Abstract
The reconfigurable intelligent surface (RIS) is one of the promising technology contributing to the next generation smart radio environment. The application scenarios include massive connectivity support, signal enhancement, and security protection. One crucial difficulty of analyzing the RIS-assisted networks is that the channel performance is sensitive to the change of user receiving direction. This paper tackles the problem by categorizing the RIS illuminated space into four categories: perfect alignment, coherent alignment, random alignment, and destructive alignment. These four categories cover all the possible phase alignment conditions that a user could experience within the overall $2\pi$ solid angle of RIS-illuminated space. We perform analysis for each of these categories, deriving analytical expressions for the outage probability and diversity order. Simulation results are presented to confirm the effectiveness of the proposed analytical results.
Introduction
Reconfigurable Intelligent Surfaces (RIS) are a cornerstone of 6G research, acting as "smart mirrors" that can steer radio waves to enhance signals or cancel interference. However, most existing research focuses on the "best-case scenario"—the user at the exact target angle with perfect hardware.
The real-world challenge is that RIS performance is highly sensitive to the user's direction and the quality of Channel State Information (CSI). If a user moves slightly off-center, or if the RIS uses cheap, discrete phase shifters instead of continuous ones, how much does the performance actually drop? This paper fills that gap by providing a mathematical map of how the RIS redistributes power across the entire $2\pi$ solid angle of its illuminated space.
Phase Alignment Categories
To analyze the RIS-assisted channel, the authors move away from the standard Central Limit Theorem (CLT) approach—which can introduce systematic errors—and instead use a more robust asymptotic analysis. They define four distinct physical states based on how the $M$ elementary waves from the RIS elements sum at the receiver:
- Perfect Alignment: Ideal conditions at the target angle where all waves arrive in phase.
- Coherent Alignment: Near-target angles, discrete phase shifts, or imperfect CSI where waves are mostly aligned.
- Random Alignment: Broadcasting mode or zero CSI where phases are uniformly distributed.
- Destructive Alignment: Intentional signal cancellation (e.g., for physical layer security) where waves sum to near zero.
Performance Analysis and Diversity Order
The channel is modeled as a sum of $M$ elementary waves. To find the diversity order (a measure of how reliably the signal overcomes fading), the authors apply Laplace transforms and Taylor series expansions to the Probability Density Function (PDF) of the channel amplitude near the origin.
The spatial distribution of these categories is shown in the relative power plot below. The "Perfect Alignment" occurs at the main lobe peak, while "Coherent Alignment" covers the main lobe width. "Random Alignment" corresponds to the side lobe peaks, and "Destructive Alignment" occurs at the nulls.
The derived diversity orders ($d$) for each category are:
- Perfect/Coherent Alignment: $d = M$.
- Random Alignment: $d = 1$.
- Destructive Alignment: $d = 0$.
Key Insights and Practical Implications
- The Quantization Threshold: A practical takeaway is that high-precision hardware isn't always necessary. The authors prove that as long as an RIS has a quantization level of $L \ge 4$ (at least 2 bits of phase control), it can still achieve full diversity order $M$.
- Diversity vs. Power Gain: In the "Random Alignment" category, the diversity order drops to 1. However, the average received power still increases with the number of elements $M$. This distinguishes between "reliability" (diversity) and "strength" (average power).
- The Shrinking Beamwidth: The authors derive a "full-diversity beamwidth," showing that as more elements ($M$) are added to increase gain, the angular window for full diversity gain shrinks proportionally to $1/M$.
Conclusion
This paper provides a comprehensive framework for analyzing RIS performance by categorizing the entire illuminated space into four Phase Alignment Categories. By deriving analytical expressions for outage probability and diversity order, the authors demonstrate that while RIS can achieve full diversity gain at a target angle, this performance degrades predictably as a user moves or as hardware constraints are introduced.
Future work may extend this analysis to dense urban environments where elements might experience correlated fading, or to complex multi-user scenarios like Non-Orthogonal Multiple Access (NOMA).
Summary updated