Sequential adaptive sampling algorithm to reduce the near-field measurement time

Abstract

This paper presents a sequential adaptive sampling algorithm in order to reduce measurement time of near-field scan. The originality of this approach is to use a deterministic mesh swept to a sequential progressive adaptive algorithm that defines whether a point must be captured or not. All the parameters of the proposed algorithm is set according near-field characteristics and the measurement setup. This approach is validated on two case studies.

Fast and Accurate Near-Field Measurement Method Using Sequential Spatial Adaptive Sampling (SSAS) Algorithm

“…The first version of this algorithm, introduced in [9], suffers of inaccuracies to estimate the maximum emission level. In this paper, a modified version is presented to provide a faithful representation of the near-field above the DUT.…”

“…The SSAS algorithm is based on the Multi-Level Adaptive approach [11] initially introduced in [9]. The first step consists in collecting an initial dataset of F. Let N the initial number of sampling points and = { = ( , ), ∈ Ω} =1 the initial dataset of probe positions.…”

“…According to the definition of W (with α=10dB), the maximum deviation from the mean was fixed at 10dB. (24) defines the first simple definition of k, used in [9].…”

Optimized Algorithm to Reduce the Near-Field Measurement Time on FPGA Device

“…The algorithm used for this study is described in [1]. In summary, this algorithm is based on a progressive sequential meshing as defined in Fig.…”

“…The displacement time of the probe is small compared to the acquisition time. This is the benefit of using a regular mesh (Sukharev grid) coupling to the progressive sequential sweep [1]. The distance between two consecutive points is minimized.…”