WLS-Based Interference Identification and Shape Factor Estimation for SINR Distribution Modeling in URLLC
Abstract
The evolution toward 6G networks demands deterministic guarantees for Ultra-Reliable Low-Latency Communications (URLLC), motivating precise modeling of the Signal-to-Interference-plus-Noise Ratio (SINR) distribution. We propose Weighted Least Squares (WLS)-based algorithms that leverage uplink SINR data to jointly identify large-scale interference and estimate the Nakagami-<inline-formula> <tex-math notation="LaTeX">$m$ </tex-math></inline-formula> shape parameter <inline-formula> <tex-math notation="LaTeX">$m$ </tex-math></inline-formula>. The approach achieves accurate interference identification and robust <inline-formula> <tex-math notation="LaTeX">$m$ </tex-math></inline-formula> estimation, thus enabling closed-form SINR distribution modeling. Simulation results demonstrate strong goodness-of-fit in characterizing the SINR distribution, thereby laying a foundation for future deterministic URLLC performance analysis.