A Token Scheduled High Throughput Multi-Channel Data Collection Protocol for Wireless Sensor Network

We present a high throughput data gathering protocol for wireless sensor network. Different from existing works we elaborate a token-scheduled multi-channel TDMA protocol named TKN-TWN that provides better scalability and topology adaptability. TDMA is more suitable for high throughput data collection than CSMA. However, TDMA-based protocols require complex scheduling of the transmission time slot. The scheduling burden is a primary stumbling block of topology adaptability. To ease the scheduling burden while exploiting the advantages of TDMA, we propose to use two tokens arbitrate transmission and associate the ownership of tokens with transmission slot assignment toward throughput optimization. Due to the simplified scheduling based on tokens, TKN-TWN is able to provide scalability and topology adaptability. In addition, TKN-TWN allows burst transmission by leveraging the contention-free nature of TDMA, and sustains high throughput with burst data transmission when sensor nodes have similar traffic pattern. An extensive simulation study indicates that TKN-TWN achieves robust data delivery even in densely deployed large scale sensor networks. Experimental results on Tmote Sky show that TKN-TWN is able to provide throughput over 10 KByte/s, an improvement of about 23% over a prior work.