Optimum Number of Reflecting Elements for UAV-Mounted Intelligent Reflecting Surface-Assisted Data Collection in Wireless Sensor Network

In this article, focusing on the application of wireless sensor networks (WSNs) in smart agriculture, unmanned aerial vehicle (UAV)-mounted intelligent reflecting surface (IRS) is used to tackle the issue of line-of-sight (LoS) channel between a sensor node (SN) and the fusion center being blocked by obstacles. In particular, the optimal number of reflecting elements is studied under the condition that the spectral efficiency (SE) and energy efficiency (EE) are balanced when the UAV hovers to collect sensors’ data. First, the relationship between SE and EE versus the number of reflecting elements for a single SN is analyzed, and then, the conditional judgment-binary search (CJ-BS) algorithm is proposed to seek the optimum number of reflecting elements. Second, on this basis and considering that multiple SNs send data to the UAV by adopting time-division multiple access, this article further analyzes the optimum number of reflecting elements when different SNs use the same or different numbers of reflecting elements (SNRE or DNRE scheme) at different time slots. Third, to find the optimal one in these two schemes, we analyze the EE and compare them. Under the above cases, we prove that the EE is a unimodal function of the number of reflecting elements. Simulation results show that with the increase in the number of reflecting elements and the hovering altitude, the relative EE improvement of SNRE versus DNRE first increases and then decreases. Furthermore, considering that the SNRE scheme’s complexity is lower, it is a better choice for practical implementation.