Conducting Research: A Personal Technical Experience - 3 books and stories free download online pdf in English

Conducting Research: A Personal Technical Experience - 3

Conducting Research: A Personal Technical Experience

-Part III: At the McMaster University

 

By JIRARA

 

© JIRARA, August 2022

Published by JIRARA

on matrubharti.com

 

All rights reserved. No parts of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, for any commercial purposes without the prior permission of the author and/or publisher.

**

Disclaimer: This is the work of fiction as far as all the characters, their names and the names of all the events are concerned and all these are imaginary and hence any resemblance to the persons (and their lives) dead or alive, and any places are coincidental. Even if a few events might look realistic/’real’, these are fictionalised and the associated names are changed in order to maintain their privacy, honour, and security. No intention whatsoever is meant to hurt any feelings of whosoever, irrespective of their personal/cultural beliefs, social or political inclinations, religion-orientations/practicing/philosophy, life styles, and work/business. The ‘I’, ‘my’, ‘me’, and ‘mine’ (if any used) do not necessarily mean the author of this book/story/article, and these and other such pronouns: her, hers, his, he, she, him, you, your, yours, ours, theirs…; are used for effective personification and dramatization, and the readers should not take these on their ‘own persons’.

**

The readers should take these stories/verses/thoughts with/in good spirit. The presented ideas and material are based, where feasible, on readings and (thought-) analyses of scientific/other open literature (which seemed most profound and trustworthy), with as much care as possibly taken. The readers are requested to verify these notions on their own, and use their own discretion. However, these stories/verses/thoughts/ideas (mostly original) are expressed here with an intention of increasing awareness of the readers with a hope that in an overall sense, their (and ours) consciousness would be heightened (in all and multiple directions), so that we all can live our lives on this planet with true happiness, ever-lasting peace and real joy (irrespective of our orientations). The author and the publisher will not be responsible for any negative effects/situations arising as a result of reading these stories/verses and/or following the suggestions if any; and no discussions/dispute of any kind will be entertained at any time and in any way, manner, and/or forum; because the dictum is that if you like(-d) you read, otherwise ignore, what is the point in making a fuss about it?; anyway you are independent to judge the messages in the articles/stories and utilize for your benefits if found useful, since here the idea is in the direction of ‘consciousness raising’. JIRARA.

***

 

 Conducting Research: A Personal Technical Experience-3

At the McMaster University

 

Preamble: During 1979-80, some NAL (now they are writing as CSIR-NAL) colleagues were applying to foreign universities for higher studies, so, I also followed them, and had applied to USA, and Canada; and I got admissions to the University of Chicago, Waterloo University in Canada, and McMaster University (McMU) in Hamilton; from the latter I was offered full fellowship even without GRE, and without TOEFL test; in the latter I had scored 533, it was required to get admission at other places. I was also offered the scholarship by the Waterloo, but since the amount from McMU was nearly 1000 Canadian dollars more per year, and since by that time I had two children, I had accepted this offer. Here, I was first required to do courses for my master degree, but after one semester if my performance were good, I could be continued automatically to the Ph.D. programme without submitting the thesis for the master’s degree, whereas elsewhere I was supposed to complete and submit the master’s thesis. After joining McMU, I had also got the offer from Queen’s university, but that was of no use to me.

**

Research and teaching assistantships

I had chosen to work in the area of human pilot modelling (based on my previous experience at the NAL, Bangalore), with the intention of applying nonlinear estimation methods, but having joined the program, I had changed the topic to nonlinear filtering at the behest of the chosen supervisor, Prof. N. K. Sinha of the dept. of electrical and computer engg., who had said to me that his friend Prof. V. S. Pugachev (VSP) from Mosco (USSR, Russia) had wanted him to further the research in the area nonlinear filtering theory that was propounded by him (Pugachev himself), and the theory was popularly called Pugachev Nonlinear Filters (PNF).

*

I had registered for two courses for the first semester, and then two for the next; and since I was too busy with the course assignments/project work and/or exams; I had not concentrated much on the research work, and any publication thereof.

*

As a teaching assistant (TA), we were required to handle laboratories for two afternoons in a week, and our work was to supervise the students who would conduct the experiments by themselves, which were already set up by the lab instructors, or technicians; we were supposed to answer any questions the students had while they were doing the experiments, collect the lab records/reports and later on in the next lab period we were to hand over the reports with marks, and comments. Also, as TAs we were supposed to evaluate their class-course assignments, and any tutorial classes, if the professor handed over to us to run. In these classes we were supposed to illustrate problem solving skills for the given subject for which we had to have the solutions with us in advance.

*

In one incidence, a UG student had ditto xerox-copied page-to-page the full assignment and submitted, and I was baffled what to do?: should I give zero marks or so?; then I showed this to the concerned professor (Prof. NKS), who decided to give at least 5 marks, since, although he had copied the assignment, he had submitted in time, and he had not scratched out the name of the original student, of course he had added his own name! What a humanitarian philosophy, or may be a diplomacy, because the student will not come to the professor and unnecessarily argue or whatever!

*

On the other side, a student of Indian origin, had already completed his MBA while his own wife was doing a Ph.D. in the McMU; but since it was taking time for her to complete the study, the husband joined the department to do M.E. so that he could also earn some fellowship and also concurrently be studying so he did not have to remain at home alone; the interesting aspect of him was that he would often come to me and request me to show my answer books of the assignments, and then copy these answers in his own handwriting and then submit; this guy subsequently got the M.E. degree, and got a job in IBM in USA, and his wife with a Ph.D. had joined some university in USA.

**

As a research assistant (RA), I was supposed to study the joint research papers of Prof. VSP, and then come out with some new filtering algorithms. I had xeroxed several papers by him and his co-authors, and then had started studying them, but I was not getting head and tail of his work; the papers though were in English, the symbols were much different, and there was a little text, and the papers would be full of several equations. This was so because, primarily I had not studied nonlinear filtering theory (NFT) earlier in my degree classes; and what I had used in NAL for human operator’s mathematical modelling was linear time series approach and not any nonlinear one.

*

Also, as a part of the monitoring system of the doctoral progress, I was asked to prepare a synopsis report wherein I was supposed to describe in 4 to 5 pages, what research work I would do, and then I was supposed to follow it up. I had mentioned that the PNF will be implemented in real time, this was on the insistence of the guide, though I was not very comfortable with this, despite I had worked in the development of circuits, and worked in NAL on AC20 programming, I had no experience of real time implementation; but the prof. had said that prof. Pugachev had told him that this implementation would be very easy; but I had a gut-feeling that this was not that easy, and certainly for me it was tough; but I had no go, but to agree with the guide.

*

So, I had decided to study the theory of nonlinear filtering, for which I also studied first the existing linear theory once again as refresher self-learning course, and then the existing nonlinear theory, since the PNF sounded much different (from the then existing theories in the literature). For the former, I decided to give two months’ time, and I used to daily sit in the library, almost for the full day, except when I had to monitor the lab classes for the UG/PG students. I got really a good grasp of the existing NFT; and then for the study of PNF, I decided to study it in my apartment environment, since I had completed the course requirements for the degree, and my registration was transferred to doctoral degree, and hence no classes were to be attended except the duties of the TA.

 **

Meanwhile I studied the extended Kalman filter (EKF) and its application to satellite orbit determination, for which I had obtained the master’s thesis form UTIAS (university of Toronto institute of the aerospace sciences/studies); I also had applied the factorisation version of the EKF and wrote three papers from the work so far on: i) application of UD filter for the IEEE conference, USA, ii) similar and expanded work for the IFAC conference in UK, and iii) survey of the orbit determination problem for the IEEE transactions on aerospace electronic systems (AES), USA. The conference papers were presented by my guide at the respective places.

*

During these three years, a good number of non-Indian students used to come to me to get their reports and draft papers checked by me, since their skills in English writing was not very good; and they knew that I had worked in a research lab in Bangalore, and had already written some reports and two papers from my research work. Since, our sitting cubicles were in the fourth floor of the engineering building and this part was called a penthouse, these students started calling me the editor of the penthouse and used to joke and laugh; the deeper meaning of which I came to know much later.

**

Almost after a little more than three years, I was asked that my progress would be reviewed by my doctoral committee, that constituted of a mechanical engg. professor Dr. Dokainish, our department’s professor Dr. C. R. Carter, and my thesis supervisor. At that time I had said to him, in his chamber, that I would not be able to do the real time part of the PNF, since it would be difficult for me to do that, and would take lot of time, and since my scholarship was for 4 years or a little more, it would be a difficult thing to complete my Ph.D. if I had to do this part of the implementation; to which the prof. got a bit upset, and told me that I had promised to do that and written down in the synopsis report, and the doctoral committee would consider the work as incomplete, and even not give me the writing permission for the thesis when  I approach them in the future committee meeting. I then said that it was not really necessary to stick to that, and since this is a research aspect one can change, and the research is an evolving matter; then I had told that I would substitute that part by doing some equivalent theoretical work. He agreed to this, but cautioned me that I should present the work to the committee after he comes back from India after nearly two months, and that he would neither support nor oppose my proposition and the responsibility would rest with me only; to which I had agreed; this was a big and a difficult decision, I had no clue what and how would I do any theoretical work on the PNF, where I had found that he had already done lot of work!

**

To do the detailed study of PNF, I collected and arranged the papers in a chronological order of their publication, and started studying each paper in greater details, and prepared several notes on: a) the symbols, b) their equivalence with the most prevalent symbols, c) the novelty of the work done, d) any similarity or contradistinction with the existing nonlinear filtering theory (mostly from the western literature on estimation and filtering), and e) anything that prof. VSP had left out to be done for the future; this latter aspect was very difficult to find, since by studying all his jointly authored papers I had some gut feeling that they had not left anything to do in theory; but I still kept my hopes very high.

And in the process of two/three months’ arduous study and work (during which I ate lot of walnuts and almonds every day), I had found out that the professor VSP had left out two aspects of the PNF, and then I started working on these and drafted two theoretical papers, and I had kept the draft ready to show to the guide. So, when he came back from his vacation, I gave the drafts for his perusal and study, and eventually he agreed to submit to some journals. I had said that I would like to submit one to the journal on Problems of Control and Information Theory (that was published by some joint society from Russia, Hungary, and one more) to which the guide told me that prof. Pugachev is one of the associate editors of the journal, and he was reluctant to submit the paper to that journal; but I had insisted, and then he agreed to submit the paper.

The other theoretical paper was submitted to the Canadian electrical engg. journal (Canadian IEEE).

During the initial phase of my study of PNF, I had written two sort of review papers that were presented by the guide in an IFAC conference on stochastic control in Russia.

*

So, I had four conference papers (IEEE, IFAC), and three journal papers from my work during the doctoral program from the year September 1981-April 1986.

***

Performance evaluation of some filtering algorithms

It was very important that some validation of the performance of the filtering algorithms be carried out, so the users have confidence in application of these for the real-life systems. For this I followed the M.E. thesis copy (of P. Chodas) that I had obtained from UTIAS, this report had given complete details of the procedure on what to simulate for the orbit determination problem, and this itself was an enormous task; I had to study several things before I could really write a software for this: i) first study all the aspects of satellite orbit determination, ii) study the coordinate transformations, iii) visualize how a satellite would be rotating around the Earth and generating the data, iv) study the search and rescue satellite system (SARSAT), v) study all the equations of motion of a satellite, state and observation equations, vi) study the programming language PASCAL, since my guide insisted that I code the equations in this language, vii) write the code and submit for running on the main frame computer of the McMU; it had two such systems, and for consistency of the results I had run the codes on these two computing systems.

*

It had taken nearly two months for me to get some sensible results of/on: i) application of UD-EKF (i.e. EUD, extended UD filter) in a simple basic study, ii) study of effects of some coordinate transformations, and order of utilization of the observables for satellite orbit determination, and iii) application of a linearized filter and the study of effect of direct and cross random noise; because the PNF was highly nonlinear and I did not know how to directly use this filter for orbit determination. Since, all the performance results were highly satisfactory, I could publish some papers from this simulation validation work.

*

My earlier experiences of conducting research at MSU in Vadodara, and at NAL in Bangalore were highly useful in generating correct and well-validated results that would give confidence in the applications of any newly devised filtering algorithms, especially to a complex problem of satellite orbit determination.  

***

The final doctoral committee meeting

When my guide was satisfied by the progress of my theoretical work, he had called the writing permission meeting, for which I had prepared the progress report, and in the report I had even written the tentative draft of the table of content of my thesis; by which the committee members very much impressed, and did not even ask me to make any presentation of the final progress, and they even told me to give them the copies of my report, so that they could give to their own doctoral students to follow the style so that they would get their writing permission easily; to which my guide was very happy and then had started supporting me, although he had said to me earlier that he would not speak in favour or against me.

**

My doctoral viva/defence had taken place on 26th March 1986, and due to some sad personal reasons, we had returned to India on 29th April 1986. An Indian student-colleague and friend of mine had completed his work later on, and had brought my Ph.D. degree certificate and the bound volume copy of my thesis with him, and had posted these from his place in Dehradun to me in Bangalore.

***

I have put much of my work done as a part of my doctoral thesis in my books on the related subjects, that I have had published during 2007-2017.

*****