2010
DOI: 10.1109/tap.2010.2078480
|View full text |Cite
|
Sign up to set email alerts
|

Fast Periodic Interpolation Method for Periodic Unit Cell Problems

Abstract: A fast periodic interpolation method (FPIM) is presented for rapidly computing fields in a unit cell of an infinitely periodic array. For low and moderate frequencies (for unit cells smaller than or on the order of the wavelength) the FPIM has the computational cost of ( ) and it requires only (1) periodic Green's function (PGF) evaluations, for sources and observers. For high-or mixed-frequencies the computational cost scales aswhere is the domain size within the unit cell and is the wavelength. FPIM is based… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1

Citation Types

2
11
0

Year Published

2010
2010
2024
2024

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 19 publications
(13 citation statements)
references
References 34 publications
(51 reference statements)
2
11
0
Order By: Relevance
“…Note that a value of N c = 7 suffices to obtain the interpolated MPGF with four significant figures. This means that only 64 samples of the MPGF are required to generate this interpolation, which is substantially less than the number of samples required in [25] and [26] to obtain the interpolated periodic Green's functions with the same accuracy of four significant figures.…”
Section: Numerical Results and Validationsmentioning
confidence: 99%
See 1 more Smart Citation
“…Note that a value of N c = 7 suffices to obtain the interpolated MPGF with four significant figures. This means that only 64 samples of the MPGF are required to generate this interpolation, which is substantially less than the number of samples required in [25] and [26] to obtain the interpolated periodic Green's functions with the same accuracy of four significant figures.…”
Section: Numerical Results and Validationsmentioning
confidence: 99%
“…The CPU time required for this spatial domain computation of the MoM matrix entries is substantially reduced by introducing two improvements. The first improvement is that the MPGF with 2-D periodicity for the potentials are judiciously interpolated in the spatial domain [25], [26] in terms of 2-D Chebyshev polynomials after extracting the behavior of the MGPF around the source points. The second improvement in the spatial domain computation of the MoM entries has to do with the efficient computation of 4-D singular integrals.…”
mentioning
confidence: 99%
“…As is evident from (13), each component requires a number of operations, , which is approximately commensurate with the evaluation of an Ewald Sum. As the nearfield matrix elements are themselves Ewald Sums, the precomputation cost scales as (15) While the first contribution is clearly linear, it is not immediately clear as to how the latter dependent contribution will scale. As the translation operator only depends upon the separation between boxes, and the tree structures imposes a uniform superstructure on the computational domain, there is a significant redundancy that can be exploited in this stage of precomputation.…”
Section: Computational Costmentioning
confidence: 99%
“…The subject remained largely unexplored for over a decade, until Otani and Nishimura [14] used a periodic FMM for the analysis of photonic crystals. Other types of hierarchical methods have been presented more recently, namely interpolatory methods such as those due to Li [15] and Chan [16]. Instead of exploiting addition theorems for the Green's Function, these methods hinge upon the projection of the Green's Function onto a hierarchy of increasingly sparse interpolatory grids.…”
Section: Introductionmentioning
confidence: 99%
“…An overview of techniques is provided in [12]. Additionally, a new interesting method based on fast periodic interpolations is reported in [13].…”
Section: Introductionmentioning
confidence: 99%