In radio frequency interference study, equivalent dipole moments are widely used to reconstruct real radiation noise sources. Previous reconstruction methods, such as least square method (LSQ) and optimization method are affected by parameter selections, such as number and locations of dipole moments and choices of initial values. In this paper, a new machine learning based source reconstruction method is developed to extract the equivalent dipole moments more accurately and reliably. Based on the near-field patterns, the proposed method can determine the minimal number of dipole moments and their corresponding locations. Furthermore, the magnitude and phase for each dipole moment can be extracted. The proposed method can extract the dominant dipole moments for the unknown noise sources one by one. The proposed method is applied to a few theoretical examples first. The measurement validation using a test board and a practical cellphone are also given. Compared to the conventional LSQ method, the proposed machine learning based method is believed to have a better accuracy. Also, it is more reliable in handling noise in practical applications.

Machine Learning Based Source Reconstruction for RF Desense

Ahstract- In this paper, the dipole-moment based reciprocity method is used to perform desense prediction and mitigation from the DDR noise source to the nearby RF antenna. The noise source is from the DDR signals between the application processor and the memory IC. Radiation physics of the noise source is analyzed by understanding of the current flow. Firstly, the random nature of the DDR signals is analyzed using the measurement data. Based on the measurement data, the setup of the near field scanning is further determined. The desense prediction procedures are decomposed into two steps: the forward problem and the reverse problem. In the forward problem, the noise source is approximately regarded as a single magnetic dipole moment based on the near field scanning above this specific electronic device. In the reverse problem, the transfer function from the magnetic dipole moment to the victim antenna is obtained by measuring the H field when the victim antenna radiates. Based on the measurements of forward and reverse problem, the coupled noise to the victim antenna can be analytically estimated. The estimated RFI results are compared with direct RFI measurement to validate the dipole-moment based reciprocity method. Lastly, a few methods to mitigate the desense are also discussed.

Desense Prediction and Mitigation from DDR Noise Source

A method of moment (MoM)-based current reconstruction method is proposed to estimate the surface current density on the ground plane. Current continuity property is automatically enforced with Rao–Wilton–Glisson basis function. Both the least square method and the optimization method are utilized to solve the inverse problem and obtain the ground current. The proposed optimization method is successfully validated with numeric simulations and also a real-world measurement example. The reconstructed current distribution on the ground plane is further used in a typical radio frequency interference (RFI) example to perform RFI estimation and provide guidelines for RFI design. The proposed MoM-based ground current reconstruction method can be valuable to estimate and debug RFI issues in early design stage.

MoM-Based Ground Current Reconstruction in RFI Application

A transfer function based calculation method is proposed to estimate radio frequency interference (RFI) problems. The closed-form equations are analytically derived from Maxwell's equations and the reciprocity theorem. The derived equations can clearly decompose the RFI problem into two parts: the noise source and the coupling transfer function to the antenna. Based on derivations, a transfer function concept is proposed to quantify the coupling coefficient from each unit dipole moment to the victim antenna. The transfer functions can be easily obtained from S -parameter measurements. The proposed method is validated through numeric simulations and real cellphone experiments. Engineering insights drawn from the method are also discussed. Overall, the proposed method is accurate, fast, and can provide physical insights for practical designs.

A Transfer Function Based Calculation Method for Radio Frequency Interference

This article demonstrates an inaudible attack on smart speakers using electromagnetic interference (EMI). The EMI induces voltages on the order of a few millivolts on conductors, which are then converted into baseband signals by exploiting the inherent nonlinearity of microphones. The EMI signal is specially preprocessed to minimize the useless harmonics generation at the microphone output signals, which significantly improves the recognition rate as well as nullify the previous countermeasures based on the harmonics detection. The sensitive carrier frequency found by our proposed method can improve the attack distance as well. A measurement-based methodology is applied to locate the sensitive regions for noise coupling without knowing the layout of the printed circuit board (PCB), and the transfer function is also obtained to insure the main coupling location. Our experiments show that in open space, intentional EMI under 2.5 W can inject commands at distances up to 2.5 m on smart speakers.

Inaudible Attack on Smart Speakers With Intentional Electromagnetic Interference

The dipole moment-based reciprocity to model intra-system electromagnetic interference (EMI) in wireless devices is proposed in this paper. In this model, the decomposition method based on reciprocity is used to calculate the coupling voltage on the victim antenna. The original noise source is replaced by equivalent dipole moments in the forward problem while electromagnetic fields on the locations of dipole moments are scanned in the reverse problem. The proposed model has been validated with full-wave simulations and measurements; it provides not only convenience to estimate radio-frequency interference (RFI), but also physical insight to understand the intra-system EMI problem.

Analytical intra-system EMI model using dipole moments and reciprocity

Heatsinks are widely used to dissipate heat in electronic devices. RF emission from heatsinks and ICs can cause radio frequency interference (RFI) issues. The existence of the heatsink affects the coupling from the noise source to the victim antenna. In this paper, the heatsink RFI problems are analyzed by two methods using reciprocity theorem. In the first method, the heatsink effect is combined with the antenna to derive the transfer function. In the second method, the heatsink can be characterized in both the source side and the antenna side on a closed 6-surface box. The physical insights and comparisons of the two RFI estimation methods based on reciprocity theorem will be investigated. The proposed methods will be utilized on a few numeric simulations for validations.

Reciprocity Theorem Based RFI Estimation for Heatsink Emission

When a wireless device is connected to wireless networks such as Wi-Fi and GSM, annoying noise in audible frequency range, named buzz noise, can appear in the audio system. The characterization and quantification of buzz noise provide the theoretical fundamental for solving this problem. It has been discovered that the buzz noise is converted from the radio frequency (RF) signal through the unexpected square-law demodulation in the microphone. Based on the buzz noise generation mechanism, this paper developed a two-port network model to quantify the RF noise coupling path and the conversion from RF noise to buzz noise. The results have been validated by repeating the measurements with different RF coupling path.

Measurement-Based Quantification of Buzz Noise in Wireless Devices

The conventional definition of shielding effectiveness (SE) is well suited for calculations of far-field electromagnetic shielding. However, in the near field, SE calculations are not as straightforward. In radio-frequency interference (RFI) problems, the majority of field coupling occurs in the near field. Thus, a well-defined method for calculating the near-field SE is needed to estimate the suppression of RFI achieved by shielding cans. In this study, a method based on near-field scanning is developed to extract the SE of board-level shielding cans. The SE is defined by modeling the shielded noise source as equivalent dipole moments. The accuracy of the equivalent source is analyzed via the least-square error and correlation coefficient as confidence verification parameters. By applying the reciprocity theorem, the voltage coupled on a planar inverted F antenna from an unshielded and a shielded source is calculated. The coupled voltage from a shielded noise source serves as a reference to validate the effectiveness of the proposed method. Practical shielding cans were used to develop and validate the SE extraction method using full-wave 3D simulations and measurements.

/pdf/near-field-scanning-based-shielding-effectiveness-extraction-7pak45y2td.pdf

Shingo Seto

Papers

A Transfer Function Based Calculation Method for Radio Frequency Interference

Analytical intra-system EMI model using dipole moments and reciprocity

Reciprocity Theorem Based RFI Estimation for Heatsink Emission

Measurement-Based Quantification of Buzz Noise in Wireless Devices

Near-Field Scanning-Based Shielding Effectiveness Extraction for Board-Level Shielding Cans