synthesis implementation details are obtained for Xilinx Virtex-7 series XC7VX485T-2FFG1761C FPGA, Verilog HDL coding is employed to develop the design and stimulus environment for the work.

In the present world, skin cancer is the most widely recognized reason for death among people. Skin malignancy is an unusual development of skin cells. Frequently created on the body part exposed to the sunlight; however, it can happen on any place on the body. The majority of the skin malignancy is treatable at the beginning phase. So an early and quick identification of skin disease can spare the patient's life. With the new innovation, early identification of skin malignancy is conceivable at the introductory stage utilizing picture handling. The skin cancer detection using image processing is based on the detection of skin cancer types at its earliest stage. There are many types of skin cancers found. It is difficult to identify the type of skin cancer at the earlier stage, manual identification can often be time consuming and inaccurate. Doctors are able to identify the symptoms of skin cancer but are unable to identify the type of skin cancer in the initial stage. So the doctors will wait until it gets blotted but by that time the disease will become out of control. So a software is developed to help the Skin Cancer Detection at its earliest stage by passing valid input images. So this chapter explains about a method to identify and classify the skin cancer using images using Convolutional Neural Network (CNN) algorithm. The accuracy obtained by using the proposed method is 89%.

We live in a period where machines have become a fundamental piece of our day-to-day existence. These machines that surround us largely depend on human assistance. To operate them, a human would nearly have to be functional. We only lose the capacity to engage with machines when these capabilities are hindered, possibly by a bodily condition or injury. This study picks the Universal Testing Machine as the machine to be operated to help physically challenged people run machinery. Additionally, it uses an Internet of Things architecture to monitor specific brain activities in the person’s brain, while controlling the machine.

Approximate computing in the era of high-speed multimedia applications increase in the demand of high-speed error-tolerant circuits. In this method, accuracy is compromised to achieve high performance. The main criterion is to reduce hardware complexity, power dissipation, and timing delay at the cost of accuracy. The main aim of this paper is to design and analyze two approximate compressors with minimum hardware complexity, delay, and power with reasonable accuracy, when compared with the existing compressors in literature. The proposed designs are implemented and verified using 90 nm Cadence NC-Verilog standard library cells. The parameters of interest in the approximate computing are Error Rate (ER), Error Distance (ED), and Accurate Output Count (AOC). The proposed compressors are used to implement 8 × 8, 16 × 16 unsigned multipliers, and signed 8 × 8 approximate multipliers. The application of the proposed design in image smoothing, multiplication, and LIF neuron modeling is discussed.

Particle swarm optimization (PSO) is among the prominent computing approaches that rely on population-based optimization. It is coupled to a swarm intelligence cluster and is used in global optimization challenges. Sliding Mode Control (SMC) is a first-order control approach which has a broad range of mechanical device applications. But due to its disadvantages such as chattering effect, a higher order control mechanism is necessary. Super-Twisting SMC (ST-SMC) is a second order control mechanism, has advantages like reduced chattering effect, and achieves convergence in time. In this review article, first a comprehensive review on PSO and its applications is performed. Later, ST-SMC is reviewed in detail and then optimization of SMC parameters using PSO for autonomous vehicle is discussed.

The Middle East and Dubai are specifically well known for their large number of Hypermarkets, Supermarkets, and Groceries, generating a huge demand for high quality consumables such as vegetables and fruits. Modern Retailing will benefit much from offering fruit and vegetables sorted by quality (based on physical characteristics such as weight, size, color, shape, smell, etc.)—an essential step in post-harvest management for offering quality agricultural produce to the consumer. The proposed work focuses on sorting fruits and vegetables based on Color and Shape. It uses a Convolutional Neural Network Architecture (Deep Learning) for Fruit/Vegetable Sorting (Inception V3). High torque geared DC motors are used to operate the Conveyor belts. Graphical User Interface (GUI) is to be created and incorporated to make the system user friendly. Raspberry Pi 3B has been used to interface the entire system. System is programmed using Python 3 with Open CV and TensorFlow modules. System has been designed to use Solar Energy to make it sustainable. System is designed to sort 30 fruits/vegetables in a minute.

The effective operation of photovoltaic (PV) systems is substantially deterred by various factors like irradiation, temperature, soiling (dust accumulation), icing, and air mass. This paper considers the effect of soiling on the performance and operational life of a PV panel. Initially, the physical and chemical properties of the five most commonly encountered dust particles are examined using the images and data obtained through scanning electron microscope (SEM) and electron dispersive spectrometer (EDS). The impact of these soiling agents on the electrical performance and operational life of a PV panel is investigated in laboratory conditions using a test setup. I-V curves for each of the dust samples are obtained and the reduction in power output is also documented. Furthermore, the paper highlights the invalidity of generalized cleaning techniques for PV panels by taking into account the heterogeneity of the dust particles encountered in PV installations across the globe.

In this study, a DC/DC Boost (step-up) converter operating in Continuous Conduction Mode is designed and analyzed (CCM). The state space averaging technique is used to analyze both the open-loop and closed-loop analysis of the boost converters in order to calculate the transfer function. The step-up converter boosts the output voltage to 4.6 V from an input voltage range of 2.9–4.5 V, and a maximum output current of 300 mA. The Pulse Width Modulation (PWM) technique generates the closed-loop control signal for a fixed frequency of 1.7 MHz. The complete system is modeled using MATLAB & Simulink and simulated in UMC 180 nm technology on Cadence Virtuoso platform.

The Smart cities with Information and Communication Technology (ICT) have made our lives simple, efficient, effective, and reliable. With the support of different technologies, the development of Smart Cities reached to the highest level by enabling cloud applications. In the case of smart street lighting, the primary objective of energy management in Smart Cities is to reduce power consumption, maintenance costs, and self-designed inbuilt energy meter for analysis. In this paper, the features of a Centralized Control Monitoring Systems (CCMS) system for LED street lighting, maintenance, and forecasting analysis of power consumption using a power saving strategy with real-time data are proposed for the development of Smart Cities.

This article focused on the development of microstrip patch antenna with linear and planar array distribution structures. Ansoft HFSS electromagnetic simulation software was used to analyze the radiation and reflection parameter of the designed element. Initially, single element antenna was realized and further 1 × 2, 2 × 2, and 1 × 4 array structures were analyzed with a due consideration of two different substrate materials such as Duroid-5880 and FR-4. 1 × 4 array antenna with Duroid-5880 offers highest gain of 14.12 dB. To validate the design concept, 1 × 2 array antenna was fabricated. S₁₁ parameter was measured using a vector network analyzer and it has downward frequency shift by 54.5 MHz as compared to simulated results. The radiation characteristic of the antenna was measured on anechoic chamber, and the E/H plot reveals that antenna is unidirectional. Measured gain is enhanced by 7.82 dB as compared to simulated results.

Network on Chip (NoC) is very useful in connecting the different components of a chip. Many NoC topologies like 2D mesh, torus etc. are already in use. RiCoBiT (Ring Connected Binary Tree) is a new topology which was recently proposed and studied. RiCoBiT is a topology where performance parameters like maximum hop and average hop are better than the other topologies. Area parameters like wirelength and number of wire segments are approximately the same as compared to other topologies, but one of the biggest problems with this topology was the lack of an adaptive routing algorithm. Thus, we propose a new routing algorithm which is adaptive in nature. The algorithm is tested for all cases: destination and source in the same ring, destination is above the source ring, and the destination is below the source ring. The testing of the algorithm was performed by using a simulation setup for E-RiCoBiT- II (Enhanced RiCoBiT). It is observed that the topology is 100. The average hop, maximum hop, and total time are studied for the proposed architecture and algorithm.

This paper discusses the configuration of an autonomous hybrid VTOL (Vertical Take-Off and Landing) aircraft is systematic for the last mile delivery of medicines to rural and remote places. The developed aircraft has a payload capacity of 4 kg, a flying period of up to 45 min, and a high stability speed of 65 km per hour. This UAV is fitted with intelligent systems that enable it to operate autonomously even in places without a Global Positioning System and away from obstructions, which is essential in disaster-prone areas. This VTOL's folding wing structure reduces the Take-off and Landing area, making it more reliable.

This paper discusses an implementation of autonomous navigation functionality on a differential drive mobile platform called AMAR (Agile and Multipurpose Autonomous robot) based on the Robot Operating System (ROS). The paper's primary contribution is to describe an unmanned platform capable of mapping, localizing, and navigating in an enclosed environment that is well-suited for industrial operations. The approach used for mapping is through the ROS framework, a technology stack containing multiple software stacks for controlling, visualization, and debugging the robot, wrapped with the help of sensors such as RGBD camera and LiDAR (Light Detection and Ranging). For navigation, Adaptive Monte Carlo Localization is used for localizing the vehicle and thus leading to autonomous operation. The proposed solution is computationally efficient and capable of dynamic obstacle avoidance. The simulations were performed using the 3D physics simulator Gazebo and the results were visualized using RViz.

Pain is caused by particular nerve fibers that transmit impulses to the brain, where they are interpreted. Pain has a multidimensional nature, making it hard to decipher, because of its subjectivity. Chronic or recurring pain is a common impairment among the elderly, significantly bringing down their quality of life. Assessment and treatment of pain are critical challenges in providing interventions for a variety of illnesses, especially because their origins are closely related to deep rooted issues. There is a rising interest in finding objective, nonverbal methods to quantify pain in people who are unable to self-report discomfort, such as those suffering from dementia or those in a minimally conscious state. Though, self and subjective reports are the common means of measuring the impact of pain and assessing its intensity, studies suggest that there are multiple sensory means through which a cumulative and concise data on occurrence of pain (tonic or phasic) can be retrieved, such as, through HRV (Heart Rate Variation), EEG/ECG/EMG readings and health performance metrics based on everyday activities. Our project aims to utilize the space of sensory technologies, with a focus on EEG data and IoT to conduct pain assessments with the help of machine learning techniques and explore IR therapy as well as electrical stimulation in subjects that experience chronic pain. Our intervention in pain assessment looks to combine subjective reporting and sensor based crucial data points on occurrence of pain to derive a reliable method of pain reporting.

In this work, a miniaturized, broadband, coaxially fed rectangular microstrip patch antenna (MPA) is designed for the traffic alert and collision avoidance system (TCAS) application. The TCAS is a legally mandated aircraft surveillance system that is installed in all aircraft to prevent accidents. A novel structure is proposed to minimize the overall size of the conventional TCAS antenna while increasing the bandwidth (BW). Many methods, such as multiple symmetrical slits, parasitic patches, and defects on the ground structure, have been incorporated into the conventional design to achieve miniaturized broadband antenna. The proposed antenna structures are designed and simulated using the High-Frequency Structure Simulator (HFSS) software. For the proposed design, antenna parameters such as impedance, VSWR, BW, reflection coefficient, gain, and radiation pattern are evaluated. The conventional antenna resonates at 1.06 GHz with 31.30 MHz BW (1.07–1.04 GHz), VSWR of 1.18, and a maximum gain of 4.68 dB, whereas, the proposed TCAS antenna operates at 1.07 GHz with 75.4 MHz BW (1.1–1.03 GHz), VSWR of 1.12, and a maximum gain of 3.31 dB. In the proposed design, the antenna gain is reduced by 1.38 dB, and it is mainly due to the reduced antenna size. The proposed antenna has improved the impedance BW from 31.3  to 75.4 MHz, which is more than double the BW of a conventional antenna. Furthermore, the overall antenna size has been reduced by 30.77%, making it an excellent choice for aircraft surveillance/TCAS applications.