R& D at Amrita
Amma places a strong emphasis on research activities in all of Her educational institutions. Rather than simply memorising all the knowledge that has been accumulated in a particular field, Amma wants Her to children to delve deeper, to explore, to discover for themselves. She is always encouraging the students and faculty of Amrita Institutions to undertake research and development work that will benefit society.

They are heeding Her call: exciting, original research is taking place in the fields of information technology, engineering and medicine, among others.

Ph.D Candidates

 

Ph.D Candidate

Name: B. Rajathilagam

Title: Visual Data Reconstruction over Wireless Sensor Networks

Brief Note:
Information Reconstruction is a challenging task. Moreover visual media such as image and video add to the complexity with their mere size and stringent real-time constraints. It is intriguing to observe that some tiny insects/animals with multiple reflective surfaces in their eyes can get the image of an object. Taking the cue, this PhD work aims at placing many tiny sensors in a geographic location and design some suitable protocols for information transfer across sensors to central information server over Wireless Sensor Network. This work also proposes novel techniques for information reconstruction.

 

 

Ph.D Candidate

Name: Ms.P.Supriya

Title: FPGA based four quadrant electric energy meter with remote demand management facility.(Tentative one)

Brief Note:
(a)Development of algorithm for the measuring of real,reactive and harmonic power in a three phase electric power system

(b) Development of a VLSI Circuit for the above purpose with remote reading using wireless technology

(c) Incorporating remote load management facility in the VLSI circuit.

 

 

Ph.D Candidate

Name: B.Yamuna

Name of the Guide:Dr.T.R.Padmanabhan

Title: Some investigations into the Physical layer of 802.16(WIMAX) and allied issues

Brief Note:
Some aspects of implementation possibilities of communication between Base station and Service stations at the PHY layer of 802.16 (WIMAX) are studied.Typically each communication takes place through datapackets in a multiplexed frame format.The communication is supported by QOS constraints and a chain of IDs.Bandwidth usage,data rate realization,number and variety of user pairs supported ,increase in distance between Base station and associated Service stations are the main/key issues in focus.

 
 

Ph.D Candidate

Name: Ms.S.Veni

Name of the Guide: Dr.K.A.Narayanankutty

Title: Vision based Hexagonal Image Processing

Brief Note:
The work explores the biological vision systems and Synthetic Radar Images with Hexagonal Image Processing (HIP) with an effort to enhance processing speed and quality of the image. The major work will be in the area of Image processing. The initial effort had been to develop a boundary contour system for dynamic compression and edge enhancement on which results are published. But work in the framework of HIP, research work is rapidly taking place and similar works have appeared in the
very recent papers. Therefore, the direction of exploration has been now changed more towards image enhancement, target tracking and fusion.

 
 

Ph.D Candidate

Name: Mr.K. Venugopalan

Name of the Guide: Dr.K.A.Narayanankutty

Title: New algorithmic techniques for Non-smooth H-infinity control
applied to Power Control

Brief Note:
Research on H-infinity control design has been going on for the past 15 years and newer and better techniques are evolving. Recent algorithms using semi-definite programming have given fresh insights. Search methods once discarded for some time is again emerging as a powerful technique. Dynamic power control and filtering techniques are now vastly used in power systems, which call for special type controllers. The controllers used are often non-smooth and thus require special design techniques. This research effort is towards developing newer techniques for design.

 
 

Ph.D Candidate

Name: C.Santhosh Kumar

Name of the Guide: Dr.Mohandas

Title: Multilingual speech recognition - A unified approach

Brief Note:
In speech recognition many languages need to be handled at the same timefor multilingual societies like India . In todays commercial system itis done by having many monolingual systems working in parallel and thecorrect model is selected by a language identification system. It isvery expensive and time consuming to enroll the new languages to suchsystem.

In our approach , we focus on building a unified acoustic model for allthe languages.When a new language is to be enrolled to the system, thesuitable mapping between the multilingual phone set and the phone set of the new language will be derived automatically using a small amount ofbootstrap data.

 

 

Ph.D Candidate

Name: Bhagavathy sivakumar

Name of the Guide: Dr.Mohandas

Title: Predictive modeling of signals in financial markets

Brief Note:
The financial market signals(which are viewed as a mixture of random anddeterministic components) are very challenging to model.An attempt to develop a predictive model and its empirical validation on Indian marketsis underway.Potential strength of soft computing,signal processing anddynamic system theory are exploited in this work.The innovation is in the model’s ability to tackle the complexity of the signal because of it’srandom plus deterministic combination.

 

 

Ph.D Candidate

Name: Prasanth Nair

Name of the Guide: Dr.Mohandas

Title: Reinforcement learning in financial markets

Brief Note:
The technique of Reinforcement learning(RL) has been evolved jointly by the AI and control engineering communities. RL makes use of goal-directed,adaptive agents that attempt to maximize reward in an uncertainscenario. RL algorithms provide an attractive alternative to DynamicProgramming, since the requirements on a complete model of the underlyingsystem are relaxed to some extent. Here the power of RL is investigatedin solving challenging problems in financial markets.

 
 

Ph.D Candidate

Name: Saravanan N

Name of the Guide: Dr.K I Ramachandran

Title: Fault diagnosis of Bevel Gear Drive using Machine Learning Techniques

Brief Note:
The work involves condition monitoring of bevel gear by monitoring the vibration signals. Typical faults are created in bevel gears and vibration signals are acquired. Relevant features are extracted therefrom and classification is performed using various machine learning techniques.

 
 

Ph.D Candidate

Name: Sugumaran V

Name of the Guide: Dr.K I Ramachandran

Title: Fault diagnosis of Roller Bearing using Machine Learning
Techniques

Brief Note:
The work involves application of machine learning techniques to fault diagnosis of roller bearing. The problem is modeled as pattern recognition problem, which involves feature extraction and feature classification. Relevant features are extracted therefrom and classification is performed using various machine learning techniques.

 
 

Ph.D Candidate

Name: M.Elangovan

Name of the Guide: Dr.K I Ramachandran

Title: Studies on Web enabled PDM systems as applied to SMEs in a manufacturing cluster

Brief Note:
Small manufacturing companies join to form a Virtual Enterprise to beat competition. To the external world the group functions as a single enterprise, while internally it is made up of group of individual member companies. They make the whole product, but in fragments of the original order to collectively reach the demanded capacity within the manufacturing deadline. This collaborative effort helps to take big orders beyond their plant capacity and also helps them to share their best practices. The thesis studies the various issues in sharing knowledge, infrastructure, data, order, and responsibility. Issues of accountability, individual identity and distinguishes between private and common data. The above framework is to be brought out in the form of a prototype that is web-enabled.

 
 

Ph.D Candidate

Name: Ms Meera

Name of the Guide: Dr. S S Bhagavan

Title:Acrylonitrile-butadiene [Nitrile] rubber based Nanocomposites : Mechanical behaviour and Barrier Properties

Brief Note:
The use of nanoparticles as filler makes it possible to producefundamentally new materials with a range of unusual mechanical andphysical properties. In addition they have good heat stability,flame resistance, improved gas barrier properties, reducedshrinkage, etc. Interestingly, these changes are achieved at lowfiller loadings (¡Ö 5%). The investigations proposed relate todevelopment of nanofilled nitrile rubber composites with referenceto mechanical and barrier properties, viscoelastic behaviour andenvironmental effects. The nanofillers of interest are nanoclay[Montmorrolonite], nanosilica and nanocalcium carbonate withsuitable treatments. The formulation would be optimised using DoEtechnique and characteristics of control system bladed would be evaluated.

 
 

Ph.D Candidate

Name: Mr. Suresh Kumar

Name of the Guide: Dr. M. Jayakumar

Title: (Tentative) Estimation and studies on the strength of security in a selected cryptosystem in a programmable communication system

Brief Note:
The parameter estimation using reported results and finding the
barrier at which the given values of the system is still a challenging task to be explored.Strength of the security estimation parameters are the main subject of research

 
 

Ph.D Candidate

Name: Mrs. M. Renu

Name of the Guide: Dr. M. Jayakumar

Title: (Tentative) Design, Studies and Implementation of efficient block ciphers using cryptanalysis.

Brief Note:
Designing of efficient block ciphers and its analysis based on
various application related parameters are the basic interest of this research

 
 

Ph.D Candidate

Name: Mrs. Bijay Shankar Mishra

Name of the Guide: Dr. M. Jayakumar

Title: Design and analysis of dynamic source routing algorithm for mobile based wireless networks

Brief Note:
Short write-up on the PhD problem Enhancement in TCP/IP algorithm makes it suitable for mobile networks. This proposal interm degrade the functioning of certain other communication port algorithms and the studies are directed to solve possible bugs in order to make the channel error free.

 
 

Ph.D Candidate

Name: Ms Meera

Name of the Guide: Dr. S S Bhagawan

Title: [Tentative] : Acrylonitrile-butadiene [Nitrile] rubber based Nanocomposites : Mechanical behaviour and Barrier Properties

Brief Note:
The use of nanoparticles as filler makes it possible to producefundamentally new materials with a range of unusual mechanical andphysical properties. In addition they have good heat stability,flame resistance, improved gas barrier properties, reducedshrinkage, etc. Interestingly, these changes are achieved at lowfiller loadings (¡Ö 5%). The investigations proposed relate todevelopment of nanofilled nitrile rubber composites with referenceto mechanical and barrier properties, viscoelastic behaviour andenvironmental effects. The nanofillers of interest are nanoclay[Montmorrolonite], nanosilica and nanocalcium carbonate
suitable treatments. The formulation would be optimised using DoEtechnique and characteristics of control system bladed would be evaluated.

 
 

Ph.D Candidate

Name: Deepak O. M.

Name of the Guide: Dr. K. M. Sreedhar

Title: [Tentative] : Acrylonitrile-butadiene [Nitrile] rubber based Nanocomposites : Mechanical behaviour and Barrier Properties

Brief Note:
(a) Synthesis of a polymer support based on the cross linked copolymers of styrene, N-vinyl pyrrolidone and 4-vinyl pyridine. The cross linkers can be divinyl benzene, hexanediol dimethacrylate, N, N’-methylene bisacrylamide etc.

(b) Synthesis of peptides on this polymer support.

(c) Solid phase synthesis of antimicrobial peptides on these cross linked support.

(d) An attempt to enhance the activity of these synthesized antimicrobial peptides.


 
 

Ph.D Candidate

Name: T. Palanisamy

Name of the Guide: Dr. J.Ravichandran

Title: “Applications of Wavelets in Regression Analysis"

Brief Note:
The Wavelets are being tried to smoothen the hetroscedastic variance functions in Regression analysis. Some problems related to this area will be studied and Illustrated with examples.

 
 

Ph.D Candidate

Name: P. Devajothi

Name of the Guide: Dr. J.Ravichandran

Title: “Wavelet Estimation Of Time series Regression”

Brief Note:
Estimation of a time series regression model using Wavelet Transform. We will focus on the discrete time realization of regression process and apply Discrete Wavelet Transform to a real time process.

 
 

Ph.D Candidate

Name: Babu Devasenapati.S

Name of the Guide:

Title: "Design and optimisation of a continuous flow reactor for biodiesel production"

Brief Note:
This work aims at optimising various parameters involved in the design and optimisation of a continuous flow biodiesel reactor. It also aims at studiying the effect of various mixing techniques and catalysts on the reactor efficiency. The currently available batch processes consumes a lot of time and space.

 
 

Ph.D Candidate

Name: Ms.B.Bala Tripura Sundari

Name of the Guide: Dr.T.R.Padmanabhan

Title: "Investigation into possible realization of Reconfigurable VLSI Systems through optimization in scheduling techniques. "

Brief Note:
The uncommitted nature of functional units and routing resources in FPGAs allow designers to define connections/configurations repeatedly and differently. The possibility of altering configuration after each application is the motivation for reconfigurable design.

Some work on configuration and related optimization to suit specific image processing tasks is available. Reconfigurable schemes have been studied to yield optimal configurations for specific tasks. However applications which demand on-line /real time reconfigurations to carry out a succession of differently structured tasks in a optimal manner can be more challenging. This forms the focus of the work here.

The approach /task demands a critical examination of scheduling algorithms, identification of optimal schedule, for each task and defining reconfigurations after each stage of the structured tasks.
It can also open up possibilities of additional scheduling algorithms. Transferring the above into HDL code and realization in FPGA form tangible realization of the scheme.

 
 

Ph.D Candidate

Name: Ms. A. Vijayakumari

Name of the Guide: Dr. Isha T B

Title: "Control of switched reluctance generator"

Brief Note:
About the problem: As a generator, a switched reluctance machine is unique in its operation. An SRM is operated in the generating mode when the rotor is positioned such as phase inductance is decreasing. Because the SRM does not employ any permanent magnet or field winding, the nature of its excitation during generation, especially under fault conditions, invokes a special study. Also, the capability of the converter to isolate the faulty section A complete simulation and experimental validation concerning this aspect is aimed in this work.

 
 

Ph.D Candidate

Name: J. Gokulachandran

Title: "Reliability analysis and prediction for multi-systems with multi-state components" (Tentative)

Brief Note:
About the problem: As a generator, a switched reluctance machine is unique in its operation. An SRM is operated in the generating mode when the rotor is positioned such as phase inductance is decreasing. Because the SRM does not employ any permanent magnet or field winding, the nature of its excitation during generation, especially under fault conditions, invokes a special study. Also, the capability of the converter to isolate the faulty section A complete simulation and experimental validation concerning this aspect is aimed in this work.

 
 

Ph.D Candidate

Name: S.P.Anbuudayasankar

Title: "A meta-heuristic approach for the multicommodity network flow problems "

Brief Note:
In this research work a hybrid meta-heuristic is being developed for the multicommodity network flow problems which is aimed at the near optimal solution and is to be tested in the real time applications in the Industries. The distribution of the multicommodity from ware houses to retailers is considered here and this NP-hard problem finds important applications in logistics of most of the Industries. The concentration is on the multicommodity stochastic flow networks in which each arc has the capacity and cost attributes as well as lower and upper bounds in the flows.

 
 

Ph.D Candidate

Name: Karuppath Narayanankutty

Thesis Advisors : Prof. Dr. P.Achuthan, Prof. Dr. C.S.Shastry

Title: "Reality in Physics"

Brief Note:
The much-debated interpretation of the Quantum Mechanics is almost three fourth of centaury old. There had not been in the history of mankind such a debate that lasted so many years by the top most brains of the world as this one. The debate that started with Einstein’s school and Bohr’s school culminated in the famous EPR paradox paper. The debate still continues to be live and relevant today and is one of the major unsolved problems of today.

This foundational problem of physics is highly related to the problem of determinism, measurement, wave-particle duality and reality in Quantum Mechanics. The resolution of this is most vital and urgent one. Its relation with the question of nature of time and causality makes it all the more exciting and rewarding. The mysterious nature of time has philosophic, religious and psychological connotations. The (apparent?) arrow and irreversibly of time on one hand and the dimensionality of time on the other as regards to special and general theory calls for close scrutiny. Again the apparent inconsistency between the two major branches of physics (Quantum Mechanics and Relativity) may be linked to the above problem.

 

 


 

 

 
     

©2004-2006, Amrita Vishwa Vidyapeetham
s