Sunteți pe pagina 1din 15

Department of Electrical Engineering Ph.D.

Admissions 2013-14 (Spring Semester)

Abstracts of Projects (only for TAP/RAP category)

Note: 1) These positions are other than TA positions. 2) Candidates admitted to TA category can choose any supervisor; however, candidates admitted under TAP/RAP category may have to work with the specific Project Investigators listed in this booklet. 3) Besides projects listed in this booklet, Bharti Centre will sponsor one Ph.D. student in the general area of communication and signal processing.

Project Investigator: Prof. Jayanta Mukherjee Title of the project: VCO design at millimeter wave frequencies Desired specialization: EE 4 Abstract of the project: 60 GHz low phase noise VCO design The 60 GHz frequency band is increasing coming into attention of the communications community due to the huge bandwidth available. In 2009 IEEE provided the 802.15-3c standard for WPAN networks operating at this frequency band. This band of frequencies offers the potential for giga bit data transfer wirelessly. In order to do so we need the appropriate hardware. A low phase noise VCO is proposed in this project which will provide lower phase noise than what has been demonstrated so far. The publication potential of this project is immense since very few people are working at this high frequency band unlike the lower frequency bands like Bluetooth, WiMax etc where a lot of work has been done. Our Lab has recently acquired the test equipment necessary for testing VCO chips at these high frequency bands.

Project Investigator: Prof. Bipin Rajendran Title of the project: Bio-inspired learning algorithms for spiking neural networks Desired specialization: EE 5 Abstract of the project: In this project, we propose to develop algorithms for spiking neural networks that are inspired by biological networks. Most current neural networks employ back-propagation or other similar algorithms that are based on internal membrane potentials. However, these are not very amenable to hardware implementation. In this project, we will develop learning algorithms that are based on spike timing, which will allow really efficient hardware implementations. An example biological network is the 7-neuron network in the worm C. Elegans that performs thermotaxis (learning and tracking the temperature where the worm was grown). We will demonstrate the acceleration and efficiency in computation from such algorithms in FPGA and GPGPU platforms.

Project Investigator: Prof. Bipin Rajendran Title of the project: Nanoscale devices for Neuromorphic computing Desired specialization: EE 4 Abstract of the project: In this project, we propose to develop bio-inspired nanoscale electronic devices approaching the scale (sub 100nm) and power efficiency (100 mV, 100nA) of biology. We will demonstrate that memristive devices could be engineered to mimic the properties of biological synapses in 4F2 area, and explore the architectural level implications of building large learning systems based on such compact devices. The most important computational element is a two-terminal, low power nanoelectronic device mimicking the properties of biological synapses i.e., it conducts information in one direction, and its effective conductance is determined by the time of arrival of pulses at its input and output terminals. We will use the facilities of IIT Bombay Nanofabrication Facility to build such nanoscale devices based Metal-insulator-metal RRAM devices.

Project Investigator: Prof. Bhaskaran Muralidharan Title of the project: Investigation of spin-torque based nanomagnetic strategies for spintronic logic and memories Desired specialization: EE 4 Abstract of the project: The project carries the primary objective of developing a quantum transport simulator that couples spin transport phenomena such as spin injection, transport and detection with magnetization dynamics. Two classes of systems to be considered here are based on the spin torque effect on the magnetization dynamics of (a) the collective magnetization and (b) the individual spin dynamics. In the collective case we focus on spin-transfer torque based MRAM structures and all spin logic devices while in the individual spin case, on spin torque based transport across quantum dots and single molecule magnets. The developmental stage of the project will build phenomenological models to benchmark with existing literature and the advanced stage will develop a semi-empirical/ab-initio framework to tackle realistic device structures. The planned simulator will prove instrumental in investigating emerging and exploratory nanomagnetic logic paradigms such as spin transfer torque MRAM, all spin

logic, and possibly spin based neuromorphic computing. We will also investigate some of the state-of-the art spin caloritronic devices that are currently being explored in the new and exciting frontier of spin based energy conversion.

Project Investigator: Prof. Maryam Shojaei Title of the project: Analog, Mixed-signal and RF IC Development and Test for biomedical applications Desired specialization: EE 4/ EE 5 Abstract of the project: In this project we explore, design, implement and test energy efficient techniques for signal or data transmission from low-power sensors for biomedical applications. The ultimate goal is to develop a prototype to demonstrate effectiveness of the technique(s).

Project Investigator: Prof. Sibi Raj Pillai Title of the project: Distributed Power control and Rate-adaptation in Wireless Desired specialization: EE 1/ EE 2 Abstract of the project: The design of distributed multiple access for wireless systems is the stated objective of the project. Relying on the information theoretic capacity region of multiple access channels, we will propose novel schemes to improve the communication data-rates under challenging conditions, where users have varying levels of information about the medium. A dual problem tackles energy minimization under delay constraints in multi-user networks. The larger scope of the project also gives freedom to investigate any model of communication-theory/information-theory pertaining to wireless. Candidates with strong enthusiasm to learn/practice communication theory are encouraged to consider this project.

Project Investigator: Prof. Gaurav Kasbekar Title of the project: Resource Allocation in Cellular Networks Desired specialization: EE 1 Abstract of the project: The broad goal of the project is to design and analyze mechanisms for resource allocation in cellular networks. Some examples are as follows: (i) To design algorithms for Inter Cell Interference Coordination, wherein different base stations jointly assign the available resources (e.g., time slots, frequency channels) to their associated mobile users so to limit interference and enhance network performance. Such algorithms may be centralized, semicentralized or distributed. (ii) To analyze and compare the performance of various frequency reuse schemes, e.g., reuse factor 1, fractional and selective frequency reuse schemes. (iii) To design and investigate economic mechanisms such as auctions and pricing for the allocation of radio spectrum to cellular operators and the sharing of spectrum among them. (iv) To investigate issues specific to heterogeneous networks, which contain cells of various coverage ranges and power levels. A combination of analytical tools such as probability and stochastic processes, algorithm design and game theory, and network simulations will be used in the project.

Project Investigator: Debraj Chakraborty and Madhu N. Belur Title of the project: Coordination and Control of Unmanned Aerial and Ground Vehicle (UAV / UGV) Groups Desired specialization: EE2 Abstract of the project: Design, fabrication and control of Unmanned Aerial and Ground Vehicles (UAVs and UGVs) have become popular research topics during the last decade. Apart from various laboratories around the world, these vehicles have now been used in various military and civilian applications including surveillance, reconnaissance, search and rescue, and remote warfare. However, even today, all these operations require intensive human supervision and fully autonomous operation is still an interesting but challenging research problem. The situation becomes even more complicated when groups of UAV/UGVs are involved since computation, sensing, and communication are distributed over multiple agents and coordinating them to implement desirable collective behaviors is extremely difficult in general. Despite the difficulty, coordination and control of multiple UAVs and UGVs have many applications including automated highway systems, formation ight control, satellite

clustering, mapping of unknown or partially known environments, and distributed manipulation. Given the numerous utilities of such coordinated UAV/UGV teams, effective design and control of such groups is an area of intense current research. This project aims to investigate this challenging problem by developing and implementing algorithms for autonomous coordination and control of multiple UAV/UGV teams. The development of new theories for multi-agent coordination would require multiple tools from sophisticated mathematics. This project will primarily use techniques from algebraic geometry, algebraic graph theory and optimization theory for this purpose. In addition to the associated theory, a testing facility consisting of multiple UAVs and UGVs will be developed as part of this project. After testing various prototype vehicles, the research group at IIT Bombay, has developed a fully programmable quadcopter model as the primary model to be used in the multi-UAV testbed. This project will integrate these (and possibly more) vehicles into a team of completely autonomous groups capable of cooperative operations. Such operations include, but not limited to, formation flying, multiagent surveillance, search and rescue, group pursuit and evasion. Each UAV will have a suite of inertial sensors for low-level stabilisation, and a camera-based motion capture system will be used to provide real time position information to UAV/UGV teams. Onboard IEEE 802.15.4 compatible wireless transceivers will provide communication between vehicles and the base station as well as direct inter-vehicle communication. Project Investigator: Prof. V. Ramgopal Rao Title of the project: Centre of Excellence in Nanoelectronics: Phase-2 Desired specialization: EE 4 Abstract of the project: The Centre of Excellence in Nanoelectronics (CEN) at the Indian Institute of Technology Bombay was initiated in the year 2006 under which a world class nano-fabrication facility was established at IIT Bombay. The project is funded by the Ministry of Communications and Information Technology (MCIT), Government of India and a significant number of leading semiconductor industries currently participate in its programmes. CEN offers multidisciplinary research opportunities involving faculty from various departments.

CEN Project has deliverables in the area of fabrication and characterization of traditional CMOS nanoelectronic devices, novel material based devices (III-V compound

semiconductor, spintronics, opto-electronics), micromechanical systems, Bio-MEMS and polymer based devices. Selected students will be given an option to work in one of these areas, provided there are positions available in the specific deliverable.

Project Investigator: Prof. Pradeep Nair Title of the project: Nanostructured electronic devices for solar energy conversion Desired specialization: EE 4

Abstract of the project: Novel nanostructured solar cells promise a low cost alternative to crystalline Silicon for efficient solar energy conversion. However, the carrier transport and hence the functional parameters that influence the performance of such nanostructured solar cells are yet to be clearly elucidated. In this project, we aim to identify the fundamental performance limits of such devices and further explore novel device designs to improve efficiency. This project is funded by SERIIUS the Solar Energy Research Institute for India and the United States to develop and ready emerging and revolutionary solar electricity technologiestoward the long-term success of India's Jawaharlal Nehru National Solar Energy Mission and the U.S. Department of Energy's SunShot Initiative.

Project Investigator: Prof. Pradeep Nair Title of the project: Nanoscale electronic devices for bio/healthcare Desired specialization: EE 4 Abstract of the project: Biology and healthcare are often identified as the most promising fields for semiconductor industry beyond the CMOS era. Indeed, the progress in this technology over the last decade dwarfs the achievements of Integrated Circuit (IC) technology by a significant margin. Nanoscale electronic devices (like Si Nanowire, Carbon nanotubes) and MEMS based devices are increasingly being explored for applications like early detection of diseases, genome sequencing, hazardous chemicals, etc. In this project, we aim to understand the fundamental scaling principles of such devices and explore optimization routes.

Project Investigator: Prof. Abhay Karandikar Title of the project: Dual-Connectivity LTE-WiFi Solutions for Broadband Wireless Networks Desired specialization: EE 1

Abstract of the project: Effective use of the cellular 4G network in conjunction with the WiFi network gives several advantages to the operators and the mobile users. The dual access propositions reduce the overall cost of the offering to the subscribers by making effective use of the license free bands. However this is possible only when the LTE and WiFi networks are tightly integrated and well managed by the operators and users. Interworking between the 3G/4G and WiFi networks is currently supported at the Core Network level but to make the optimal use of both the networks and provide good Quality of Experience to the end user, interworking in the Radio Access Network segment is essential. LTE-WiFi interworking is a topic of interest globally as well and is being discussed in forums like 3GPP and IEEE. There are a number of challenges which are to be addressed in order to achieve tight integration and coordination between LTE and WiFi networks. This project proposes to addresses some of those challenges related network, discovery, mobility, quality of experience and energy efficiency. Project Investigator: Prof. Abhay Karandikar Title of the project: Test-Bed for Affordable Broadband in India using Backhaul in TV White Space Desired specialization: EE 1 Abstract of the project: The objective of this proposal is to investigate the feasibility of a backhaul network in India by utilizing the UHF or TV band spectrum. For this purpose key technical issues need to be addressed which are as follows: 1. An experimental setup or test-bed will have to be deployed in a sub-urban or rural area and broadband/Internet connectivity will have to be demonstrated using TV band frequencies. A selection of appropriate protocol has to be made either from existing choices such as IEEE 802.11af and IEEE 802.22, or development of new protocols will be needed if these are not feasible.

2. A throughput or capacity planning of the backhaul network will have to be done to ensure that backhaul is sufficient for providing broadband speeds in the rural areas. This will, in turn, require radio-propagation model and novel network scheduling algorithms. The scheduling algorithm should be fair in a multi-operator environment. 3. Coexistence or interference management of multiple operators providing backhaul services in a single band will have to be addressed. Since the UHF bands, whether unlicensed or in white space access, will be adjacent to other licensed channels (such as LTE-A), study of adjacent channel coexistence will also be important. 4. Based on the technology present in the adjacent bands, and the technology adopted in the channel used for backhaul networks, appropriate spectral mask and roll-off characteristics need to be defined. The out-of-band emissions must be extremely low so as to ensure minimum interference to the primary services in the adjacent bands. 5. Indian rural terrain is different from the developed countries and India-specific propagation models are not available in the literature. Hence, we will study the propagation characteristics in the 470-585 MHz band under different terrain conditions in the Indian context. ================================================================== Project Investigator: Prof. Anil Kottantharayil Title of the project: Indian Nanoelectroincs Users Program (INUP) Desired specialization: EE 4 Abstract of the project: Indian Nanoelectroincs Users Program (INUP) is a unique research enabler platform funded by the Department of Electronics and Information Technology of the Government of India and operational at the IIT Bombay Nanofabrication facility (IITB-NF) at the department of Electrical Engineering. Through this project, researchers from academic institutions from across India are allowed to use the IITB-NF facilities for fabrication and characterization of electronic devices. Since 2008 more than 100 institutions have used this facility thereby bring high level of visibility to IITB-NF. The project also offers opportunities for aspirants to pursue Ph. D. while mentoring INUP users in their endeavour. The topics of research offered by faculty members are list below. The students offered positions under this program are expected to devout 8 - 20 hours per week on mentoring INUP users and parallely do their Ph. D. A model which is quite successful since 2008. This offers you a unique opportunity to learn a broad spectrum of applications of nanoelectronics in close proximity in addition to a deep understanding of your own chosen theme of research.

MEMS & Sensors- To build an ultra-low power (or self sustaining) wireless sensor network platform based on the nano-electro-mechanical-systems through the integration of energy scavenging technologies with the sensor technologies. Ultra-dense memory using multi-bit complementary resistance- In this project, we propose to develop multi-bit complementary resistance switches (CRS) to build the nextgeneration solid state memory technology. Recently one-bit CRS switches have been demonstrated which integrate both access device and memory device in the same physical device, thereby permitting efficient integration of cross-bar array based memories. However, it will be essential to store more than 3-4 bits of information per physical device in order to develop a commercially viable memory technology. In this project, we will use the facilities of IIT Bombay Nanofabrication Facility to build such nanoscale Metalinsulator-metal memristive devices. Study of vertical gate-all-around SiGe/Ge devices- Study of vertical gate-all-around SiGe/Ge devices: This project will investigate vertical gate-all-around devices for future CMOS technology nodes. Issues such as strain relaxation, vertical etch damage and its passivation and effects of quantum confinement on device performance will need to be investigated using device fabrication and characterization. Study of vertical source-drain diodes in SiGe/Ge layers- This project will investigate vertical source and drain junctions in SiGe/Ge layers for future CMOS technology nodes. Issues such as strain relaxation, vertical etch damage and its passivation, doping, contact and series resistance, and effects of quantum confinement on device performance will need to be investigated using device fabrication and characterization. III-nitride semiconductor quantum wires as electron waveguides for quantum information processing- It may be possible to achieve massively parallel quantum, information processing using electron waveguides. In this project, we will fabricate electron waveguides based on III-nitride semiconductor heterostructures, and study the electron transport properties therein. Engineering the interface of High k dielectrics for next generation CMOS devices- In this project, we plan to investigate the interface of ultrathin high-k dielectrics grown/deposited on Si, Ge,SiGe and III-N substrates for next generation CMOS devices. Various high-k dielectrics such as HfO2, Gd2O3, ZrO2 and their alloys will be growndeposited on several substrates primarily using molecular beam epitaxy (MBE) and atomic layer deposition (ALD) techniques. The interface of these high-k dielectrics will be engineered by both pre and post deposition treatments and characterized via several in-situ and ex- situ diagnostic tools such as reflection energy high electron diffraction (RHEED), high resolution x-ray photo electron spectroscopy (HRXPS), high resolution x-ray diffraction (HRXRD) and transmission electron microscopy (TEM) techniques.Several electrical measurements such as capacitance-voltage (C-V), leakage current (J-V) analysis, interface traps density estimation, etc. will be carried out on MOS capacitors fabricated under various experimental conditions. Finally long channel MOSFETs will be fabricated to study the effect interface treatments on their performance. This project will also involve significant modeling and simulation to understand and explore the basic electrical characteristics and hence to identify the functional dependence of various process parameters for better device performance.

RRAM Devices and Systems- Resistance RAM (RRAM) based systems are proposed for replacement of DRAM, Flash and 3D embedded memory. The research is focused on materials (Si, SiGe, metal oxides), device to circuits and systems for improving memory performance and unique functions. The students will develop wide set of skills in theory, simulations and experimental research like fabrication of nanoscale devices and measurements. Spin-dependent transport in devices- In this project the student will fabricate nanoscale spintronic devices such as magnetic tunnel junctions which are useful for memory applications, spin-transistor etc. Characterization of devices by dc and rf transport measurements will be carried out. The results obtained will be interpreted in terms of existing or new models of spin-dependent phenomena.

Project Investigator: Prof. Souvik Mahapatra Title of the project: CMOS device reliability (BTI, HCI and mixed BTI HCI) Desired specialization: EE 4 Abstract of the project: This is a sponsored project from Synopsys, and the work will involve development of models for CMOS device reliability (BTI, HCI and mixed BTI HCI), working together with Synospsys in implementation of those models in Sentarus 3D TCAD, and model parameter calibration and model validation against experimental data. Work will also involve reliability characterization of FinFET devices for sub 14/16nm node.

The candidate should have background knowledge of CMOS device physics, and working knowledge of TCAD simulators is a plus. Project Investigator: Prof. Shalabh Gupta Title of the project: High-Speed Integrated Circuits for Next Generation Optical Communication Systems Desired specialization: EE 4/EE5 Abstract of the project: The student will be designing high-speed ICs, designing test setup, and testing the integrated circuits design by him/her for high-speed low-power coherent optical communication links.

The project would mostly involve analog circuit design using blocks such as mixers, limiting amplifiers, oscillators and phase locked loops. Some digital design may also be required.

Project Investigator: Prof. Shalabh Gupta Title of the project: Advanced Modulation Techniques and Signal Processing Algorithms for Next Generation Optical Communication Systems Desired specialization: EE 1/EE5 Abstract of the project: The student will be studying and developing new modulation and coding techniques for next generation (coherent) optical communication systems. The student will be expected to work closely with others to help in the implementation of the proposed algorithms in hardware and testing the same.

Sponsored Project: National Programme on Perception Engineering, Phase II Investigators: Prof PC Pandey, Prof V Rajbabu, Prof S Chaudhuri, Prof P Rao, Prof Sharat Chandran EE Specialization Areas: EE1 (Communication), EE5 (Electronic Systems) Topic 1) Speech training aid (PCP/VR) Objective: Providing a visual feedback of articulatory efforts involved in speech production to assist in the acquisition of speech production despite a lack of auditory feedback. Required background: Signal processing, speech processing, communication, biomedical engineering, instrumentation. image processing,

Topic 2) Speech perception aids (PCP) Objective: signal Signal processing in a hearing aid to compensate for various distortions associated with sensorineural hearing loss to improve speech perception. Required background: Signal processing, speech processing, communication, biomedical engineering, instrumentation.

Topic 3) Haptic system for analyzing motor disorders (SC) Objective: Developing a haptic system for interacting with soft, virtual objects to analyze the tremors which would give an account of the amount of vibrations in the movement of the hand of the subject, for application in early detection, progress analysis, and recursion avoidance of various nervous disorders. Required background: signal processing, communication, biomedical engineering, instrumentation. 1. Speech Training Aid (PCP/VR) Most children with severe to profound hearing impairment do not develop speech due to the lack of the auditory feedback of their own speech and this impairment becomes a handicap in developing language skills. A speech training can provide visual feedback of articulatory efforts involved in speech production to help a hearing-impaired child in speech and language acquisition. The major challenges are (i) signal processing for getting the information on the articulatory efforts from the speech signal, (ii) visual display of the information in a manner that helps in correcting the articulatory errors without straining the perception process, and (iii) evaluation of the effectiveness of the training system and devising methods for fine tuning the system for requirements of the individual learner. The estimation of the vocal tract shape involves speech processing based on a model of the oral cavity as a concatenation of several sections of variable cross-sectional areas and the processing gives ratios of these areas, using Wakita's LPC based method. The earlier methods work only during steady-state vowel segments and not during the consonantal segments, show a high level of variability during steady state vowels, and the estimates vary with speakers. Work by our group has resulted in estimation of the articulatory efforts during stop consonants and in improving the consistency of the estimation. PC Pandey and MS Shah, IEEE Trans. Audio, Speech, Language Process., vol. 17(2), pp. 277-286, 2009. KS Nataraj, Jagbandhu, PC Pandey, and MS Shah, Proc. National Conference on Communications 2011 (NCC 2011), Bangalore, India, paper SpPrII.4. Jagbandhu, KS Nataraj, and PC Pandey, Proc. 13th Annual Conference of the International Speech Communication Association (Interspeech 2012), Portland, Oregon, September 9-12, 2012, paper no. 497. Further work in this area will involve (i) estimation of the vocal tract shape for fricative and nasal sounds, and (ii) improving the consistency of vocal tract shape estimation by estimation of time-varying area of lip opening during speech utterances. 2. Speech Perception Aids (PCP) The objective is to develop speech processing techniques for improving perception by persons with sensorineural hearing impairment and also by persons with normal hearing under adverse listening conditions. The following are the major problems faced by most of the hearing-impaired listeners, particularly those with sensorineural loss: i) Frequency-dependent increase in hearing thresholds (particularly at higher frequencies), severe reduction in dynamic range of hearing (due to increase in hearing threshold levels

without any corresponding increase in upper comfort levels), and abnormal growth of lodness. ii) Reduction in perception of weak speech phonemes due to increased intra-speech spectral and temporal masking. iii) Increased masking effect of noise in the listening environment. A partial solution to the first problem is available in most of the modern hearing aids, while the solutions to the other problems are under research. Our group has earlier developed two methods for reducing the effects increased intra-speech spectral masking: Binaural dichotic presentation using a pair of perceptually balanced comb filters for persons with moderate bilateral loss. Ref. P.N. Kulkarni, P.C. Pandey, and D.S. Jangamashetti, "Binaural dichotic presentation to reduce the effects of spectral masking in moderate bilateral sensorineural hearing loss", Int. J. Audiology, vol. 51(4), pp. 334-344, April 2012. Multi-band frequency compression for monaural presentation. Ref.: P.N. Kulkarni, P.C. Pandey, and D.S. Jangamashetti, "Multi-band frequency compression for improving speech perception by listeners with moderate sensorineural hearing loss", Speech Communication, vol. 54(3), pp. 341-350, March 2012. The test results have shown significant improvement in speech perception. Further research will involve developing the techniques for reducing the effects of increased intra-speech temporal masking for improving the perception of transient parts of speech, and integration of the techniques with techniques for compensating for frequency-dependent increase in hearing thresholds and decrease in dynamic range. 3. Haptic System for Analyzing Motor Disorders (SC) Our earlier haptics related work has involved developing (a) interface through a pointbased force rendering interface, (b) a haptic based game that involves playing snooker through haptic interface, and (c) theoretical analysis of Weber sampling of haptic signals. We are currently working on (a) distance field based object reconstruction from haptic data, (b) evaluation with haptic texture rendering, high frequency (texture) component generation, and rendering of haptic force, (c) reconstruction of object with varying stiffness and rendering of surfaces from point cloud data, and (d) scaling (zooming in and out) of objects during haptic rendering and rendering of deformable objects. As a continuation of the earlier and ongoing work in haptics and as an application with a significant societal impact, it is proposed to develop a haptic system for analyzing the motor disorders in human subjects. The system will be capable of interacting with soft, virtual objects to measure the tremors which would give an account of the amount of vibrations in the movement of the hand of the subject. Such type of system would help in early detection, progress analysis and recursion avoidance of various nervous disorders like that of Parkinsons disease or numbness of fingers.

Project Investigator: Prof. B.G. Fernandes/Prof. M.B. Patil Project title: National Centre for Photovoltaic Research and Education (NCPRE) Desired specialization: EE 3/EE 4

Abstract of the project: Solar photovoltaic technology is an important alternative to fossil fuel based energy sources. The technology is attractive due to the abundance and easy availability without causing environmental damage and scalability (small systems powering digital watches to large systems of several mega watt for utility scale). Crystalline silicon solar cell technology has the largest market share. However the commercial cell technologies lag behind the highest efficiencies reported. The main reason is the cost of high efficiency solar cells. One of the research topics being pursued under the project National Centre for Photovoltaic Research and Education (NCPRE), funded by the Ministry of New and Renewable Energy (MNRE) is the development of cost effective crystalline silicon solar cell technology. A key element of this is the development of thin films for effective passivation of crystalline silicon surfaces and integration of the same in solar cells. The activities would involve all aspects of solar cells including fabrication, simulation and characterisation and power electronics related to photovoltaic applications.

Project Investigator: Prof. Madhu Belur Project title: Efficient data based approaches for real-time fault detection and diagnosis of Self Powered Neutron Detectors (SPNDs) Desired specialization: EE 2 Self Powered Neutron Detectors (SPNDs) are sensors that are widely used in nuclear power industry for estimating the neutron flux in a nuclear reactor. However, over a period of time, these detectors can develop faults and become un-usable due to a variety of reasons. Recently, we have developed data based models for SPNDs using Principal Component Analysis (PCA). These models can be used to detect a faulty SPND as well as estimate the corresponding true flux. In this work, we propose to implement the developed models in realtime on an existing nuclear power reactor to assess their computational requirements as well as their diagnostic performance. To deal with widely varying operations in a nuclear reactor, we propose to develop a basket of models. Implementing these strategies in real time will require development of computationally efficient algorithms. This work is in close collaboration with BARC and NPCIL. Scope of Work: Computational, theoretical.