Strategic Deployment of Reconfigurable Intelligent Surfaces

20 concepts19 papers1 proposal

Reconfigurable Intelligent Surfaces (RIS)—also referred to as Intelligent Reflecting Surfaces (IRS)—have emerged as a paradigm-shifting technology for next-generation wireless networks. By tile-mapping an environment with software-controlled passive phase shifters, an RIS can dynamically reshape wireless propagation environments, boosting signal-to-noise ratios (SNR) without the power consumption or noise amplification of traditional active relays. As detailed by Ying-Chang Liang et al. in their foundational overview, Large Intelligent Surface/Antennas (LISA): Making Reflective Radios Smart, an RIS can scale received signal power by the square of its reflecting elements ($M^2$).

However, an enduring challenge limits this potential: where should the surface physically sit? While substantial research focuses on optimizing real-time passive beamforming matrices, the physical positioning of the RIS establishes the upper bound of its performance.


The Tyranny of Product Path Loss

To understand why placement is so critical, we must look at how signals decay when bouncing off a passive surface. In a conventional direct wireless link, the signal path loss scales with distance $d$ according to a power law, roughly proportional to $d^{-\alpha}$, where $\alpha$ is the path loss exponent.

An RIS-assisted link, however, is a two-hop cascade: from the Base Station (BS) to the RIS ($d_1$), and from the RIS to the User Equipment (UE) ($d_2$). Because the surface acts as a passive reflector rather than an active amplifier, the total path loss is proportional to the product of the two link distances:

$$\text{Path Loss} \propto d_1^{-\alpha_1} \cdot d_2^{-\alpha_2}$$

This multiplicative relationship introduces a severe geometric penalty known as the "double path loss" effect.

Imagine a simplified one-dimensional deployment where the total distance between the BS and the UE is fixed ($d_1 + d_2 = D$). If we plot the total path loss as a function of the RIS position, the curve is concave. The path loss is maximized precisely when the RIS is placed at the midpoint between the transmitter and receiver ($d_1 = d_2 = D/2$).

Max Path Loss (Worst Performance)
              v
BS ----------- RIS ----------- UE
<--- d1 --->        <--- d2 --->

To minimize this product path loss, we must make the product $d_1 \cdot d_2$ as small as possible. Mathematically, this occurs at the boundaries: either $d_1 \to 0$ or $d_2 \to 0$.


The Endpoint Advantage

This geometric reality forms the core intuition behind optimal RIS placement strategies. To bypass the worst of the product path loss, an RIS should be deployed as close as possible to either the transmitter or the receiver.

Min Path Loss (Optimal)              Min Path Loss (Optimal)
         v                                    v
BS - RIS ------------------ UE       BS ------------------ RIS - UE

In Performance Analysis of Intelligent Reflecting Surface Assisted NOMA Networks, Xinwei Yue et al. demonstrate that positioning an IRS closer to the base station yields superior outage probability and ergodic rates. This proximity ensures a strong, unobstructed Line-of-Sight (LoS) channel for the first hop, effectively turning the IRS into an extension of the base station's antenna array.

Similarly, Xinwei Yue et al. extend this analysis to active surfaces in Exploiting Active RIS in NOMA Networks With Hardware Impairments, confirming that even when accounting for hardware impairments, base-station-proximal deployment remains highly resilient.

This "endpoint advantage" is further validated across multi-user topologies. In Joint Association, Beamforming, and Resource Allocation for Multi-IRS Enabled MU-MISO Systems With RSMA, Huijun Xing et al. show that deployments situated near either the access point or the cluster of end-users drastically outperform mid-link placements, directly translating geometric positioning into network-wide energy efficiency gains.


Joint Optimization: Co-Designing Space and Phase

While the endpoint rule provides a strong geometric heuristic, real-world networks are rarely one-dimensional or static. Optimal deployment requires a holistic approach that co-designs physical placement alongside real-time resource allocation and beamforming.

For instance, in Non-Orthogonal Multiple Access (NOMA) networks, users share the same frequency resources through power-domain multiplexing. In Joint Location and Beamforming Design for STAR-RIS Assisted NOMA Systems, Qiling Gao et al. examine Simultaneously Transmitting and Reflecting (STAR)-RIS systems. The authors demonstrate that optimizing the physical coordinates of the STAR-RIS alongside active and passive beamforming vectors is essential for maximizing the weighted sum-rate of the network. By shifting the surface's location, the network can artificially balance or differentiate the channel strengths of paired NOMA users, which is a prerequisite for effective successive interference cancellation (SIC).

Similarly, Jianfeng Shi 0001 et al. show in Energy Efficient Reconfigurable Intelligent Surface Enabled Mobile Edge Computing Networks With NOMA that the physical placement of an RIS directly dictates the energy efficiency of mobile edge computing offloading. Strategic positioning minimizes the transmission power required by edge devices to upload computational tasks.

Adapting to Traffic Dynamics

Static deployments can become suboptimal when user distributions change over time. To bridge this gap, Xiao Liu et al. propose a predictive framework in RIS Enhanced Massive Non-orthogonal Multiple Access Networks: Deployment and Passive Beamforming Design. Instead of placing surfaces at geographic centers, the authors pair Long Short-Term Memory (LSTM) networks for traffic hotspot prediction with a Decaying Double Deep Q-Network ($D^3QN$). This allows the network to optimize the positioning of surfaces based on where demand actually clusters, maximizing long-term energy efficiency.


Scaling to Multi-Surface and Sensing Networks

The deployment problem becomes combinatorial when a network utilizes multiple distributed surfaces to bypass structural blockages or to perform spatial sensing.

Submodular Placement for Sensing

In radar and localization systems, the goal of an IRS is often to maximize target visibility or mutual information rather than data throughput. In Submodular Optimization for Placement of Intelligent Reflecting Surfaces in Sensing Systems, Tara Esmaeilbeig et al. prove that the multi-IRS placement problem exhibits submodularity—a mathematical property akin to the law of diminishing returns. By exploiting submodularity, the authors develop a greedy placement algorithm that provides a provable, near-optimal performance guarantee, demonstrating that strategic multi-surface placement is just as critical as phase-shift optimization for radar sensing accuracy.

Centralized vs. Distributed Trade-offs

When deploying a fixed budget of reflecting elements, engineers face a structural choice: should they deploy one large centralized surface or multiple smaller distributed surfaces? Ming-Min Zhao et al. tackle this in Intelligent Reflecting Surface Aided Full-Duplex Communication: Passive Beamforming and Deployment Design. While a centralized surface maximizes the $M^2$ beamforming gain in a single localized area, a distributed deployment strategy offers superior spatial diversity, creating alternative communication pathways in complex environments with severe blockage.

This spatial sensitivity is mirrored in localization and sensing applications. For example:


Open Frontiers

Despite clear geometric principles, establishing a universal deployment framework remains an open research direction due to several real-world constraints:

Key Foundational Work

RIS Enhanced Massive Non-orthogonal Multiple Access Networks: Deployment and Passive Beamforming Design (2020) — contributed 1 concept elements to this direction.
Large Intelligent Surface/Antennas (LISA): Making Reflective Radios Smart (2019) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_optimal_ris_positioning_994d07a3,
  title = {optimal RIS positioning},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/optimal-ris-positioning/rdr_994d07a339054c6ab9f488a29f429407},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_optimal_ris_positioning_994d07a3_v1,
  title = {optimal RIS positioning},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/optimal-ris-positioning/rdr_994d07a339054c6ab9f488a29f429407/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake