Thursday 10 January 2008

Smart Dust

One of the sensor node technology is Smart Dust, it can be utilized as tiny sensor node. Smart Dust technology was originally developed in the mid 1990’s at university research institutions, the University of California, Berkeley, for defense and intelligence applications. As the research world rallied around the concept of “Smart Dust” ,the first project was conducted by Kris Pister, a Professor at UC Berkeley and founder and CTO of Dust Networks, universities and research institutions worldwide started to take a closer look at the impact of ubiquitous sensing and control. Smart Dust (Silicon Mote) can collect , analyze data and transmit them to 30m range distance by wireless
SmartDust was designed in very small Silicon Mote and integrated with SoC and solar-cell technology in early days.

Intel Mote
One component of the sensor node is low-power CPU, this component has been developed by Intel, ARM, and Motorola. Intel developed motes which use micro processor based on ARM 32bit CPU , as following image shows Intel Mote . It has 64KB RAM, 512 KB memory and 12MHz CPU clock. RF Module is based on 2.4GHz Blue tooth which is developed by Zeevo, and its transmit ration is up to maximum 720kbps.

Problems and other approach
However, the sensor node technology has still outstanding issues such as sensor size , battery life cycle, cost for installation and data accuracy. To solve these issues many research and development project is still going on. By the Dust technologies, current sensor mote size is as small as 1cent, however, it can be recognized by eye, so it can easily exposed to enemy or general human beings who may regard it as “dust” and throw it away.
Second, limitation of battery life cycle can be an obstacle for the deployment. To extend battery’s life cycle there are several approaches has been researched and developed. In initial stage, it is proposed instead of battery, that solar cell is used. However, solar cell cannot minimized mote size and not efficient in energy transformation. Since then, instead on solar cell, acoustic noise and kinetic energy by vibration between the motes are replaced.
l Solar cell
l Energy source from human body (Temperature)
l Source from building(heat)
l Kinetic Energy (by Vibration, by Motion
l Acoustic Noise

Nowadays, other approach for better life cycle is researched by communication engineering area, which is to optimize sensor network To minimize battery exchange cost, there are several efficient communication routing protocol is proposed for example, SMedia Access Control by prof. Wei Ye.
Third, sensor mote’s price is not cheap. Recently, MEMS technology has dramatically led to the development of ultra small in size, low-power, low-cost sensor devices. However, cost is shrink down from 50$/mote to 1$/mote. Still, it will cost huge amount of money to apply and create sensor network in real world.

No comments: