The Joy of being a computer engineer lies in the amazing ability of complexity synthesis that lies at my disposal. The inexplicable sense of magic, which I felt when my first “hello world ” got displayed on screen, Is something I have never experienced in any other phase of my academic pursuit “Hello, World”; the flirts and perhaps the simplest of the programs any novice programmer executes. Yet, its seeming simplicity lent a magic-like feeling to programming, something that I have found hard to ascribe to any science.
Since all the things that made the program work were yet to e explained, I was left with an uneasiness that persisted through the many progressively complex programs that I wrote. It was during my fifth semester at college, when I was trying to make a stepper motor work In the Microprocessors laboratory, that I saw how a program could influence something in the real world. The fact that an “unreal” piece of text that I had keyed in could move something in the real world fascinated me. The sight of the stepper motor turn addressed the discomfort in a subtle but very important way.
Soon, I realized that looking at orgasm from the hardware perspective and understanding how things worked “under the hood” fascinated me and added a new dimension to my programming skills. It is this fascination for the hardware software co;design that led me to pursue a career in embedded systems development.. It was during my Junior year project that I first got a glimpse of embedded systems world. MIM task was to designed and implemented a length measuring device . I singed a 8051 controller Interfaced with a stepper motor and a TDMA keypad to move a metallic pointer to the specified distance.
I achieved a least count of 0. 01 mm. This was myth first project where I had storehouse to micro-controller programming for a complex system and it gave me an in depth insight on the workings of a basic processor architecture. Next, during my senior year, I worked with a more complex micro-controller- the TTL- MSP. This was done as a part of final term Internship at GILL Labs, an embedded start-up firm In Bangor, India. Successfully wrote the code for a standalone client which read the environment temperature (using in-built sensor of MSP430F2XX rise) and sent it to a server using a wireless protocol.
During this period I understood the RISC architecture and also got hands on experience of writing and debugging Industry level embedded code. During the first few months of working at Samsung had the opportunity to work on Java Virtual Machine during my tenure in the CDMA JAB team. During the course of this project, I got acquainted with the core VIM fundamental concepts (generic SUN JAB architecture, role of Class loader, execution of botched, virtual machine instruction set etc. ) and also worked on certain procedures for optimizing he VIM performance for speed of execution. Optimized manner of using the LCD buffer, usage of Inline functions and writing assembly Instructions for certain key Limited Device Configuration) VIM Apish onto the Qualm Mobile Platform (BREW) with REXES as the Operating System. Samsung introduced me to the world of ARM. In my time in Samsung Wireless Terminal Division, I analyses and studied almost everything about mobile devices experience at Samsung has given me exposure to almost every aspect of mobile telephony like Boot up procedure, BSP, Java Virtual Machine, Performance Analysis of
Applications , etc.. As a part of the User Interface Framework team in Samsung Linux Platform, my primary task was the optimization of the graphics pipeline for the various processor cores. The platform ran on the likes of Scales based Paxar and the Cortex-AH based Mapping. The primary bottleneck in both the devices was the DRAM since it ran several orders of magnitudes slower than the processor. However’ came up with the idea of by using pre-loading of pixels and using SIMD ,and was able to get a frame rate improvement of over 80% over the current pipeline on
Monsoons processor and over 110% on MAP. I am also currently working on the analysis of engaged in studying the cache behavior of the Mapping toss that I can maximize its pre- loading potential. This has given an opportunity to gain enhanced in depth understanding of the embedded processor architecture and its interaction with the main memory subsystem. I have also become fairly proficient with the ARM architecture. As of writing this, viable processor options for any mobile device manufacturer are either the Scale based cores or the MAP line of processors.
Having worked on both of them, I can confidently say that thinner is capabilities of theses cores in of handling most complex tasks like content-based retrieval techniques, speech recognition and AD games especially at higher resolutions like ‘VGA are quite restricted.. I feel the current scope of mobile applications is severely limited by the available processing power. I wish to explore the concept of using a massively parallel architecture on a mobile device and the various problems associated with it like power consumption and heat dissipation.
The concept would bets similar to the CELL architecture except for the fact that it would be in a much smaller form factor and it should be able to run on a portable battery. I think that such an architecture coupled with an operating system that can fully take advantage of it will truly make the mobile device a unique platform having its own class of applications rather than a smaller copy of the desktop. Apart from this project, I had the opportunity to work on Java Virtual Machine during my tenure in the CDMA JAB team at Samsung for six months.
During the generic SUN JAB architecture, role of Class loader, execution of botched, instruction set etc. ) and also worked on certain procedures for the optimizing the VIM performance for speed of execution. (Optimized manner of using the LCD buffer, usage of inline functions and writing assembly instructions for certain key sections of code etc. ). As a part of the project, I had also ported the CLC (Connection Limited as the Operating System. The firm foundation laid by my undergraduate studies and three years of industrial experience will assist me in my graduate studies.