About this site:
This is a website solely dedicated to my MSc Project where I am attempting to keep a log of progress and manage my files, bookmarks and problems efficiently. Please read further for more information about the project. The website/blog is still a work in progress and is mainly used for my own benefit rather than anyone else! To know more about me please visit the about page on my personal blog.
About the Project:
Sampling theory dictates that any signal must be sampled at a rate that is at least two times greater than the maximum frequency content of the signal. This offers certain limitations for very high frequency signals – mainly the requirement of sampling at very high rates and the amount of data generated as a result of sampling. After sampling, the data is usually compressed e.g. a JPEG image represents one such compressed form. Thus after sampling at high speeds, data is compressed (and lost too). This represents an inefficient use of resources. A technique developed at Rice University, USA, called Compressive Sensing/Sampling is a method to perform sampling and compression simultaneously, at a much reduced rate! (For more info on Compressive Sensing, please visit http://dsp.rice.edu/cs).
My project requires the development of a hardware of very low power consumption that can perform Compressive Sensing on EEG signals. The hardware is to be connected to a source of signal (of course), sample, condition and process the signals (at a much reduced rate) and store the data on a SD card. The sampling and processing part is where Compressive Sensing (CS) kicks in and requires some algorithm to be implemented which I’m a bit unsure of at the time of writing this page. The data saved on SD card is to be reconstructed on a PC using the complex mathematical way described in the CS papers.
The microcontroller to be used in the project is Texas Instrument’s MSP430F5438, a 100-pin, very small and ultra-low power MCU.
That’s the bigger picture of the project. The progress and implementation details shall be posted in a Twitter-like format as and when they happen. Feel free to chip in with comments.