Analytical and DNN-Aided Performance Evaluation of IRS-Assisted THz Communication Systems

This paper investigates the performance of an intelligent reflecting surface (IRS)-assisted terahertz (THz) communication system, where the IRS facilitates connectivity between the source and destination nodes in the absence of a direct transmission path. The source-IRS and IRS-destination links are subject to various challenges, including atmospheric attenuation, asymmetric $α$-$μ$ distributed small-scale fading, and beam misalignment-induced pointing errors. The IRS link is characterized using the Laguerre series expansion (LSE) approximation, while both the source-IRS and IRS-destination channels are modeled as independent and identically distributed (i.i.d.) $α$-$μ$ fading channels. Furthermore, closed-form analytical expressions are derived for the outage probability (OP), average channel capacity (ACC), and average symbol error rate (ASER) for rectangular QAM (RQAM) and hexagonal QAM (HQAM) schemes over the end-to-end (e2e) link. The impact of random co-phasing and phase quantization errors are also examined. In addition to the theoretical analysis, deep neural network-based frameworks are developed to predict key performance metrics, facilitating fast and accurate system evaluation without computationally intensive analytical computations. Moreover, the asymptotic analysis in the high-signal-to-noise ratio (SNR) regime yields closed-form expressions for coding gain and diversity order, providing further insights into performance trends. Finally, Monte Carlo simulations validate the theoretical formulations and present a comprehensive assessment of system behavior under practical conditions.

💡 Research Summary

**
The paper presents a comprehensive performance study of an intelligent reflecting surface (IRS)‑assisted terahertz (THz) communication system in which the direct link between source and destination is blocked. The authors model the two‑hop link (source‑to‑IRS and IRS‑to‑destination) by incorporating three dominant impairments: (i) atmospheric molecular absorption, (ii) asymmetric α‑μ small‑scale fading, and (iii) beam‑pointing errors caused by misalignment of highly directional THz beams. Atmospheric attenuation is expressed through Buck’s empirical formula, yielding an exponential attenuation factor hₐ that depends on temperature, pressure, and humidity. The α‑μ fading is allowed to be different for the two hops, characterized by parameters (α₁, μ₁) and (α₂, μ₂). Pointing errors are modeled with a Gaussian‑beam based PDF parameterized by φᵢ and S₀ᵢ for each hop.

To capture the statistical behavior of the composite IRS‑reflected channel, the authors employ a Laguerre series expansion (LSE) approximation. Unlike the conventional central‑limit‑theorem (CLT) approach, LSE remains accurate even when the number of IRS elements N is moderate, and it can handle the joint effect of non‑identical α‑μ fading and pointing errors. The end‑to‑end signal‑to‑noise ratio (SNR) λ is expressed as

λ = (Pₛ/σₙ²)·hₐ·N·∑{j=1}^{N}∏{i=1}^{2}h_{f,ij},

where h_{f,ij} denotes the small‑scale fading‑plus‑pointing‑error term for the i‑th hop and j‑th element.

Closed‑form analytical expressions are derived for three key performance metrics:

1. Outage Probability (OP) – obtained from the cumulative distribution function (CDF) of λ, which is expressed in terms of Meijer‑G functions after applying the LSE. The OP formula explicitly contains the α‑μ parameters, pointing‑error coefficients, and the atmospheric attenuation factor.

2. Average Channel Capacity (ACC) – derived by integrating log₂(1+λ) over the PDF of λ. The result is represented using multivariate Fox‑H functions. In the high‑SNR regime, the expression simplifies, revealing a diversity order

   d = (α₁μ₁ + α₂μ₂)/2

   and a coding gain

   G_c =