Saket Vora

Update: Vimeo has a great way of letting authors create ‘channels’ for videos that center around a theme. You can follow all EE 414 related videos here: http://www.vimeo.com/ee414. How about that…they even let you use a custom URL. You can even grab an RSS feed of a channel too. Well done.

This week’s lab experiment was to design and build a low noise amplifier. Most people worked on simulations, but the last couple of days people started making stuff in the lab. Here are some pictures and video.

Kamal works on his board

Andy and I work on our boards…I’m soldering a surface mount capacitor.

EE 414 – Lab 4 – LNA Design 2 from Saket on Vimeo

Jonathan and Andy scheme ways to get a better noise figure…

Andy plays the skeptic as Jonathan gets desperate trying to lower his noise figure

This is the first rev of my LNA.

Testing the LNA for the noise figure.

After two weeks of relatively chilly weather, today was nice and warm. There was a lot more activity on campus today, with Freshman Admit day going on. Accepted students are invited to come tour campus while the university tries to convince them to come to Stanford. These “ProFros” (prospective frosh…Stanford has a strange colloquial use of acronyms) seem like newbies walking campus. I almost hit three of them as their go meandering about totally oblivious to bikers.

Got rev 1 of the LNA done in the lab, but haven’t tested it. Went over to Bikiran’s place to work on our embedded systems homework and working on that a bit now. Will go back to the lab and test the LNA later. I’ll post videos of action in the lab as I get it.

A friend of mine from Sweden, Andreas Teske, went to a zoo and took some wonderful pictures of my favorite animal: the bear. Nader, inspired after seeing a collection of these photos, took them and used the photos to express the stages of life.

I encourage you to enjoy his work, if nothing else for the smile it’ll bring.

Update

[I]t just came to me as i was looking at the photos. To me it is the triple-truth of Bear, Life, and All Civilization.

Everything has these stages. The bear is a microcosm of one man’s lifetime.

And one man’s lifetime is a microcosm of the entire civilization.

And the civilization is a microcosm of the Great Bear which makes all things.
~ Nader Moussa

This brought a smile to my face and I had to post it. This is an infoviz but with another dimension added — audio. As the user hovers over areas on the map (controllable in size by the user too) they hear an audio representation of the traffic. This is a great idea because of the different ways you can encode information into audio — volume, pitch, tone, etc. For instance, if the vehicle distribution is tilted more towards larger trucks or vehicles, the low frequencies (bass) can dominate, where as smaller roads where small cars, motorcycles, etc are found you can let the high frequency sounds dominate. Traffic flow could be done with volume, etc. Combine that with the visualization of traffic nodes and bottlenecks, and you have a fairly sophisticated interface for conveying information. Imagine using this in a DOT control room for instance.

Read an in-depth interview about the project here.

In what other areas can we utilize this approach?

Cascade on Wheels: Traffic Mixer from steph thirion on Vimeo

Here is another one he did, sans audio, showing car traffic data in Madrid. Pretty sophisticated data mapping! I’m impressed. Unlike most dataviz projects, this combines nice aesthetics & what appears to be meaningful data.

Cascade on Wheels: Walls Map from steph thirion on Vimeo

::phone rings, I see the number is from Iowa::

Me: “Hello?”
Guy: “Hi is this Justin?”
Me: “No, it’s Saket. 919.244.9095″
Guy: “Oh man, sorry about that, wrong number.”
Me: “No problem.”

1 minute later

::phone rings, I see it’s the same number from Iowa::
Me: “Hey.”
Guy: “Hi Justin.”
Me: “No, it’s the guy you called a minute ago. What number are you trying to call?”
Guy: “What? Is this 919.244.9059?”
Me: “No, I’m 9095, not 59.”
Guy: “Ohh. My bad, man. Hey, you wouldn’t by any chance be an Obama supporter would you?”
Me: “Actually I am. I did some phone banking during the California primary. I’m out here in California.”
Guy: “Oh wow. California, huh? Yeah, I’m here at ECU working on the Obama campaign.”
Me: “Yeah I did my undergrad at NC State and now I’m at Stanford for grad school. Let’s hope Obama takes it all the way.”
Guy: “Yeah, definitely. Ok, I promise I won’t call you again.”
Me: “Not a problem, take it easy!”

Hehe.

On Saturday I attended the Startup School started and organized by Y-Combinator. The Startup School brings many notable speakers to talk about all things entrepreneurship, funding, the web, ideas, etc. This year’s speakers included:

• Jeff Bezos, founder and CEO of Amazon.com
• Marc Andreessen, creator of Mosaic, Netscape, Ning
• David Heinemeier Hansson, creator of Ruby on Rails
• Michael Arrington, founder of Techcrunch.com
• Paul Buchheit, creator of Gmail
• Sam Altman, founder of Loopt
• Paul Graham, founder of Y-Combinator
• Greg McAdoo, partner at Sequoia Capital
• Jack Sheridan, partner at Wilson Sonsini Goodrich & Rosati (the tech law firm)
• David Lawee, VP of Corp. Dev. at Google
• Peter Norvig, Dir. of Research at Google

There are over 500 people there who flew in from all over the country. The talks ranged from how VCs think about ventures, the legal issues of starting a company, how to get publicity, how to scale your business, and core principles can serve you well. My personal favorite was David Heinemeier Hansson, who gave a very entertaining but solid presentation on the flip side to web startups – you don’t have to be the next Facebook or YouTube or Google! Imagine if you (gasp!) charged money for your service, you could get profit! I mean, he had a picture of Eric Cartman on one slide and a lolcat reference (”I can haz scaling probs plz?”) on another. It was great. Paul Graham stressed the value of being good, Paul Buchheit described the importance of listening, and Michael Arrington gave a rather thoughtful (perhaps a tad bit melodramatic) about the troubling state of the community here in the Valley.

It was neat to see and hear from so many famous names, and during the breaks there was always a buzz of…yes…the next Facebook or YouTube or Google. Overall it was a great way to spend a Saturday.

This past Saturday, WaterPLUS presented in the final round of the Duke Startup Challenge and won $15,000 in awards. Joel, Naman, Kari, and Will gave an outstanding presentation, winning 1st Place in the Social Track ($5k) and taking 2nd Place Overall ($10k). This is a great achievement and I’m personally so honored to work with this team. We plan on using the money to fund Kari for the summer to conduct technology evaluation and perhaps start some early prototyping. See here for more information about the winners.

I had a second go at the the microstrip filter today, starting from scratch. Instead of our center frequency being 50 MHz over, I ended up 50 MHz under our target of 1 GHz. So I spent a while carefully shaving off millimeters and trying to get our center frequency correct.

I improved it about 10 MHz at a time, and miraculously enough the bandwidth started to get better too (at first it was like 75 MHz and the target is 50 MHz). Finally, I got really close…close enough that Catherine the TA said it looked great and that I was done with it.

Center frequency was maybe 1 MHz off and the bandwidth was 57 MHz. Good enough!

when I see a demo setting up by Lomita Mall. I took out my camera and shot this:

It’s the STARMAC quad rotor aircraft that the Hybrid Systems Lab in the aero/astro department is working on. Vijay, a grad student came over and filled me on what the project is about. This flyer is a test platform for multi agent control. They want to operate like 8 of these aircraft and have them work together intelligently to accomplish missions or tasks. I expressed my interest in ’swarm behavior’, but Vijay accurately explained that swarm behavior involves simple behaviors, whereas what their group is trying to achieve is highly intelligent decision making in a multi-agent setting.

Pretty cool stuff.

This morning Jess made a brunch dish I’d never had before; last night she stewed/reduced sliced oranges (rinds and all) with sugar and let them sit overnight. In the morning, she toasted bread and spread a good amount of ricotta cheese, then put a generous heap of the oranges on it. Very interesting dish!

So I was in the RF lab from 3pm to 11pm. Seriously. Spent most of the time figuring out how to design this microstrip bandpass filter (3^rd order Chebyshev with 0.5dB ripple and 50MHz bandwidth). The lab handout does a good amount of design setup, but there are some rather critical parameters left to solve. My partner had done the reading for this section while I was working on an earlier part, so when he told me the kind of calculation we had to do I went into Excel Ninja mode. We spent about 1 hour crunching through Excel to find our two needed parameters, but I started to realize that it was impossible with our setup. I went back to the lab handout and discovered a ’supplemental’ section which not only validated some calculations we did earlier, but then just gives a strange-where-did-that-come-from equation for a critical parameter we needed. I then proceeded to build and start testing the design.

Kamal came in and said “free food!” and the whole lab emptied out up the stairwell. We looked really sketchy….8 students walking briskly in a line straight for the back door to the food and making out like bandits. Didn’t even stay for the presentation it was meant for! I also later stepped out for a few minutes to ask Max about a problem we had with our embedded systems project and to bounce my idea to him.

As much work as it sounds (just the 3^rd week and I’ve been spending like ~15-20 hours a week in there), I’m having a really good time. There are around 8 of us who really have a strong camaraderie. We all help each other out, talk about all sorts of things, heckle one another’s designs, and laugh a lot. It’s a nice place to come to. The TAs are also very ‘with it’ and seem to have a good time. After leaving the lab at 11pm, Bikiran and I headed back to Blackwelder to work on our embedded system project. There were some stupid things to solve with the LCD circuit, but we got most of the issues solved so tomorrow there will be some cleanup to do.

It’s 4:10am….should get some sleep.

I attended a conference the whole day today about water in the developing world held at Stanford University. Though I didn’t originally plan to, I stayed for the whole day and I’m very glad I attended. The conference featured two tracks of panels and talks; one that focused on water technologies and social enterprise issues, the other that looked at water scarcity in terms of geography and geopolitical conflict. The best part of the conference was getting to meet a lot of terrific people and talking with them about their own projects and ideas. I met many from Berkeley, who has a very nice focus on water technologies for the developing world. In particular, I met people from the UV Tube, Haath Mein Sehat (HMS India), Vestergaard Frandsen and the LifeStraw Family, AquaStar UV, and Stanford’s program in the Entrepreneurial Design for Extreme Affordability.

Last week Mary pointed me to an article on Science and technology for water purification in the coming decades, which made a reference to a new technique of having water pass through carbon nanotubes at really high flux rates. One of the lead scientists on that paper, Olgica Bakajin, was also here. I also heard about an interesting microfluidic inspired centrifugal spiral technique for separating suspended particles from water at a fraction of the energy compared to other methods from a scientist at PARC and a scientist from CalTech who is working on nanoparticles that envelope and remove harmful molecules from water too.

I made a lot of good contacts and will be following up with them for sure in the weeks to come.

I invite readers to check out a profile of the WaterPLUS project recently featured at CollegeMogul, a site that tracks ventures and ideas from colleges around the country.

Mike Glennon of CollegeMogul found out about the WaterPLUS project through YouNoodle.