Optimizing the Phase Shifts of RIS by MM Algorithm in RIS-Aided Positioning Systems

In Reconfigurable intelligent surface (RIS)-aided positioning systems, a proper design of RIS phase profile is important to fully explore the potential of RIS. Generally, the position error bound (PEB) is used as a metric of positioning performance, and the phase shifts of RIS are optimized to minim...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2023 3rd International Conference on Intelligent Communications and Computing (ICC) s. 168 - 173
Hlavní autoři: Hong, Sheng, Liang, Mengdan, Zhou, Rumeng, Xu, Pengzhen
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 24.11.2023
Témata:
Complexity theory
MM algorithm
Optimization
Parameter estimation
position error bound
Positioning
Real-time systems
reconfigurable intelligent surface
Reconfigurable intelligent surfaces
Signal to noise ratio
Transforms
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In Reconfigurable intelligent surface (RIS)-aided positioning systems, a proper design of RIS phase profile is important to fully explore the potential of RIS. Generally, the position error bound (PEB) is used as a metric of positioning performance, and the phase shifts of RIS are optimized to minimize the PEB. However, the optimization for PEB is very complex and time-consuming, because (1) the expression of PEB is complex and not compact, (2) the phase shifts are usually time-variant to enable the parameter estimation, (3) the number of RIS units is large. To ensure a real-time positioning system, it is essential to develop a low-complex optimization scheme. Thus, in this paper, we transform the minimization of PEB into the maximization of signal-to-noise ratio (SNR) and propose to optimize RIS phase shifts by the majorization-maximization (MM) algorithm. It is shown that optimizing the SNR can achieve the purpose of reducing the PEB, and the optimization complexity is significantly reduced at the cost of negligible performance loss. Moreover, the superiority of the proposed optimization scheme becomes pronounced when the time slots and RIS units are numerous.
DOI:10.1109/ICC59986.2023.10421115