Model-driven transmission power management for wireless sensor networks

Abstract

Recently wireless sensor networks (WSNs) have been deployed for several dataintensive sensing applications such as structural monitoring and habitat monitoring. Sensor nodes in these applications often sample the physical environments at high rates. Supporting data-intensive applications poses several major challenges to the design of WSNs. Due to tight power budget, radios on sensor nodes have very limited bandwidth. In addition, sensor data usually must be delivered to the sink through multiple hops. The achievable delivery rate of aWSN is thus limited by the interference among transmitting nodes. As a result, a fundamental tension exists between the sheer amount of data generated by sensor nodes and the low capacity of WSNs. Moreover, the low throughput of a network also leads to poor energy efficiency as nodes must remain active for a longer period. In an attempt to deal with the problem, this thesis first develops empirical signal decay and interference models. Based on these models, it then proposes C-MAC, a new MAC protocol designed to achieve high-throughput bulk communication for data intensive sensing applications. Nodes running C-MAC estimate the level of interference based on the physical Signal-to-Interference-plus-Noise-Ratio (SINR) model and adjust the transmission power accordingly for concurrent channel access. C-MAC has been implemented in TinyOS-1.x and extensively evaluated on Tmote nodes. The experimental results show that C-MAC significantly outperforms the state-of-art CSMA protocol in TinyOS with respect to system throughput, delay and energy consumption. Moreover, this thesis extends to study inter-channel interference models and then apply the models to both link capacity analysis and channel assignment protocols. The experimental results from a testbed of 802.15.4 TelosB motes show that inter-channel interference models not only are accurate but also incur low measurement overhead. The results also demonstrate that channel assignment protocols for WSNs significantly benefit from using overlapping channels.