Statistics of the electromagnetic response of a chaotic reverberation chamber

This article presents a study of the electromagnetic response of a chaotic reverberation chamber (RC) in the presence of losses. By means of simulations and of experiments, the fluctuations in the maxima of the field obtained in a conventional mode-stirred RC are compared with those in a chaotic RC in the neighborhood of the Lowest Useable Frequency (LUF). The present work illustrates that the universal spectral and spatial statistical properties of chaotic RCs allow to meet more adequately the criteria required by the Standard IEC 61000-4-21 to perform tests of electromagnetic compatibility.

💡 Research Summary

This paper investigates the electromagnetic (EM) response of reverberation chambers (RCs) in the vicinity of the Lowest Usable Frequency (LUF), focusing on statistical properties that are essential for electromagnetic compatibility (EMC) testing. Conventional RCs rely on a mechanical stirrer and intrinsic losses to achieve modal overlap, thereby approximating Hill’s hypothesis, which treats the internal field as a random superposition of plane waves. However, near the LUF the modal overlap is moderate (d ≈ 0.45) and the hypothesis breaks down, leading to non‑uniform, anisotropic fields that do not satisfy the IEC 61000‑4‑21 uniformity criterion (σ_dB < 3 dB).

To address this limitation, the authors transform a commercial mode‑stirred RC into a chaotic cavity by attaching three metallic half‑spheres to its walls, thereby destroying parallel planar surfaces and suppressing regular modes. Spectral analysis of 880 lossless eigenmodes shows that the conventional chamber’s normalized spacing distribution deviates from the Wigner surmise (GOE prediction), whereas the chaotic chamber follows it closely, confirming its chaotic nature.

The EM response with losses is modeled using the dyadic Green’s tensor (DGT) expressed as a sum over complex resonances. Complex eigenfrequencies and eigenfields are obtained via finite‑element simulations; losses are introduced through low‑conductivity patches on the walls, yielding a realistic mean quality factor Q≈1500–2000. In chaotic cavities, the complexness parameter of each mode and the resonance widths obey the statistical predictions of random matrix theory (RMT) for open chaotic systems.

A key statistical descriptor is the phase rigidity ρ = ⟨E²⟩/⟨|E|²⟩ of each Cartesian field component. The normalized intensity Ĩ = |E|²/⟨|E|²⟩ follows the P_ρ(Ĩ) distribution, which interpolates between Porter‑Thomas (closed) and exponential (fully open) limits. The authors demonstrate that, for the chaotic RC, the empirical intensity histograms across many frequencies and stirrer positions match the theoretical P_ρ(Ĩ) when the measured distribution of ρ is used. In contrast, the conventional RC exhibits significant deviations, reflecting non‑universal, anisotropic statistics.

The IEC uniformity criterion is evaluated by computing σ_dB, the standard deviation (in dB) of the maximum field amplitude across eight measurement points for each stirrer configuration. Numerical results show that the chaotic RC satisfies σ_dB < 3 dB in virtually all cases (average ≈ 2.1 dB), while the conventional RC exceeds the limit in roughly 25 % of cases (average ≈ 2.6 dB). Increasing the number of measurement points (8, 16, 64) does not alter the chaotic chamber’s average σ_dB, whereas the conventional chamber’s average drifts upward and its spread remains large, underscoring the lack of statistical universality.

Experimental validation is performed on the same commercial chamber, both in its original configuration and after adding the three half‑spheres. Measurements of the S‑matrix between a dipole and a monopole antenna over a 20 MHz band centered at 400 MHz, for 30 stirrer angles and eight antenna positions, confirm the numerical findings: the chaotic configuration yields a lower mean σ_dB (≈ 2.18 dB) and a tighter distribution compared with the conventional case (≈ 2.57 dB). The agreement between simulation and experiment validates the modeling approach, including the treatment of losses and modal overlap.

In conclusion, converting a reverberation chamber into a chaotic cavity restores the universal spectral and spatial statistics predicted by RMT, even when modal overlap is moderate. This universality ensures that the IEC 61000‑4‑21 field‑uniformity criterion is reliably met near the LUF, improving the accuracy and repeatability of EMC tests. The study thus positions chaotic reverberation chambers as a practical and superior alternative for high‑fidelity electromagnetic compatibility assessments.