Reflected wireless signals under random spatial sampling

We present a propagation model showing that a transmitter randomly positioned in space generates unbounded peaks in the histogram of the resulting power, provided the signal strength is an oscillating or non-monotonic function of distance. Specifically, these peaks are singularities in the empirical probability density that occur at turning point values of the deterministic propagation model. We explain the underlying mechanism of this phenomenon through a concise mathematical argument. This observation has direct implications for estimating random propagation effects such as fading, particularly when reflections off walls are involved. Motivated by understanding intelligent surfaces, we apply this fundamental result to a physical model consisting of a single transmitter between two parallel passive walls. We analyze signal fading due to reflections and observe power oscillations resulting from wall reflections – a phenomenon long studied in waveguides but relatively unexplored in wireless networks. For the special case where the transmitter is placed halfway between the walls, we present a compact closed-form expression for the received signal involving the Lerch transcendent function. The insights from this work can inform design decisions for intelligent surfaces deployed in cities.

💡 Research Summary

The paper investigates a fundamental phenomenon that arises when a wireless transmitter is placed at random locations in space and the deterministic received‑power function is non‑monotonic, i.e., it contains oscillations due to multipath or reflections. The authors first formulate a generic propagation model S(r)=A(r)·W(r), where A(r)≥0 is a monotonically decreasing attenuation term and W(r) is an oscillatory (possibly complex) wave term. The received power is P(r)=|S(r)|²=A(r)²·|W(r)|². Because |W(r)|² can produce local minima and maxima, P(r) possesses turning points t₁,…,t_m where P′(t_i)=0.

When the transmitter position r is treated as a continuous random variable U on an interval (r_L,r_U), the random power V=P(U) can be analyzed via a change‑of‑variables. By partitioning the interval into monotonic sub‑intervals A_i=(t_i,t_{i+1}) and defining g_i(r)=P(r) on each A_i, the inverse g_i⁻¹ exists. The probability density of V on the corresponding range (v_i,v_{i+1}) is

 f_V(v)=f_U(g_i⁻¹(v))/|g_i′(g_i⁻¹(v))|.

Since g_i′(t_i)=0 at a turning point, the denominator vanishes and f_V(v) diverges like |v−P(t_i)|^{-1/2}. Thus the histogram of received power exhibits unbounded peaks (singularities) at the power values associated with the deterministic turning points. This “inverse‑square‑root singularity” is a direct consequence of the geometry of the deterministic model and is independent of the specific distribution of U, provided U has a smooth density.

To demonstrate the physical relevance, the authors study a canonical two‑parallel‑wall scenario, a model often used for urban canyon or waveguide analysis. Using the classical method of images, each wall reflection is represented by an infinite series of image transmitters. The total field at the observer (origin) is a sum over all images:

 S(r)=∑_{n=−∞}^{∞} A_n e^{j k d_n},

where d_n is the total path length of the n‑th image, A_n incorporates distance‑dependent attenuation and reflection coefficients, and k=2π/λ is the wavenumber. For the symmetric case where the real transmitter sits exactly halfway between the walls (a=b=d/2), the distances simplify to d_n = 2n d for even images and d_n = (2n+1) d for odd images. The infinite series can be summed in closed form using the Lerch transcendent Φ(ζ,s,γ), yielding a compact expression for the received signal and consequently for the power:

 S(r)=C·Φ(ζ,s,γ), P(r)=|C·Φ(ζ,s,γ)|².

The resulting power function has precisely two turning points, corresponding to constructive and destructive interference of the forward and reflected waves. Their locations depend on the wall separation d and the wavelength λ (e.g., minima at d/2±λ/4). The analytical form confirms the presence of the inverse‑square‑root singularities in the probability density of P when the transmitter position is randomized.

Numerical simulations (Monte‑Carlo sampling of random transmitter locations) corroborate the theory: the empirical histogram of received power shows sharp spikes at the predicted power values, while away from those points the distribution follows the smooth transformation of the underlying spatial density. The authors contrast this behavior with standard fading models (Rayleigh, Nakagami) that assume smooth PDFs and highlight the challenges such singularities pose for statistical estimation, confidence‑interval construction, and performance analysis.

Beyond the specific two‑wall example, the paper discusses broader implications for intelligent (reconfigurable) surfaces (IRS). Although IRS can manipulate the phase of reflected waves to mitigate destructive interference, the geometric turning points are intrinsic to the environment and cannot be eliminated solely by phase control. Hence, optimal IRS deployment must consider the placement of transmitters/receivers relative to walls to avoid power values that correspond to singularities. The authors also outline how stochastic geometry (e.g., Poisson point processes) could be combined with their hybrid deterministic‑random model to study large‑scale networks where many transmitters are randomly distributed, and how the singularities might affect network‑wide metrics such as outage probability and throughput.

In summary, the paper makes four key contributions: (1) a rigorous proof that random transmitter placement induces singularities in the PDF of received power whenever the deterministic power function has turning points; (2) a closed‑form Lerch‑function expression for the symmetric two‑wall configuration; (3) a hybrid propagation model that bridges deterministic ray‑tracing with stochastic fading, illustrated through the wall‑reflection scenario; and (4) insight into how these findings inform the design and analysis of intelligent surfaces and future 5G/6G deployments. The work opens new avenues for incorporating non‑smooth probability densities into wireless‑channel modeling and for developing mitigation strategies in environments rich with reflective surfaces.