Random signals and parameterization

During my internship at IBM, I was working on a process of verification of hardware implementations by injecting signals into the design under test. Here the verification engineer has to be smart enough to guess the potential breakdown points of the hardware and inject those signals. To further test for robustness, random signals are injected. For another level of robustness, partialy random signals - biased towards the "guessed" potential breakdown points.

The whole idea when applied to life, seems to make so much sense that I was startled when I first thought about it. To test other persons' reaction, dont we drop hints suggesting things that we are not sure of the reaction from the other person?

This same thing, in a different setting sounds so much like a pre-product launch market survey. "Have a look at the need to be catered to(hardware) - Guess the ways which the market may respond in the way we want it to(guess the potential breakdown signals) - Give the market some products that hover around the previous guesses, see how the market reacts(Random signals biased towards the guessed potential breakdown points) - All this is done before the final product (chip) is launched into the market"

Fascinating, how concepts learnt in one sphere of life can be used in other spheres as well.

Smart grids and the concept of Distributed Generation

Here is something I wrote for the NERD (Notes on Engineering Research and Development) magazine. This magazine aims to motivate Indian students to write about their field of interest and research. It has got contributions form all the IITs plus a few from Indians outside India.



Distributed Energy generation typically refers to a multitude of small generators instead of having one huge generator that takes care of the entire load.

A smart grid is defined as one which integrates advanced sensing technologies, control methods and integrated communications into the current electricity grid.

So whats so different about these modern grids as compared to older grids? - The major difference is the fact that it is designed to deal with both ways power, rather than the traditional design where the grid just gives out power.

Comparing only the power delivering part, at the transmission level today’s grid is efficient, smart, intelligent. Its at the distribution level that there is a difference. At the distribution level and at the customer level, there are opportunities for automation, intelligent appliances, advanced data collection networks. This in part is because there was no need previously as there was little demand matching done. The Smart Grid makes it possible to integrate large scale intermittent generation through demand response.

The typical components of a smart grid are :-

1) SCADA (Supervisory Control and Data Acquisition), PMUs (Power Management Units), FACTs (Flexible AC transmission systems), Advanced Conductors - At the generation, transmission, substation levels. These things help in maintaining the power quality, reliability and efficiency .

2) Substation Automation. - Helps in resource utilization and demand matching.

3) Distribution Automation, MicroGrid. - Enables synchronization of distributed generators.

4) Advanced metering, Demand response, and distributed resources - at the customer end.
- Helps in demand matching at the consumer level. Ensures correct pricing as there is power both in and out of a building.(As the building also has a generator, when it is producing more than it needs, it sells power to the grid and when in deficit, it buys from the grid.)

So what does it achieve?

1) Reduce peak demand by actively managing consumer demand: The ratio of available appliances and equipment that can respond to both consumer and grid operator priorities continues to grow. Because these grids can manage power both out and in the grid, it will reduce the need for power, especially during high-use periods. like hot summer afternoons when the cost of producing and delivering power is extremely high.

2) Balance consumer reliability and power quality needs: Although some uses of electricity require near perfect reliability and quality, others are almost insensitive to these needs.
For eg- A device working on a resistor heating up or a motor rotating, does not really care a lot about the quality. But a device using electronics, needs to care more about the quality of power in. It cant afford to have a lot of frequency changes or voltage sags or swells.

Similarly, there can be some critical loads that need a very reliable power - Like a server or some central controller type thing cant afford to go off.

A smart grid will be able to distinguish the difference and adjust power reliability and quality accordingly at an appropriate cost.

3) Mine energy efficiency opportunities proactively: A smart grid will furnish consumers and utilities with accurate, timely, and detailed information about energy use. Armed with this information, one can identify ways to reduce energy consumption with no impact on our safety, comfort, and security.
This would mean that just be managing our demand and supply better, we can reduce the total amount of energy required. This will help us gain some understanding and insight into how our energy use affects our environment, and economy.

4) Improve overall operational efficiency: A smart grid is automated, and smart sensors and controls are integral to its design and operation. This will help the grid operators to easily identify, diagnose, and correct problems, and will even have the capabilities to anticipate problems before they happen.

5) Seamlessly integrate all clean energy technologies: Clean energy is so central to the idea of sustainable development that it cant be left behind, especially by a modern technology.Roof-top/side-wall solar systems, wind farms, small community hydro-plants and storage devices will become a fundamental part of the grid. These clean energy technologies will generate not only energy and power, but perhaps more importantly save on the fuel consumption.

More popular are the ones more talked about

Just read this interesting article which says that people look for common grounds to talk rather than talking about the better things. This is in a way a reiteration of "Matthew Effect" (The rich get richer, the poor get poorer).

But this research being done in the social domain, could mean a lot about the way marketing is perceived. The gurus knew all along that it works, but to be proved that it does work is a different thing.

Understanding this concept can help a lot in marketing, especially on the web. If somehow one could get their product to be discussed on the "right" forums - it automatically starts generating traffic. If you pay a battery of people to discuss your product on twitter, and it features on the twitter search list, it is very likely that other people will start discussing it - as it will seem like the "IN" thing. Or if you can get your product discussed on slashdot or technorati, it is very likely that it will be discussed beyond that as well.

In fact this love for likeliness is seen in at the school and college level : A popular guy/girl is talked more about, and he/she gets more popular. If a group of 5 people start saying out loud that a certain dress is cool, the whole school gradually comes to accept the same as cool. So it can be used in adolescent marketing too. FMCG markets typically have a huge chunk of the demand coming from teens and just-left-teens . To be able to project the product as cool, they need to be able to make an impression at school, colleges, FaceBook, MySpace - and in all likelyhood, it will catch up.

भारतीय प्रोद्योगिकी संसथान, मद्रास की पत्रिका "दी फौर्थ इस्टेट" आपकी रचनाएँ आमंत्रित करती है।

ये बहुत ख़ुशी की बात है कि "आई आई टी मद्रास" की पत्रिका "दी फोर्थ एस्टेट" अब ब्लॉग के जरिये अपने पाठकों से संपर्क बना सकेगी। अब तक कई विषय जो पत्रिका के मुद्रित होने के पहले टिपण्णी का इन्तेजार करते हुए दम तोड़ देते थे, अब इस ब्लॉग में जगह पा सकेंगे। नया सम्पादकीय दल आगे के संस्करणों को और अधिक रोचक बनाने से प्रयास में लगा हुआ है। उम्मीद है कि इस बार भी हम पहले से बेहतर होते "दी फोर्थ एस्टेट" के हिंदी खण्ड को नई दिशा में सफल रहेंगे।

"दी फोर्थ एस्टेट" अपने अगले संस्करण के लिए रचनायें आमंत्रित करता है। रचनायों केप्रकाशन हेतु प्रस्तुतिकरण पर कोई बंधन नहीं है।

कोई भी बात हो, गुस्से के साथ हो,
झूझते जज्बात हों या प्यार की तकरार हो
कहानी या लेख हो या डेली सोप हो,
जीने की तलब हो या मौत ही बैखोफ हो
प्रोजेक्ट की बात हो या हॉस्टल की रात हो
राजनीति का खेल हो या कोर कोई बेमेल हो
शास्त्रा के शस्त्र हो या सारंग के रंग हों
हो शुरू ले के कलम छोड़ के सारे भरम
फोर्थ इस्टेट के नाम कर लेख अपने ताजे गरम

आप अपनी रचना झटपट tfe.hindi.iitm@gmail.com पर भेज दीजिये. यदि आपको लगता है "दी फोर्थ एस्टेट" में कुछ कमी है तो अपने सुझाव तुंरत मेल कर डालिए.

आपके इन्तेजार में
आदित्य अंकित सत्येन्द्र

PS- For people for could not read the above few lines, dont despair. It was an announcement calling for entries for the Hindi (national language of India) campus magazine of IIT Madras.

PPS- For people who smell plagiarism, let me remind you that I am one of the hindi editors.

Different career options

This is something I wrote for the kids in school who want to choose a career which is non-traditional. As the academic coordinator for my school alumni association, We have posted it on my school notice board so to help students there.

I chose the path less traveled.

1) Art manager - One problem with Artists all over the world is that they are poor at marketing their art. This often means that they dont get the kind of money while they are alive as they get later on. As an art manager, you will have to take care of an artist's business. So that it allows him/her to make best use of his/her time while not worrying about a source of income.

This would require an understanding of the art and good convincing powers to get buyers to exhibitions.

2) Physio - Sports persons need a physio to look after their daily exercise and training while returning from an aliment.

This would require a good understanding exercises and what diet helps in what function besides some idea about the load a injured body can take so that it recovers fast.

3) Safety Engineer - Buildings often require someone who can certify that the building is safe for its designated use. All buyers of residential flats are advised to look for such a certification about its safety. Likewise, In an industrial setting, Companies often want to make sure that their building is safe to work in. That involves besides the safety of the structure, the environment at the place.

This requires knowledge about protection measures, lighting and ventilation requirements for residential and work usage, building structure testing and evacuation plans.

4) Environmental effect analyzer - There is a growing concern about human activity's detrimental effects on the environment. An EEA will have to look at different actions being taken by corporates and advising them so that the adverse effects are minimal.

With Copenhagen conference in less than six months from now, expect such jobs to be on a boom very soon.

This would require an aptitude for understanding various factors which affect the local and global eco-system, and more importantly, an ability to transform practices from worse to bad.

5) Energy Auditor - Energy costs are soaring right now. Everyone wants to cut down on their electricity wastage. In such situations, an energy auditor can pitch in with information on where the bulk of the wastage is happening and ways to reduce it.

An energy auditor to be effective will have to be a jack-of-all-trades. A person who has a working knowledge of various electrical and mechanical equipments, operations, and is aware of the various technologies available to solve a particular problem.

6) Web 2.0 Evangelist - The internet has revolutionized the way the world does business today. Your local shopkeeper can tell you about the latest offers at his place, you buy things from half way across the globe with the click of a button.

Businesses that fail to realize this potential, are losing their advantage. Most companies are awaking up to the reality that web 2.0 is, and this means that they need someone to promote their brand and business on the web - through twitter, facebook, youtube etc.

Passion for internet technology, and a knack for advertising without irritating the viewer is all that is required. On the net, the consumer has much more control over content than it the more traditional methods of advertisement - thats what makes it more challenging.

7) Event manager - Organizing an event is no mean job. One needs to ensure that everything is in place when its required. Besides the onus of publicizing the event also falls on the event manager.

People who know how to deal with people from all walks of life, can handle pressure and know how to get resources when in need are ideally suited for such a job.

8) Compering - This is the job for the fun-loving! Hosting an event, keeping the audience involved, pulling the crowd!

But, this is not for people who are shy and reserved. Only someone who gets along with people very well and fast, is a quick thinker and is spontaneous can make the cut here.

9) Architect - The infrastructure in India is expected to grow rapidly in the next few years. For that, we would need a lot of architects who can come up with engineering marvels that the country requires.

Creativity, along with building fundamentals is the quality to look for here.

The little book of Flow

I generally don't post an entry here about something I read somewhere else unless I have something to say about it. But this time I am writing this just so that you guys and read it too. I read this 'essay' "The little book of flow" by Nick Smith, based on Mihaly Csikszentmihalyi's celebrated work - " Flow: The Psychology of Optimal Experience"

It seems as though you have your entire life's purpose described here. Just have a look. And yes, I don't really have anything to say about it except that its just too awesome.

Functionality is NOT the king!

I was thinking, what makes a product a hands down market leader and an also-ran. One thing I realised was that functionality is not the only criteria. If it were the same, FriendFeed would be much bigger than Twitter. Or for that matter, LinkedIn wouldn't be the market leader either. Or Google might have been oust by one of the many search engines that came up after it.
Same is true for mobile phones too. If functionality was the only criteria, the lot of "prettier" phones would have been replaced by the more "functional" ones. iPhone might just have gone extinct if it were only to be compared on functionalities with the blackberries. I had read somewhere that SAP proudly stated that they never show the product during the selling stage. Reason - because "nobody ever got sacked for buying SAP " !

The race for functionality is never ending. You could go on adding features to your product. It will never reach the marketing stage in this way. Functionality is not the main thing and the marketing people know it too! The product has to be only "good enough" on functionality. The rest is about the acceptability, portability and the perception. "If the whole industry uses SAP, then it must be good enough". "All my friends use Twitter, I don't want to use something else."

If it is all about functionality then it is an argument about diminishing returns, marginal differences and small variations. And often this is not nearly enough to win a client, cause a revolution, or shift opinions.


Frankly, as long as our needs (present, and foreseeable in the near future) are met, do we really care what more is on offer? After this point, perception takes over. If something is viewed as the "poor man's cow", it will not be bought by someone who wants to project a rich image. If google is the in-thing, why would someone use AOL?

The next thing is the change. The trouble of moving from one product to another. People love familiarity. Getting people to switch without showing them the huge advantages is not going to work. That's the reason that analysts claim that despite Wolfram Alpha's great features, it might not be able to overtake google. For the cost and disruptions would be too huge given the present slight edge in functionality.

Suicide Vs Murder

Of late, I have been having many murderous thoughts. And I really dont know how to get rid of them. So this post, so that I can vent out the thoughts.

I remember having discussions with my friends about whether or not one should commit suicide in the event of insurmountable difficulties. I held the view that one should rather murder than commit suicide. But now I see why people choose suicide over murder.

So what is it that invokes suicidal thoughts in humans? When a person is subjected to double talk - when on the surface everything seems very friendly and cordial but dig just a little bit and the irony glares on your face. In such a situation, the person might find it disturbing to kill someone who has been so friendly. But the person is being subjected to utter frustration of not doing anything and no one being able to tell him why. In such a situation, where the guilt of killing someone else is substantial but the person finds himself in an situation frustrating to death - the person can find suicide a better option.

The next big thing on the net

Someone directed me to a link to the new project by Wolfram people called Wolfram Alpha saying that it was the next big thing on the net. My initial reaction to the claim was - "Ah, yet another big claimer!"

But 5 minutes into the site, and I was taken over by it. This project seeks to integrate all knowledge into one. You can put in any query and it gives you the "right" results. Much unlike Google, which directs you to further sites that might or might not (though they generally do) contain relevant information - Alpha gives you numbers and statistics about anything that can be represented by numbers. Like you could key in GDP of South Africa - And know all that you wanted to know in terms of numbers. Likewise if you key in "GE" in the stocks section, you could see how the trend for GE stocks has been for the last 1 week, 1 month, 1 year and 5 years! besides it shows you its market capitalization and the beta. If you search for AIDS deaths, then you get statistics for deaths across the world.

This thing could indeed become the next big thing in the net!

PS- The afore mentioned friend was Abhishek Gandhi, fondly(?) known as Tool here at IITM. He claims his connect to the first family of the country, but we are all too used to people who fake surnames to gather mileage !

Of modelling Human beings and pseudo random number generators

While have a chat with my friend Abhishek over my previous post on irrational behavior of humans, I started thinking on how and if humans can be represented by random numbers (Afterall each one us is unique and essentially random).

I believe (as do a lot of evolutionary scientists) that our reaction to any stimulus depends on a lot of factors, like the genes we carry, the natural environment we have lived in, the kind of thoughts we have, our experiences, stimuli from the recent past etc. Evidently, there are far too many variables that can affect our decisions.

So is there a way to this apparent madness ? Can we call these effects totally random? Abhishek sure seems to think so. But I doubt it. If everyone had totally random characteristics, then we would have had a zero (neutral) average reaction to any stimulus. But, on the contrary, masses behave more in unison. So the seemingly random "personality trait numbers" have a non-zero mean. But this mean can only be seen in a "large enough" group.

So can traits of human beings be thought of as being generated by a pseudo-random number generator? After all, these can have a mean, are random for all practical purposes but everyone knows that these have been derived from some determinstic method. I really dont know. But the idea itself seemed too good not to be posted