電子情報通信学会ソサイエティ大会講演要旨
B-5-94
Evaluation of Fast Connectivity Prediction by using RSSI and SNR in LoRa
○Guibing Zhu・Chun-Hao Liao・Theerat Sakdejayont(The Univ. of Tokyo)・Masayuki Iwai(Tokyo Denki Univ.)・Yoshiaki Narusue・Hiroyuki Morikawa(The Univ. of Tokyo)
In order to conduct a fast connectivity prediction in LoRa, we evaluate the feasibility of using a small number of packets with SF7 to perform accurate connectivity prediction of all SFs. First, our experiments show that the conventional link indicators, i.e., Received Signal Strength Indicator (RSSI) and Signal-to-noise ratio (SNR), are inconsistent over both time instant and SFs. Second, we strive to clarify: 1) a number of packets which is required for accurate prediction overcoming the fluctuation in time and 2) the degradation in the prediction when using SF7 for predicting the connectivity of all other SFs. To this end, we suggest the necessary number of packets for accurate prediction while keeping a low overhead.