Main Dissertation Objectives and Scope

The dissertation is a very crucial part of the MTech program in Biomedical Engineering at Amrita Vishwa Vidyapeetham. The students are given the freedom to choose from a variety of specialisations like Biosensors, Biomaterials, Computational Drug Design, Biomedical Image / Signal Processing and Biomedical Instrumentation, making the programme truly inter-disciplinary. They are actively encouraged to think out of the box and go beyond the confines of their curriculum, to come up with work that is outstanding – Amrita demands no less. A few of the projects done in the last couple of years are shown here.

Some of the Significant Projects

Project 1: Design and Development of Low Profile Substrate Integrated Dielectric Resonator Antenna for Space Applications

A low cost and high radiation efficiency antenna structure, Substrate integrated waveguide (SIW) is designed for Space Applications at 5.8 GHz. Substrate integrated waveguide (SIW) is a guided-wave structure that has the benefits such as high quality factor (Q), high power handling capacity, low loss and is suitable for microwave and millimeter-wave systems. With the development of SIW technology, several components, including active circuits, passive components, and antennas are designed. Designing low profile antenna with good radiation performance is in great demand, especially in space applications such as satellite, aircraft and radar in some military platforms. Slot antennas, for their attractive characteristics such as low profile, conformability to planar or curved surfaces, easy integration with planar circuits and better isolation from feed network, are very suitable for these applications. Dielectric resonator antennas (DRAs) are interesting especially at mm-wave range for their wide advantages such as high radiation efficiency, small size, low cost, and low losses. Substrate integrated waveguide (SIW) is used as a feeding network to further reduce conductor loss compared to other traditional feeding networks. As compared to planar antennas, using current technologies DRA are more difficult and costly to fabricate. In this work, a simple and low cost substrate integrated waveguide fed dielectric resonator antenna (SIW-DRA) element at 5.8 GHz is analyzed.

Top view of SIW fed CDRA with different slot configuration
(a) Existing Antenna (b) Proposed Antenna

Project 2: Multi-User Detection in Sporadic 3GPP Massive M2M Communication via Compressed Sensing

Massive MIMO facilitates the potential drive behind the next generation communication systems, the most awaiting 5G technology. 5G technology aim at Human to Human communication with high data rate as well as the evolution of technologies like Internet of Things (IoT) and Machine-to- Machine (M2M) communication, which can change the face of current communication systems. M2M communications deals with low data rate and short packet transmissions requiring a good quality of service. This also demands congestion free multiple access with low control signaling overhead. This work mainly concentrated on studying the possible supporting techniques for a high performance Massive M2M system under 3GPP standard. Along with that need to reduce the control signaling overhead in M2M communication system is addressed.

M2M Uplink network layout of WINNER II channel model

Project 3: Machine Independent Fault Diagnosis : A Unified Approach

In recent years, condition based maintenance (CBM) of systems has received a lot of attention because of its potential to enhance productivity, maximize operating lifetime and conjointly minimize maintenance and working cost. Support vector machine (SVM) is a widely used machine learning algorithm in the development of CBM fault identification systems

Experimental setup of machine/system independent fault diagnosis system

Project 4: Tri-band Microstrip Patch Antenna for GPS Application

This project presents a tri-band microstrip patch antenna for global positioning system (GPS) receiver.

Project 5: Automating Anxiety Detection using Respiratory Signal Analysis

Anxiety is an emotional disorder which can cause fear, nervousness and worry in individuals. In this work, we explored the development of an anxiety detection (AnD) system using the respiratory signal as its input.

Project 6: Emulation of MIMO Satcom System using Software Defined Radio

In comparison with terrestrial communication systems, satellite communications have gained huge significance over the last decade. It is far superior and economical compared to the other communication systems. Satellite communications have many advantages like broadcast possibilities, provision of service to remote or underdeveloped areas, user has control over own network, etc. Apart from the advantages, satellite communications also possess some disadvantages like huge initial cost, noise and interference, propagation delay and congestion of frequencies. The motivation behind this project is to overcome the above limitations of satellite communication, that is to reduce the cost and improve the performance of satellite communication system and to add more flexibility to such system.To overcome all these issues, MIMO SATCOM system is emulated using Software Defined Radio to provide high spectral efficiency and high reliability at low cost.

Block diagram of proposed system

Project 7: Improving the Intelligibility of Dysarthric Speech

Dysarthria refers to a group of neuro-motor disorders resulting from impairment of one or more organs used for speech production: respiration, phonation, resonance, articulation and prosody. Dysarthria affects the articulation of consonants and vowels causing a substantial decrease in the intelligibility of speech. This work aims at enhancing the intelligibility of dysarthric speech towards developing an effective speech therapy tool. In this therapy tool, enhanced speech is used for providing auditory feedback with a delay to instill confidence in the patients, so that they can improve their speech intelligibility gradually through relearning. In this work, feature level transformation techniques between dysarthric speech and normative speech are used to enhance the intelligibility of dysarthric speech. Speech utterances from Nemours dataset with mild and moderate dysarthria are used to study the effectiveness of the proposed algorithms. The quality of the transformed speech is evaluated using subjective and objective measures.

Project 8: Recognition of Emotions from Video using Acoustic and Facial Features

Emotions are extensively exploited by the human beings for conveying messages. Emotion recognition is a painless task for human beings but for the machine to identify the emotion is challenging. It offers a natural interface between machines and humans, by which the system can understand, interpret and respond to the human emotions. In investigation department they use emotion recognition system to predict the activities of the criminals by analyzing the taped conversations. In this work, the speech and facial features extracted from the video data is explored to recognize the emotions. Since both these features are compliment to each other, on combining them will result in higher performance. The features used for emotion recognition from video data are appearance based features and Speeded Up Robust Features while prosodic and spectral features are employed for speech signal. Support Vector Machine classifier is used to capture the emotion specific information. The basic aim of this work is to explore the capability of speech and facial features to provide the emotion specific information.

Major Projects

Fingerprint Based Detection of Hardware Trojan in VLSI Circuits


Manufacturing of Integrated Circuits (ICs) done mostly by the third party vendors at distant foundries makes the ICs more vulnerable to tampering by the addition of Hardware Trojan (HT).>This work proposes a fingerprint based methodology to detect the presence of the Hardware Trojans present in the design. Hardware implementation of this technique is done by considering the Xilinx FPGA environment. The technique is implemented on a set of ISCAS89 and ISCAS85 benchmark circuits. The results shows the proposed segmentation technique is better than compared to Region based segmentation technique based on reduced overlapping of gates.

Hardware Implementation - Setup for taking Current Fingerprint

Reconfigurable True Random Number Generator on FPGA

Hardware Implementation


A TRNG is a generator which generates the random bits by utilizing the physical process in the circuit. A random number generator includes a noise source which generates the continuous time signal and that continuous signal is digitized using a sampler. The commonly employed method is based on jitter. The usage of jitter in ring oscillator aids in obtaining a high-speed real-time random number generator (RNG). The statistical tests along with internal tests are conducted as an added advantage to ensure security to the architecture. National Institute of Standards and Technology (NIST) tests validated the unpredictability and randomness of the True random number (TRN) generated.

Test Automation Board for Communication Interfaces


Infotainment systems use various communication interfaces to connect to various other peripherals within and outside the Head unit. The project focuses on developing a test platform along with suitable software stack to automate driver level testing of communication interfaces. The study evaluates the limitations of existing verification methods and addresses the feasibility of utilizing Commercial-Off-The- Shelf (COTS) hardware platforms. The design of a software stack to control the test board and automate the test process has been described. Finally, the design of the system using a FPGA platform which provides an additional avenue has been explored.

Hardware Implementation

High Performance CMOS based LC_VCO Design Using High Q-Factor, Field Shield layered Substrate Inductor

Hardware Implementation


The project focuses on designing a CMOS based current reuse voltage controlled oscillator(VCO)with a lower phase noise and improved figure-of-merit (FOM) to operate in a frequency range of (2.07-2.79)GHz. This aims to design on-chip spiral inductor with improved L-density and Q-factor and making use of it in the design of VCO. The use of ferrite film as magnetic enhancement results in an improvement in L-density and Q-factor of the inductor. The proposed inductor improves the L-density by 78.1% and the Q-factor by 53.8% as compared to air core inductor. The 65 nm LC-VCO using the proposed inductor exhibits lower phase noise of -115.96 dBc/Hz and -135.01 dBc/Hz at an offset of 1 MHz and 5 MHz respectively. Measurements show that the proposed VCO consumes a power of 2.85 mW from a supply of 1.1 V.

Ongoing External Funded Projects

Hardware Trojan detection and consistency based diagnosis.
VLSI architecture for reliability based soft decision decoding of turbo codes for satellite communication.

Improved Microaneurysm Detection in Fundus Images for Diagnosis of Diabetic Retinopathy

This work addresses the development of a computer-aided diagnosis (CAD) system for early detection of diabetic retinopathy (DR), a sight threatening disease, using digital fundus photography (DFP). More specifically, the proposed CAD system is intended for detection of microaneurysms (MA) which is the earliest indicator of DR. The system addresses the common challenges in candidate MA detection, which includes detection of subtle MAs and MAs close to each other and those close to blood vessels, thus improving their sensitivity. The use of modified morphological contrast enhancement and multiple structuring elements has significantly improved the detection rate of MAs. Further, a combination of Principal Component Analysis (PCA) and Fourier Transform has yielded a good false positive elimination performance.

Automated Sleep APNEA Detection Using Polysomnography Signals

Sleep apnea is a common sleep disorder which often goes undiagnosed and leads to serious problems like stroke, heart attacks etc. Conventional diagnosis of sleep apnea is done by continuous recording of physiological signals for 6 to 7 hrs during sleep and then manually marking the events. This process is inconvenient, expensive and not affordable to people in rural areas. In this work the detection of obstructive sleep apnea (OSA) has been automated using machine learning techniques. The heart rate variability (HRV) and respiratory rate variability (RRV) parameters, derived from electrocardiogram (ECG) and respiratory effort signals (RES) respectively, serve as input as to the support vector machine (SVM) backend classifier.

Stand - Alone Pulse Wave Velocity Meter

Cardiovascular diseases (CVD) are the main cause of mortality in both men and women. The functionality of the heart and the associated blood vessels have to be monitored regularly for the early detection of CVDs. ECG and PPG are the biological signals usually used for this purpose. Therefore, biomarkers extracted from ECG and PPG would be of great help for the monitoring. One of the potential biomarkers of CVD is found to be the Pulse Wave Velocity (PWV). PWV is the velocity with which the arterial pulse propagates through the circulatory system. Studies have shown the significance of PWV in determining the blood pressure and arterial stiffness. A device which measures PWV is not available in the market. This project aims at the development of such a device. Here, PWV is determined from Pulse Transit Time (PTT), which is the time interval between the R peak of the ECG and the peak of the PPG signal. The device will have a display of PWV, the ECG and the PPG waveforms.

A3B Gene Silencing Control Breast Cancer - A Pharmacogenomic and Genetic Engineering Approach

Breast cancer is one of the most prominent types of cancer. Genetic and epigenetic modifications in the genes are the major reasons behind development of breast cancer. Although numerous genes are associated with breast cancer, up-regulation of APOBEC3B gene is considered as an endogenous enzymatic cause of other major mutations. However, proneness of the disease as well as susceptibility of patients towards drug action varies from person to person, necessitating a pharmacogenomic P4 (Participatory, predictive, preventive and personalized) approach in drug designing. In the present work, a comprehensive pharmacophoric analysis of ‘A3B gene responsiveness’ towards bresst cancer and designing of a novel P4 (anti-breast cancer molecule have been carried out. The results support the possibility of using the template (model drug) or a further optimized molecule for controlling A3B mutated ‘breast cancer’ subject to further clinical trials.

NIRF 2017