A cross-layer data gathering scheme for heterogeneous sensor networks based on polling and load-balancing

In this paper, we propose a contention-free polling protocol and a load-balancing algorithm for two-layered heterogeneous sensor networks, based on the energy consumption features of the physical layer, MAC layer and network layer. By introducing some powerful nodes, sensors can be organized into clusters. Each powerful node works as a local data sink and gathers data from sensors in the cluster. We first analyze the power consumption model of IEEE 802.15.4-based transceivers. Based on physical layer features of such low-power, low-rate transceivers, we design a contention-free polling protocol for inner-cluster data gathering, so that packet collision can be avoided. By deactivating the transceivers of sensors which do not need to transmit or receive packets, energy consumption on idle listening can be saved. We also propose a load-balancing algorithm at network layer to maximize the network lifetime by taking into consideration of power consumption features at the physical and MAC layers. Simulation results show that the proposed scheme can achieve more than one order of magnitude improvement compared to the FLOW-CSMA combination and reduce the duration of sensor duty cycle by at least 50% under various offered traffic loads.