Hierarchical codebook design for IRS assisted angular estimation of user position

Abstract

With the increase in demand for high-speed connectivity, millimeter-wave (mmwave) communication has received increasing attention from academia and industry due to its exceptional advantages. Compared to existing wireless communication tech- niques, such as Wi-Fi and 4G, mmwave communications operate at higher carrier frequencies and thus come with benefits including massive bandwidth, narrow beam, high transmission quality, and strong detection ability. Furthermore, highly direc- tional nature of mmwave help in localization of users in the network. On other hand, due to its high frequency, mmwaves are highly susceptible to attenuation and block- ages. Consequently, if the user is at a great distance from the source, then due to high atmospheric attenuation, that user is out of coverage. To deal with such scenar- ios, we have proposed a system which employs Intelligent Reflecting Surfaces (IRS), that helps in angular estimation of the positions of users that are far away from the Base Station(BS). Thus, it helps in estimating positions of users with higher accu- racy and provides better localization in a cellular network. Furthermore, we have proposed a hierarchical codebook approach to estimate angular positions of users, and we compared it with tradition methods like exhaustive search, showing how the proposed method is computationally more efficient. And, we have used Cramer Rao Lower Bound (CRLB) to observe the degradation of angular estimation performance as distance between IRS and users increase. We also used it to demonstrate improve- ment in angular estimation performance by increasing number of IRS elements. We found that IRS assisted system provides much better angular estimation of users po- sitions than a system with only a BS. Furthermore, we also found that Hierarchical codebook method is more computationally efficient than traditional methods like ex- haustive search.