Supporting Data-Intensive Wireless Sensor Applications using Smart Data Fragmentation and Buffer Management

• Main Author: Masilela, Mbonisi
• Format: Others
• Published: VCU Scholars Compass 2007
• Subjects: wireless sensor; programming; data transmission; NesC; Java; protocol structure; internet; Computer Sciences; Physical Sciences and Mathematics
• Online Access: http://scholarscompass.vcu.edu/etd/779; http://scholarscompass.vcu.edu/cgi/viewcontent.cgi?article=1778&context=etd

Recent advances in low power device technology have led to the development of smaller powerful sensors geared for use in Wireless Sensor Networks. Some of these sensors are capable of producing large data packets in a single reading. This becomes a challenging problem given the constraints imposed by current MAC and Transport Layer implementations since a single data packet can exceed the MTU of the protocol stack. Little has been done in the way of addressing this issue in Wireless Sensor Networks. This paper proposes a novel solution to this issue. Proposed is a Lightweight Data Transportation Protocol that uses smart data fragmentation and efficient pipelined transmission and buffer management schemes to solve this problem. The methodology outlined in this paper ensures that data is successfully transmitted from source to destination with minimal delay or packet loss.