Modeling and Analysis of Energy Harvesting Sensor Nodes in IEEE 802.15.4 Protocols

在物聯網中，裝置的大小限制了本身能夠儲存的電量，使得網路的 壽命受到影響，而藉由能量收集，讓裝置能夠自行充電，被視為一種 延長網路壽命的方式。但是因為環境的影響，有別於以往用電池供電 的方式，能量收集的電量大小會隨著時間變動，進而影響無線感測網 路的運作，因此，以能量收集為供電方式的無線感測網路的效能需要 被重新評估。在這篇論文中，我們嘗試用模型去描述使用能量採集來 充電、並採用IEEE 802.15.4 標準下的分槽式載波感測多重存取/ 碰撞 避免(CSMA/CA) 為MAC 協定的裝置行為，並以平均充電時間、延遲 和吞吐率做為效能評估的標準，我們提出的馬可夫模型成功的描述了 通道競爭和充電的過程，並且比較使用不同的能量收集技術(例如太陽 能、振動能量收集等) 的裝置效能。模型會藉由模擬進行驗證，而分析 出的吞吐率結果和模擬結果的誤差不會超過6%。In the Internet of Things (IoT), the size constraint of those small and embedded devices limits the network lifetime because limited energy can be stored on these devices. In recent years, energy harvesting technology has attracted increasing attention, due to its ability to extend the network lifetime significantly. However, the performance of IoT devices powered by energy harvesting sources has not been fully analyzed and understood. In this paper, we model the energy harvesting process in IoT devices using slotted Carrier Sense Multiple Access with Collision Avoidance (CSMA /CA) mechanism of IEEE 802.15.4 standard, and analyze the performance in terms of charging time, throughput and delay. Our new model successfully integrates the energy harvesting process and binary backoff process through a unified Markov chain model. Finally, the new model is validated by simulation and the throughput errors between simulation and analytical model are no more than 6%. We demonstrate the application of the model with different energy harvesting rate corresponding to different sources such as solar and vibration energy harvesters