Friday, May 28, 2010

last week of May 2010

This week was a pretty busy week with less of research work and more of documentation. This week was spent on data pooling from the network. This activity was carried out to measure performance of the network and the motes in different modes of operation.
Micaz motes are programmed into distinct groups of operation. The first group of motes are programmed to measure ambient temperature, second group to measure vibration and sound intensity of transformers and circuit breakers, third group programmed to measure the surface temperature of equipments and the fourth group of motes to detect SF6 gas leaks.

The motes can also be classified based on the frequency of packet forwarding. The high forwarding or periodic motes are the ones who transmit data packets every 15 minutes. The low forwarding or level crossing motes are the ones who transmit packets only if any parameter (i.e. temperature or vibrations) increase above a certain preset threshold.

The data pooling activity involved recording difference or drop in battery voltages of motes depending on their location or proximity to the base station and a comparison of voltage drops in periodic Vs Level crossing motes. As expected the periodic motes will have a higher voltage drop as compared to the level crossing one's because of more number of packet transmissions. The activity also involved measuring the solar irradiation in the region so as to speculate the viability of "heliomotes" which convert solar energy into usable electrical energy. The final activity involved a comparative study of the average number of packets transmitted by motes in each of the region (region depends on proximity to the base station). The motes which were at a greater distance from the base station were seen to transmit slightly lesser number of packets than the ones near to the base station. The reason is probably because packets have to traverse greater distance over multiple hops. Even though the Xmesh protocol exhibits reliability of transmission, some packet losses are involved over multiple hops.

No comments:

Post a Comment