Wireless signals attenuate rapidly as they propagate through space. When a signal encounters a conventional passive wall or mirror, it experiences what engineers call "double path loss." The signal degrades on its journey from the transmitter to the surface, and then degrades again on its way from the surface to the receiver. Mathematically, if the two distances are $d_1$ and $d_2$, the received power plummets proportionally to $1/(d_1^2 d_2^2)$. This multiplicative fading historically made passive relays impractical for long-distance signal enhancement.
Reconfigurable Intelligent Surfaces (RIS)—also known as Intelligent Reflecting Surfaces (IRS)—fundamentally change this equation. When we scale up the number of passive reflecting elements ($N$) on an intelligent surface, the received signal power does not increase linearly. Instead, it scales quadratically ($\propto N^2$). This quadratic power scaling law allows large passive surfaces to overcome double path loss entirely without needing active radio frequency (RF) amplifiers.
The Intuition Behind the $N^2$ Power Gain
To understand why the power scales with the square of the number of elements, we must separate the physics of the surface into two distinct mechanisms: energy collection and energy focusing.
Transmitter ------> [ RIS Surface (N elements) ] ------> Receiver
(Signal 1. Captures power ∝ N (Coherent addition
spreads) 2. Focuses beam width ∝ N yields total power ∝ N²)
1. The Aperture Effect (Energy Collection)
Think of an RIS as a collection of small mirrors. As you add more elements to the surface, its physical size (aperture) grows. A larger surface physically intercepts more of the transmitted electromagnetic wavefront, acting like a larger bucket catching rain. The total amount of environmental energy captured by the surface scales linearly with the number of elements, or $\propto N$.
2. The Coherent Beamforming Effect (Energy Focusing)
If the surface were a flat, static mirror, the captured energy would simply scatter back out into the environment. However, an RIS is intelligent: each element can independently alter the phase of the wave it reflects. By precisely tuning these phase shifts, the surface ensures that the reflections from all $N$ elements arrive at the target receiver exactly in-phase.
This alignment causes the waves to interfere constructively, focusing the scattered energy into a tight, highly directional beam aimed directly at the user. The precision and directivity of this beamforming focus scale linearly with $N$.
Combining the Effects
Because the surface first captures an amount of power proportional to $N$, and then concentrates that power with a directional beamforming gain also proportional to $N$, the final power received at the destination scales as:
$$\text{Received Power} \propto N \times N = N^2$$
As detailed in the comprehensive overview by Mengnan Jian, this far-field quadratic scaling stands in stark contrast to traditional active beamforming at a base station, where doubling the number of antennas under a fixed power budget only yields a linear ($\propto N$) increase in received signal-to-noise ratio (SNR).
Overcoming Double Path Loss
The practical magic of $N^2$ scaling is its ability to neutralize the severe attenuation of the indirect channel. While the path loss product $1/(d_1^2 d_2^2)$ heavily suppresses the signal, deploying a surface with a large number of elements provides an $N^2$ multiplier that counteracts this drop.
The authors of Statistical CSI based Design for Intelligent Reflecting Surface Assisted MISO Systems demonstrate that this total power gain is proportional to the square of the number of reflecting elements multiplied by the number of transmitter antennas. By making $N$ sufficiently large, the path loss bottleneck is broken, turning what would have been a dead zone into a robust communication link.
Passive vs. Active Scaling Laws
It is valuable to contrast this phenomenon with active surfaces. One might assume that adding power amplifiers to the surface elements would preserve or improve this scaling. However, the asymptotic analysis by Zijian Zhang et al. in Active RISs: Signal Modeling, Asymptotic Analysis, and Beamforming Design reveals an ironic twist: active surfaces scale linearly ($\propto N$) rather than quadratically in terms of SNR.
This occurs because active elements amplify not only the incoming signal but also their own internal thermal noise, which scatters omnidirectionally. While active surfaces offer a higher baseline power at lower element counts, passive surfaces rely on their clean, noise-free $N^2$ scaling to dominate as the array size grows large.
System-Level Implications
This scaling behavior fundamentally alters how wireless networks can be expanded. Traditional methods for boosting network capacity require adding more active base stations or complex antenna arrays, both of which introduce significant power overhead due to their power-hungry RF chains.
As argued by Xianghao Yu et al. in Robust and Secure Wireless Communications via Intelligent Reflecting Surfaces, passive surfaces offer a highly scalable alternative, allowing networks to achieve substantial beamforming gains without a corresponding spike in energy consumption.
By exploiting this law, operators can drastically lower base station transmit power requirements. For example, Huimei Han et al. in Intelligent Reflecting Surface Aided Network: Power Control for Physical-Layer Broadcasting show that increasing the density of passive units provides the necessary degrees of freedom to maintain strict quality-of-service targets while minimizing active power expenditure. Ultimately, as proven by Wanli Ni et al. in Intelligent Reflecting Surface Aided Multi-Cell NOMA Networks, scaling the density of these low-cost passive components allows network sum-rate capacity to scale nearly linearly, offering a clear, energy-efficient path forward for next-generation wireless infrastructure.