Year: 2026
Type: preprint
Abstract: This paper examines the performance degradation of receivers caused by imperfect channel state information (CSI), arising from training-based and semi-blind channel estimation techniques. We focus on the uplink of massive multiple-input multiple-output (MIMO) systems operating in time-division duplexing (TDD) mode, considering the asymptotic regime where the number of base station (BS) antennas grows large while the number of users remains fixed. Leveraging random matrix theory (RMT), we analyze the asymptotic behavior of the signal-to-interference-plus-noise ratio (SINR) under linear MRC and MMSE detection schemes. We derive closed-form approximations of the achievable rates, prove their asymptotic tightness, and validate their accuracy through simulations for systems of finite dimensions. Numerical results under realistic non-terrestrial network (NTN) conditions further demonstrate that semi-blind channel estimation enables MRC and MMSE receivers to outperform their training-based counterparts, while also quantifying the number of pilot symbols required to achieve a desired SINR level.
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Keywords: non-terrestrial networks, massive MIMO, uplink, SINR analysis, imperfect channel state information, semi-blind channel estimation, training-based channel estimation, maximum ratio combining, MMSE detection, random matrix theory, asymptotic analysis, achievable rate, pilot overhead, time division duplex
Open Access status: green
OpenAlex ID: W7226015847
