Computational approaches for generating electromagnetic Gaussian Schell-model sources

Santasri Basu; Milo W. Hyde; Xifeng Xiao; David G. Voelz; Olga Korotkova

doi:10.1364/OE.22.031691

1. Introduction

The electromagnetic Gaussian Schell-model (EGSM) source/beam was introduced as an extension of the scalar Gaussian Schell-model (GSM) beam [1, 2]. Since then, it has attracted special attention due to the interesting polarization evolution that can occur on its propagation and the reduction in scintillation that is possible in free-space optical communications, imaging through turbulence, and remote sensing applications [3–10]. The ability to customize the EGSM attributes can lead to improved performance for particular applications and scenarios.

An EGSM beam can be described by a 2 × 2 cross-spectral density (CSD) matrix that characterizes second-order correlations between two mutually orthogonal components of the fluctuating electric field at a pair of spatial arguments and frequency [10]. Substantial progress has been made on the theoretical understanding of these beams including their propagation aspects, correlation features, and realizability conditions [11–15]. Concurrently, various methods have been proposed to produce EGSM sources numerically and experimentally [16–23]. These efforts include an approach to experimentally synthesize EGSM sources with the same mutually orthogonal electric field components [18], an experimental measurement to verify the validity of the EGSM beam parameters [19, 22], and a practical method of producing a general EGSM source [20]. Most recently, a reduction in scintillation for a particular subclass of EGSM beam (completely unpolarized) was successfully demonstrated for propagation in thermally simulated atmospheric turbulence [23]. These studies provide practical techniques to physically realize the EGSM beam and successfully validate the existing theory; however, a practical ability to design and control the EGSM beam characteristics was not the primary emphasis of these efforts.

In this paper, the fundamental relationships between the two orthogonal polarization components of an EGSM beam are examined and a computational approach for creating numerical random screens that are associated with the components is presented. The desired EGSM beam parameters determine the selection of the screen parameters. The concept is that a pair of such screens is applied at the source plane to two orthogonally polarized coherent waves. The two resulting wave components constitute an instantaneous electromagnetic beam realization. Sufficiently large, mutually independent sequences of the screen pairs are then applied and the resulting intensities for each field component are averaged over these ensembles of realizations. The four average intensities, two representing self-correlations and two representing joint correlations, between the two components comprise the EGSM beam. The beams produced in this manner are consistent with the EGSM realizability conditions stemming from the fundamental properties of the CSD matrix [14, 15]. The produced ensembles of screens can be used in a numerical wave optics simulation or in the laboratory with spatial light modulators (SLMs).

In Section 2, two screen methodologies, the phase screen (PS) and the complex transmittance screen (CS), are introduced. The relationships between the screen parameters and the desired EGSM beam parameters are explored and the benefits and constraints of the two approaches are discussed. The screen methodologies are validated in Section 3 via numerical modeling of typical EGSM beams and comparison of the results with theoretical predictions. Some final remarks and future research directions are given in Section 4.

2. Methodology

Figure 1 shows a proposed experimental schematic for generating EGSM sources. Note that this proposed set-up is similar to that presented in Ref [20]. Light leaves a laser and traverses a beam expander (BE) and half-wave plate (HWP) before being split along two paths by a polarizing beam splitter (PBS). The initial HWP is used to control the relative amplitudes of the fields along each path. In paths 1 and 2, the light is polarized vertically (denoted by the circle) and horizontally (denoted by the two-sided arrow), respectively. It is assumed here that the SLMs control only vertically polarized light; thus, a HWP is used in path 2 to transform horizontal linear polarization into vertical polarization.

Fig. 1 Proposed experimental schematic for generating EGSM sources. The acronyms used in the figure are beam expander (BE), half-wave plate (HWP), polarizing beamsplitter (PBS), lens systems (LS), spatial light modulator (SLM), Gaussian amplitude filter (GAF), and variable retarder (VR). The polarization state of the light passing through the system is denoted by two-sided arrows (representing horizontal polarization) and circles (representing vertical polarization). When both are present, the light is in a general polarization state, i.e., polarized, partially polarized, or unpolarized.

Download Full Size | PDF

The light in both paths is then incident on the SLMs. Because of their widespread use, it is assumed that the SLMs in Fig. 1 are reflective, phase-only SLMs. The SLMs impart random, correlated phases to the light in paths 1 and 2. After the SLMs, the light enters general lens systems (LS). These LS could be spatial filters, 4-f systems, etc. and are included to remove unwanted diffraction orders, produced by the SLMs, which may corrupt the desired EGSM source output.

After traversing the LS, the light in both paths passes through Gaussian amplitude filters (GAFs) which set the desired Gaussian amplitude widths of the EGSM source (discussed in more detail below). The light from path 1 and path 2 is then recombined using a PBS. Note that the HWP, located before the GAF, on path 1 is required to transform the polarization state from vertical to horizontal polarization so that the light from both paths can be recombined. Lastly, a liquid crystal variable retarder (VR) is included to control the relative phasing between the vertical and horizontal polarization states.

It must be stated that the experimental set-up depicted in Fig. 1 is hypothetical. No experimental results are presented in this paper. The approaches presented here for generating EGSM sources are validated via simulation. The above description is included to provide background on how one might generate EGSM sources in practice. An experimental system similar to the one in Fig. 1 is currently in work. Experimental results will be presented in a future paper.

Two methods for generating EGSM sources are presented in this paper—the PS and CS methods. The PS approach involves generating two random phase screens, one for each polarization component. This approach can be implemented in the laboratory with two phase-only SLMs as shown in Fig. 1. The interested reader is referred to Ref [24]. for the practical aspects of generating a scalar GSM beam with a single nematic phase-only SLM. The PS approach is equivalent to that presented in Ref [20]; however, here, the derivation is presented differently.

While the PS approach is useful for practical implementation purposes, its main disadvantage is that the autocorrelation function of the screen transmittances is typically not of the desired form. This is a significant problem when the desired autocorrelation function is not Gaussian. The CS approach, on the other hand, does not suffer from this shortcoming. This approach involves generating two screens with complex transmittance functions, i.e., both the amplitude and phase of the incident wave are randomized spatially upon transmission through the screen. The CS approach is ideal for numerical simulations, but laboratory implementation is rather difficult because both the amplitude and phase of the source must be controlled.

The elements of the CSD matrix of an EGSM source are [10]

\begin{matrix} W_{α β} (ρ_{1}, ρ_{2}, 0; ω) = \sqrt{S_{α} (ρ_{1}; ω)} \sqrt{S_{β} (ρ_{2}; ω)} μ_{α β} (| ρ_{1} - ρ_{2} |; ω) \\ S_{α} (ρ; ω) = A_{α}^{2} \exp (\frac{- ρ^{2}}{2 σ_{α}^{2}}) \\ μ_{α β} (| ρ_{1} - ρ_{2} |; ω) = B_{α β} \exp (\frac{- {| ρ_{1} - ρ_{2} |}^{2}}{2 δ_{α β}^{2}}), \end{matrix}

where

α, β = x, y

,

S_{α}

is the spectral density,

μ_{α β}

is the spectral correlation function and

ρ = \hat{x} x + \hat{y} y

. Further,

σ_{α}

and

δ_{α β}

are the r.m.s. widths of the spectral density and correlation profiles, respectively. The parameters

σ_{α}

,

B_{α β}

, and

δ_{α β}

are constrained by the following relationships:

\begin{matrix} B_{α β} = 1 α = β \\ | B_{α β} | \leq 1 α \neq β \\ B_{α β} = B_{β α}^{*} \\ δ_{α β} = δ_{β α} \\ \frac{1}{4 σ_{α}^{2}} + \frac{1}{δ_{α α}^{2}} ≪ \frac{2 π^{2}}{λ^{2}} . \end{matrix}

In addition, an EGSM source must satisfy the fork inequality

\sqrt{\frac{δ_{x x}^{2} + δ_{y y}^{2}}{2}} \leq δ_{x y} \leq \sqrt{\frac{δ_{x x} δ_{y y}}{| B_{x y} |}}

to be realizable [15]. It is imperative to show that both proposed approaches produce sources whose parameters obey the above constraints. Hereafter, the dependence on the radian frequency

ω

is omitted for the sake of brevity.

2.1 PS approach

Let the electric field in the source plane, $z = 0$ , be

\begin{matrix} E (ρ, 0) = \hat{x} E_{x} (ρ) + \hat{y} E_{y} (ρ) \\ E_{α} (ρ) = C_{α} \exp (\frac{- ρ^{2}}{4 σ_{α}^{2}}) \exp [j ϕ_{α} (ρ)], \end{matrix}

where

C_{α} = | C_{α} | \exp (j θ_{α})

is a complex constant and

ϕ_{α} (ρ)

is the random phase contribution due to the screen. Performing the autocorrelations necessary to fill the CSD matrix produces

\begin{matrix} 〈 E (ρ_{1}, 0) E^{*} (ρ_{2}, 0) 〉 = W (ρ_{1}, ρ_{2}, 0) = [\begin{matrix} 〈 E_{x} (ρ_{1}, 0) E_{x}^{*} (ρ_{2}, 0) 〉 & 〈 E_{x} (ρ_{1}, 0) E_{y}^{*} (ρ_{2}, 0) 〉 \\ 〈 E_{y} (ρ_{1}, 0) E_{x}^{*} (ρ_{2}, 0) 〉 & 〈 E_{y} (ρ_{1}, 0) E_{y}^{*} (ρ_{2}, 0) 〉 \end{matrix}] \\ 〈 E_{α} (ρ_{1}, 0) E_{β}^{*} (ρ_{2}, 0) 〉 = C_{α} C_{β}^{*} \exp [- (\frac{ρ_{1}^{2}}{4 σ_{α}^{2}} + \frac{ρ_{2}^{2}}{4 σ_{β}^{2}})] 〈 \exp [j ϕ_{α} (ρ_{1})] \exp [- j ϕ_{β} (ρ_{2})] 〉 . \end{matrix}

The phase screen realizations are sample functions drawn from two correlated Gaussian random processes. Hereafter, for the sake of brevity, functions evaluated at

ρ_{1}

or

ρ_{2}

are denoted with a subscript 1 or 2, respectively. For example,

ϕ_{α} (ρ_{1})

is expressed as

ϕ_{α 1}

.

The expectation on the second line of Eq. (5) is recognized as the joint characteristic function of the Gaussian random variables $ϕ_{α}$ and $ϕ_{β}$ evaluated at $ω_{1} = 1$ and $ω_{2} = - 1$ , where $ω_{1}$ and $ω_{2}$ are radian frequencies. This expression is [25]

〈 \exp [j ϕ_{α 1}] \exp [- j ϕ_{β 2}] 〉 = \exp {- \frac{1}{2} (σ_{ϕ_{α}}^{2} + σ_{ϕ_{β}}^{2}) [1 - \frac{2 σ_{ϕ_{α}} σ_{ϕ_{β}}}{σ_{ϕ_{α}}^{2} + σ_{ϕ_{β}}^{2}} ρ_{ϕ_{α} ϕ_{β}} γ_{ϕ_{α} ϕ_{β}} (| ρ_{1} - ρ_{2} |; ℓ_{ϕ_{α} ϕ_{β}})]},

where

σ_{ϕ_{α}}

and

σ_{ϕ_{β}}

are the standard deviations of the

ϕ_{α}

and

ϕ_{β}

phase screens, respectively;

0 \leq ρ_{ϕ_{α} ϕ_{β}} \leq 1

is a correlation coefficient (

ρ_{ϕ_{α} ϕ_{β}} = 1

if

α = β

); and

γ_{ϕ_{α} ϕ_{β}}

is the normalized cross-correlation function taken here to be Gaussian-shaped, viz.,

γ_{ϕ_{α} ϕ_{β}} (| ρ_{1} - ρ_{2} |; ℓ_{ϕ_{α} ϕ_{β}}) = \exp (- \frac{{| ρ_{1} - ρ_{2} |}^{2}}{ℓ_{ϕ_{α} ϕ_{β}}^{2}}) .

The symbol

ℓ_{ϕ_{α} ϕ_{β}}

is the spatial cross-correlation radius of the phase screens

ϕ_{α}

and

ϕ_{β}

. Assuming that

(σ_{ϕ_{α}}^{2} + σ_{ϕ_{β}}^{2}) / 2 ≫ 1

, or equivalently

σ_{ϕ_{α}}, σ_{ϕ_{β}} \geq π

,

γ_{ϕ_{α} ϕ_{β}}

can be safely approximated as

γ_{ϕ_{α} ϕ_{β}} \approx 1 - {| ρ_{1} - ρ_{2} |}^{2} / ℓ_{ϕ_{α} ϕ_{β}}^{2}

. Substituting this expression into Eq. (6), then into Eq. (5), and simplifying produces

\begin{matrix} 〈 E_{α 1} E_{β 2}^{*} 〉 \approx C_{α} C_{β}^{*} \exp [- (\frac{ρ_{1}^{2}}{4 σ_{α}^{2}} + \frac{ρ_{2}^{2}}{4 σ_{β}^{2}})] \exp [- \frac{1}{2} (σ_{ϕ_{α}}^{2} - 2 ρ_{ϕ_{α} ϕ_{β}} σ_{ϕ_{α}} σ_{ϕ_{β}} + σ_{ϕ_{β}}^{2})] \\ \exp [- \frac{{| ρ_{1} - ρ_{2} |}^{2}}{ℓ_{ϕ_{α} ϕ_{β}}^{2} / σ_{ϕ_{α}} σ_{ϕ_{β}} ρ_{ϕ_{α} ϕ_{β}}}] . \end{matrix}

By comparing Eq. (8) to Eq. (1), one deduces the following relationships:

\begin{matrix} δ_{x x} = \frac{1}{\sqrt{2}} \frac{ℓ_{ϕ_{x} ϕ_{x}}}{σ_{ϕ_{x}}} & A_{x} = | C_{x} | \\ δ_{y y} = \frac{1}{\sqrt{2}} \frac{ℓ_{ϕ_{y} ϕ_{y}}}{σ_{ϕ_{y}}} & A_{y} = | C_{y} | \\ δ_{x y} = \frac{1}{\sqrt{2}} \frac{ℓ_{ϕ_{x} ϕ_{y}}}{\sqrt{σ_{ϕ_{x}} σ_{ϕ_{y}} ρ_{ϕ_{x} ϕ_{y}}}} \\ | B_{x y} | = \exp [- \frac{1}{2} (σ_{ϕ_{x}}^{2} - 2 ρ_{ϕ_{x} ϕ_{y}} σ_{ϕ_{x}} σ_{ϕ_{y}} + σ_{ϕ_{y}}^{2})] & ∠ B_{x y} = θ_{x} - θ_{y} . \end{matrix}

Note that the relations reported in the left column of Eq. (9) are coupled and cannot be chosen at will. On the other hand, the relations in the right column of Eq. (9) are uncoupled and can be chosen at will. Referring back to Fig. 1,

A_{x}

and

A_{y}

are controlled using the initial HWP,

σ_{x}

and

σ_{y}

are set by using the appropriate GAFs, and

∠ B_{x y}

is set using the VR. The remaining EGSM source parameters are determined by the statistical properties of the phases commanded to the SLMs discussed in detail in Section 2.3.

2.2. CS approach

Let the electric field components in the source plane, $z = 0$ , be

E_{α} (ρ) = C_{α} \exp (\frac{- ρ^{2}}{4 σ_{α}^{2}}) T_{α} (ρ),

where

T_{α} (ρ)

is the complex transmittance function of the screen. Performing the autocorrelations necessary to fill the CSD matrix produces

〈 E_{α 1} E_{β 2}^{*} 〉 = C_{α} C_{β}^{*} \exp [- (\frac{ρ_{1}^{2}}{4 σ_{α}^{2}} + \frac{ρ_{2}^{2}}{4 σ_{β}^{2}})] 〈 T_{α 1} T_{β 2}^{*} 〉 .

Just like

ϕ_{α}

and

ϕ_{β}

in the PS approach,

T_{α}

and

T_{β}

are sample functions drawn from two correlated Gaussian random processes. This time, however, the random processes are complex.

The expectation in Eq. (11) is recognized as the cross-correlation function of the Gaussian random processes $T_{α}$ and $T_{β}$ :

〈 T_{α 1} T_{β 2}^{*} 〉 = σ_{T_{α}} σ_{T_{β}} ρ_{T_{α} T_{β}} γ_{T_{α} T_{β}} (| ρ_{1} - ρ_{2} |; ℓ_{T_{α} T_{β}}),

where

σ_{T_{α}}

and

σ_{T_{β}}

are the standard deviations of the

T_{α}

and

T_{β}

screens, respectively;

0 \leq ρ_{T_{α} T_{β}} \leq 1

is a correlation coefficient (

ρ_{T_{α} T_{β}} = 1

if

α = β

); and

γ_{T_{α} T_{β}}

is the normalized cross-correlation function taken here to be Gaussian-shaped, namely,

γ_{T_{α} T_{β}} (| ρ_{1} - ρ_{2} |; ℓ_{T_{α} T_{β}}) = \exp (- \frac{{| ρ_{1} - ρ_{2} |}^{2}}{ℓ_{T_{α} T_{β}}^{2}}) .

The symbol

ℓ_{T_{α} T_{β}}

is the spatial cross-correlation radius of the complex transmittance screens

T_{α}

and

T_{β}

. Substituting Eqs. (12) and (13) into Eq. (11) and simplifying produces

〈 E_{α 1} E_{β 2}^{*} 〉 = C_{α} C_{β}^{*} σ_{T_{α}} σ_{T_{β}} ρ_{T_{α} T_{β}} \exp [- (\frac{ρ_{1}^{2}}{4 σ_{α}^{2}} + \frac{ρ_{2}^{2}}{4 σ_{β}^{2}})] \exp [- \frac{{| ρ_{1} - ρ_{2} |}^{2}}{ℓ_{T_{α} T_{β}}^{2}}] .

By comparing Eq. (14) to Eq. (1), the following relationships are deduced:

\begin{matrix} δ_{x x} = \frac{ℓ_{T_{x} T_{x}}}{\sqrt{2}} & A_{x} = σ_{T_{x}} | C_{x} | \\ δ_{y y} = \frac{ℓ_{T_{y} T_{y}}}{\sqrt{2}} & A_{y} = σ_{T_{y}} | C_{y} | \\ δ_{x y} = \frac{ℓ_{T_{x} T_{y}}}{\sqrt{2}} \\ | B_{x y} | = ρ_{T_{x} T_{y}} & ∠ B_{x y} = θ_{x} - θ_{y} . \end{matrix}

While not yet evident, the relations reported in the left column of Eq. (15) are coupled and cannot be chosen at will. The relations in the right column are uncoupled and can be chosen at will.

2.3. Generating phase screens (PS approach)

In this section, a method for generating the required discretized $ϕ_{x}$ and $ϕ_{y}$ is presented. Of the two approaches discussed above, the PS approach is the most applicable to laboratory research because of the commercial availability of phase-only SLMs. The specifications of commercial SLMs (size of active area, number of pixels, pixel pitch, etc.) vary by vendor. Here, the specifications of the Boulder Nonlinear Systems (BNS) Model P512-0635 SLM are adopted, i.e., 512 × 512 pixel array with a $15 μ m$ pitch [26]. These numbers are used in the simulation results presented in Section 3.

Let $ϕ$ and $\tilde{ϕ}$ be Fourier transform pairs, i.e.,

\begin{matrix} \tilde{ϕ} (f_{x}, f_{y}) = \int_{- \infty}^{\infty} \int_{- \infty}^{\infty} ϕ (x, y) \exp (- j 2 π f_{x} x) \exp (- j 2 π f_{y} y) d x d y \\ ϕ (x, y) = \int_{- \infty}^{\infty} \int_{- \infty}^{\infty} \tilde{ϕ} (f_{x}, f_{y}) \exp (j 2 π f_{x} x) \exp (j 2 π f_{y} y) d f_{x} d f_{y} . \end{matrix}

Since

ϕ_{α}

obeys Gaussian statistics,

\begin{matrix} 〈 ϕ_{x} (x, y) 〉 = 〈 ϕ_{y} (x, y) 〉 = 〈 ϕ_{α} (x, y) 〉 = 0 \\ 〈 ϕ_{α} (x_{1}, y_{1}) ϕ_{α}^{*} (x_{2}, y_{2}) 〉 = σ_{ϕ_{α}}^{2} \exp (- \frac{{| ρ_{1} - ρ_{2} |}^{2}}{ℓ_{ϕ_{α} ϕ_{α}}^{2}}) . \end{matrix}

The phase screen

ϕ_{α}

is real; the complex conjugate on the second term in the autocorrelation is provided only for completeness.

Expanding $ϕ_{α}$ in a Fourier series yields

\begin{matrix} ϕ_{α} (x, y) = Re [\sum_{m, n} φ_{α m n} \exp (j2 π \frac{m}{L} x) \exp (j2 π \frac{n}{L} y)] \\ = \sum_{m, n} φ_{α m n}^{r} \cos [\frac{2 π}{L} (m x + n y)] - \sum_{m, n} φ_{α m n}^{i} \sin [\frac{2 π}{L} (m x + n y)], \end{matrix}

where

φ_{α m n}

, the Fourier series coefficients, are zero mean circular complex Gaussian random numbers and

L = N Δ

is the size of the discrete grid. Here,

φ_{α m n}^{r}

and

φ_{α m n}^{i}

are the real and imaginary parts of

φ_{α m n}

, respectively.

Taking the autocorrelation of $ϕ_{α}$ , making use of the assumption that $φ_{α m n}$ are circular complex Gaussian random numbers, and simplifying yields

〈 ϕ_{α} (x_{1}, y_{1}) ϕ_{α}^{*} (x_{2}, y_{2}) 〉 = \sum_{m, n} \sum_{p, q} 〈 φ_{α m n}^{r} φ_{α p q}^{r} 〉 \cos [\frac{2 π}{L} (m x_{1} + n y_{1} - p x_{2} - q y_{2})] .

This expression must be equal to the autocorrelation of

ϕ_{α}

computed using Eqs. (16) and (17); therefore,

\begin{matrix} 〈 φ_{α m n}^{r} φ_{α p q}^{r} 〉 = 〈 φ_{α m n}^{i} φ_{α p q}^{i} 〉 = Φ_{ϕ_{α} ϕ_{α}} (\frac{m}{L}, \frac{n}{L}) δ_{m p} δ_{n q} \frac{1}{L^{2}} \\ 〈 {(φ_{α m n}^{r})}^{2} 〉 = 〈 {(φ_{α m n}^{i})}^{2} 〉 = Φ_{ϕ_{α} ϕ_{α}} (\frac{m}{L}, \frac{n}{L}) \frac{1}{L^{2}}, \end{matrix}

where

Φ_{ϕ_{α} ϕ_{α}} (f_{x}, f_{y}) = σ_{ϕ_{α}}^{2} π ℓ_{ϕ_{α} ϕ_{α}}^{2} \exp [- π^{2} ℓ_{ϕ_{α} ϕ_{α}}^{2} (f_{x}^{2} + f_{y}^{2})]

is the power spectral density of

ϕ_{α}

,

〈 {(φ_{α m n}^{r})}^{2} 〉

and

〈 {(φ_{α m n}^{i})}^{2} 〉

are the variances of the real and imaginary parts of the Fourier series coefficients

φ_{α m n}

, and

δ_{m p}

and

δ_{n q}

are Kronecker deltas. The desired phase screen

ϕ_{α}

can be produced by using Eq. (18), namely,

\begin{matrix} ϕ_{α} [i, j] = Re [\sum_{m, n} r_{α} [m, n] \frac{σ_{ϕ_{α}} \sqrt{π} ℓ_{ϕ_{α} ϕ_{α}}}{N Δ} \exp {- \frac{π^{2} ℓ_{ϕ_{α} ϕ_{α}}^{2}}{2} [{(\frac{m}{N Δ})}^{2} + {(\frac{n}{N Δ})}^{2}]} \\ \exp (j \frac{2 π}{N} m i) \exp (j \frac{2 π}{N} n j)], \end{matrix}

where

r_{α}

is a matrix of zero mean circular complex Gaussian random numbers with the real and imaginary parts each having unit variance.

In order to generate correlated $ϕ_{x}$ and $ϕ_{y}$ , necessary to synthesize the “cross” terms of the CSD matrix, the cross-correlation of Eq. (21) must be computed:

\begin{matrix} 〈 ϕ_{x} [i, j] ϕ_{y} [k, l] 〉 = \sum_{m, n} \sum_{p, q} \frac{σ_{ϕ_{x}} \sqrt{π} ℓ_{ϕ_{x} ϕ_{x}}}{N Δ} \exp {- \frac{π^{2} ℓ_{ϕ_{x} ϕ_{x}}^{2}}{2} [{(\frac{m}{N Δ})}^{2} + {(\frac{n}{N Δ})}^{2}]} \\ \frac{σ_{ϕ_{y}} \sqrt{π} ℓ_{ϕ_{y} ϕ_{y}}}{N Δ} \exp {- \frac{π^{2} ℓ_{ϕ_{y} ϕ_{y}}^{2}}{2} [{(\frac{p}{N Δ})}^{2} + {(\frac{q}{N Δ})}^{2}]} \\ 〈 {r_{x}^{r} [m, n] \cos (\frac{2 π}{N} (m i + n j)) - r_{x}^{i} [m, n] \sin (\frac{2 π}{N} (m i + n j))} \\ {r_{y}^{r} [p, q] \cos (\frac{2 π}{N} (p k + q l)) - r_{y}^{i} [p, q] \sin (\frac{2 π}{N} (p k + q l))} 〉, \end{matrix}

where

r^{r}

and

r^{i}

are the real and imaginary parts of

r

, respectively. Expanding the terms inside the angle brackets, letting

\begin{matrix} 〈 r_{x}^{r} [m, n] r_{y}^{r} [p, q] 〉 = 〈 r_{x}^{i} [m, n] r_{y}^{i} [p, q] 〉 = Γ δ_{m p} δ_{n q} \\ 〈 r_{x}^{r} [m, n] r_{y}^{i} [p, q] 〉 = 〈 r_{x}^{i} [m, n] r_{y}^{r} [p, q] 〉 = 0, \end{matrix}

where

0 \leq Γ \leq 1

is a correlation coefficient, and simplifying yields

\begin{array}{l} 〈 ϕ_{x} [i, j] ϕ_{y} [k, l] 〉 = \sum_{m, n} \frac{σ_{ϕ_{x}} σ_{ϕ_{y}} π ℓ_{ϕ_{x} ϕ_{x}} ℓ_{ϕ_{y} ϕ_{y}}}{{(N Δ)}^{2}} \frac{Γ}{2} \exp {- π^{2} (\frac{ℓ_{ϕ_{x} ϕ_{x}}^{2} + ℓ_{ϕ_{y} ϕ_{y}}^{2}}{2}) [{(\frac{m}{N Δ})}^{2} + {(\frac{n}{N Δ})}^{2}]} \\ {\exp (j \frac{2 π}{N} m (i - k)) \exp (j \frac{2 π}{N} n (j - l)) + \exp (- j \frac{2 π}{N} m (i - k)) \exp (- j \frac{2 π}{N} n (j - l))} . \end{array}

Note that the complex exponential terms in the braces are discrete inverse and forward Fourier transform kernels. The discrete function being transformed in Eq. (24), equivalent to the cross-power spectral density, is even in

m

and

n

; therefore, the forward and inverse Fourier transforms yield the same result. Applying these simplifications produces

\begin{matrix} 〈 ϕ_{x} [i, j] ϕ_{y} [k, l] 〉 = \sum_{m, n} σ_{ϕ_{x}} σ_{ϕ_{y}} π (Γ ℓ_{ϕ_{x} ϕ_{x}} ℓ_{ϕ_{y} ϕ_{y}}) \exp {- π^{2} (\frac{ℓ_{ϕ_{x} ϕ_{x}}^{2} + ℓ_{ϕ_{y} ϕ_{y}}^{2}}{2}) [{(\frac{m}{N Δ})}^{2} + {(\frac{n}{N Δ})}^{2}]} \\ \exp (j \frac{2 π}{N} m (i - k)) \exp (j \frac{2 π}{N} n (j - l)) \frac{1}{{(N Δ)}^{2}} . \end{matrix}

By comparing the discrete function being transformed in Eq. (25) to the continuous cross-power spectral density function, i.e.,

Φ_{ϕ_{x} ϕ_{y}} (f_{x}, f_{y}) = σ_{ϕ_{x}} σ_{ϕ_{y}} π ρ_{ϕ_{x} ϕ_{y}} ℓ_{ϕ_{x} ϕ_{y}}^{2} \exp [- π^{2} ℓ_{ϕ_{x} ϕ_{y}}^{2} (f_{x}^{2} + f_{y}^{2})],

one obtains the following relationships:

\begin{matrix} ℓ_{ϕ_{x} ϕ_{y}} = \sqrt{\frac{Γ ℓ_{ϕ_{x} ϕ_{x}} ℓ_{ϕ_{y} ϕ_{y}}}{ρ_{ϕ_{x} ϕ_{y}}}} = \sqrt{\frac{ℓ_{ϕ_{x} ϕ_{x}}^{2} + ℓ_{ϕ_{y} ϕ_{y}}^{2}}{2}} \\ Γ = \frac{ρ_{ϕ_{x} ϕ_{y}} (ℓ_{ϕ_{x} ϕ_{x}}^{2} + ℓ_{ϕ_{y} ϕ_{y}}^{2})}{2 ℓ_{ϕ_{x} ϕ_{x}} ℓ_{ϕ_{y} ϕ_{y}}} . \end{matrix}

Using Eq. (9), the general relationships between the EGSM source parameters and the phase screen design parameters are found to be

\begin{matrix} δ_{x x} = \frac{1}{\sqrt{2}} \frac{ℓ_{ϕ_{x} ϕ_{x}}}{σ_{ϕ_{x}}} \\ δ_{y y} = \frac{1}{\sqrt{2}} \frac{ℓ_{ϕ_{y} ϕ_{y}}}{σ_{ϕ_{y}}} \\ δ_{x y} = \frac{1}{\sqrt{2}} \frac{ℓ_{ϕ_{x} ϕ_{x}}^{2} + ℓ_{ϕ_{y} ϕ_{y}}^{2}}{\sqrt{4 Γ σ_{ϕ_{x}} σ_{ϕ_{y}} ℓ_{ϕ_{x} ϕ_{x}} ℓ_{ϕ_{y} ϕ_{y}}}} \\ | B_{x y} | = \exp [- \frac{1}{2} (σ_{ϕ_{x}}^{2} - \frac{4 Γ σ_{ϕ_{x}} σ_{ϕ_{y}} ℓ_{ϕ_{x} ϕ_{x}} ℓ_{ϕ_{y} ϕ_{y}}}{ℓ_{ϕ_{x} ϕ_{x}}^{2} + ℓ_{ϕ_{y} ϕ_{y}}^{2}} + σ_{ϕ_{y}}^{2})] . \end{matrix}

In the above equations,

ℓ_{ϕ_{x} ϕ_{x}}, ℓ_{ϕ_{y} ϕ_{y}} > 0

,

σ_{ϕ_{x}}, σ_{ϕ_{y}} \geq π

, and

0 < Γ \leq 1

.

Equation (28) expresses the four desired EGSM source parameters in terms of five phase screen design parameters; thus, the system of nonlinear equations is undetermined. Upon closer inspection of Eq. (28), one notes that three of the four desired EGSM parameters can be chosen at will (recall that $A_{x}$ , $A_{y}$ , and $∠ B_{x y}$ can be chosen freely). The values of the three chosen parameters then set the value of the remaining one. This is most evident if one decides to choose the values of $δ_{x x}$ , $δ_{y y}$ , and $δ_{x y}$ . The value of $| B_{x y} |$ is then set by the values of those other parameters. This is in contrast to previous EGSM synthesis research where $δ_{x y}$ was a function of $δ_{x x}$ and $δ_{y y}$ [18–20].

While Eq. (28) could be inverted in the manner just outlined, the optimal solution is not guaranteed. Here, the optimal solution is defined as the phase screen design parameters that yield EGSM parameters “nearest to” the desired EGSM parameters. Thus, in this work, the optimal phase screen design parameters are found using constrained nonlinear optimization.

2.4. Generating complex screens (CS approach)

In this section, a method for synthesizing discretized $T_{x}$ and $T_{y}$ is shown. Because both amplitude and phase must be controlled, the CS approach is much better suited to research involving simulation. For ease of comparison, the same SLM specifications listed above are used in the simulation results presented in Section 3.

Like $ϕ_{α}$ in the PS approach, the complex screen transmittances have zero mean and a Gaussian correlation function, i.e.,

\begin{matrix} 〈 T_{x} (x, y) 〉 = 〈 T_{y} (x, y) 〉 = 〈 T_{α} (x, y) 〉 = 0 \\ 〈 T_{α} (x_{1}, y_{1}) T_{α}^{*} (x_{2}, y_{2}) 〉 = σ_{T_{α}}^{2} \exp (- \frac{{| ρ_{1} - ρ_{2} |}^{2}}{ℓ_{T_{α} T_{α}}^{2}}) . \end{matrix}

Expanding

T_{α}

in a Fourier series yields

T_{α} (x, y) = \sum_{m, n} T_{α m n} \exp (j2 π \frac{m}{L} x) \exp (j2 π \frac{n}{L} y),

where

T_{α m n}

, the Fourier series coefficients, are zero mean circular complex Gaussian random numbers and

L = N Δ

is, again, the size of the discrete grid. Taking the autocorrelation of

T_{α}

produces

〈 T_{α} (x_{1}, y_{1}) T_{α}^{*} (x_{2}, y_{2}) 〉 = \sum_{m, n} \sum_{p, q} 〈 T_{α m n} T_{α p q}^{*} 〉 \exp [j \frac{2 π}{L} (m x_{1} - p x_{2})] \exp [j \frac{2 π}{L} (n y_{1} - q y_{2})] .

Like in the PS approach, Eq. (31) can be shown to be equal to the autocorrelation of $T_{α}$ , computed using similar Fourier transform relations as given in Eq. (16) and the expression given in Eq. (29). Performing the necessary analysis, one deduces that

\begin{matrix} 〈 T_{α m n} T_{α p q}^{*} 〉 = Φ_{T_{α} T_{α}} (\frac{m}{L}, \frac{n}{L}) δ_{m p} δ_{n q} \frac{1}{L^{2}} \\ 〈 {| T_{α m n} |}^{2} 〉 = Φ_{T_{α} T_{α}} (\frac{m}{L}, \frac{n}{L}) \frac{1}{L^{2}}, \end{matrix}

where

Φ_{T_{α} T_{α}} (f_{x}, f_{y}) = σ_{T_{α}}^{2} π ℓ_{T_{α} T_{α}}^{2} \exp [- π^{2} ℓ_{T_{α} T_{α}}^{2} (f_{x}^{2} + f_{y}^{2})]

is the power spectral density of

T_{α}

and

〈 {| T_{α m n} |}^{2} 〉

is the variance of the Fourier series coefficients

T_{α m n}

. The complex amplitude screen

T_{α}

can be produced by using Eq. (30), i.e.,

\begin{matrix} T_{α} [i, j] = \sum_{m, n} r_{α} [m, n] \frac{σ_{T_{α}} \sqrt{π / 2} ℓ_{T_{α} T_{α}}}{N Δ} \exp {- \frac{π^{2} ℓ_{T_{α} T_{α}}^{2}}{2} [{(\frac{m}{N Δ})}^{2} + {(\frac{n}{N Δ})}^{2}]} \\ \exp (j \frac{2 π}{N} m i) \exp (j \frac{2 π}{N} n j), \end{matrix}

where

r_{α}

is, again, a matrix of zero mean circular complex Gaussian random numbers with the real and imaginary parts each having unit variance.

In a manner completely analogous to the PS approach presented above, the cross-correlation of Eq. (33) must be computed. Using Eq. (23) and simplifying yields

\begin{matrix} 〈 T_{x} [i, j] T_{y}^{*} [k, l] 〉 = \sum_{m, n} σ_{T_{x}} σ_{T_{y}} π ℓ_{T_{x} T_{x}} ℓ_{T_{y} T_{y}} Γ \\ \exp {- π^{2} (\frac{ℓ_{T_{x} T_{x}}^{2} + ℓ_{T_{y} T_{y}}^{2}}{2}) [{(\frac{m}{N Δ})}^{2} + {(\frac{n}{N Δ})}^{2}]} \\ \exp [j \frac{2 π}{N} m (i - k)] \exp [j \frac{2 π}{N} n (j - l)] \frac{1}{{(N Δ)}^{2}} . \end{matrix}

By comparing the discrete function being inverse Fourier transformed in Eq. (34) to the continuous cross-power spectral density function, i.e.,

Φ_{T_{x} T_{y}} (f_{x}, f_{y}) = σ_{T_{x}} σ_{T_{y}} π ρ_{T_{x} T_{y}} ℓ_{T_{x} T_{y}}^{2} \exp [- π^{2} ℓ_{T_{x} T_{y}}^{2} (f_{x}^{2} + f_{y}^{2})],

one obtains the following relationships:

\begin{matrix} ℓ_{T_{x} T_{y}} = \sqrt{\frac{Γ ℓ_{T_{x} T_{x}} ℓ_{T_{y} T_{y}}}{ρ_{T_{x} T_{y}}}} = \sqrt{\frac{ℓ_{T_{x} T_{x}}^{2} + ℓ_{T_{y} T_{y}}^{2}}{2}} \\ Γ = \frac{ρ_{T_{x} T_{y}} (ℓ_{T_{x} T_{x}}^{2} + ℓ_{T_{y} T_{y}}^{2})}{2 ℓ_{T_{x} T_{x}} ℓ_{T_{y} T_{y}}} . \end{matrix}

Using Eq. (15), the general relationships between the EGSM source parameters and the complex screen design parameters are

\begin{matrix} δ_{x x} = \frac{ℓ_{T_{x} T_{x}}}{\sqrt{2}} \\ δ_{y y} = \frac{ℓ_{T_{y} T_{y}}}{\sqrt{2}} \\ δ_{x y} = \frac{1}{\sqrt{2}} \sqrt{\frac{ℓ_{T_{x} T_{x}}^{2} + ℓ_{T_{y} T_{y}}^{2}}{2}} \\ | B_{x y} | = \frac{2 Γ ℓ_{T_{x} T_{x}} ℓ_{T_{y} T_{y}}}{ℓ_{T_{x} T_{x}}^{2} + ℓ_{T_{y} T_{y}}^{2}} . \end{matrix}

In the above equations,

ℓ_{T_{x} T_{x}}, ℓ_{T_{y} T_{y}} > 0

and

0 < Γ \leq 1

.

It is clear from Eq. (37) that two of the three correlation function widths can be chosen freely (the third is set by the other two). One is generally free to choose the value of $| B_{x y} |$ subject to the constraint that $Γ \leq 1$ . The other EGSM source parameters, $A_{x}$ , $A_{y}$ , and $∠ B_{x y}$ , can be chosen at will.

3. Validation

3.1 Simulation description

In this section, simulation results are presented to validate the PS and CS approaches described above. As stated previously, 512 points per side and a spacing of $15 μ m$ were used to discretize the fields along paths 1 and 2 in Fig. 1. These numbers were chosen to match the BNS Model P512-0635 SLM. A wavelength of $λ = 632.8 nm$ was assumed. Two different EGSM sources were simulated. The first was a linearly, partially polarized EGSM source with the off-diagonal elements of the CSD matrix equal to zero. Since for this case $σ_{x} = σ_{y}$ , the polarization state was uniform across the source plane [2]. The second was an elliptically partially polarized EGSM source with a fully-populated CSD matrix. Table 1 reports the desired, PS, and CS EGSM source parameters for both cases.

Table 1. EGSM Source Parameters

View Table

The screen parameters for the PS and CS approaches were determined by inverting Eqs. (28) and (37), respectively. For the CS approach, Eq. (37) is easily inverted. When the off-diagonal elements of the desired CSD matrix are zero (Case I), the CS approach can generate an EGSM source with the desired parameters (note that $δ_{x y}$ is irrelevant in these cases). This is not guaranteed when the desired CSD matrix is fully populated (Case II), however.

For the PS approach, Eq. (28) is a coupled system of nonlinear equations and not easily inverted. Here, constrained nonlinear optimization was used to find the phase screen parameters such that

\underset{x \in ℝ}{\arg \min} \sqrt{{(\frac{δ_{x x}^{desired}}{δ_{x x} (x)} - 1)}^{2} + {(\frac{δ_{y y}^{desired}}{δ_{y y} (x)} - 1)}^{2} + {(\frac{δ_{x y}^{desired}}{δ_{x y} (x)} - 1)}^{2} + {(\frac{| B_{x y}^{desired} |}{| B_{x y} (x) |} - 1)}^{2}},

where

x

was a vector of the unknown phase screen parameters. The constraints on

x

included the conditions given in Eqs. (2) and (3) as well as positivity. In addition, to satisfy the “strongly scattering screen” requirement, i.e., the Gaussian approximation to the joint characteristic function [see Eq. (8)],

σ_{ϕ_{x}}, σ_{ϕ_{y}} \geq π

. Like in the CS approach, when the off-diagonal elements of the desired CSD matrix are zero (Case I), the PS approach can generate an EGSM source with the desired parameters. Again, this is not guaranteed when the desired CSD matrix is fully populated (Case II).

3.2 Simulation results

Figure 2 and Fig. 3 show the simulation results for Case I and II, respectively. The figures are organized such that the PS, CS, and theoretical results are along the columns—PS results are Figs. 2(a), 2(d), 2(g), 2(j) and Figs. 3(a), 3(d), 3(g), 3(j); CS results are Figs. 2(b), 2(e), 2(h), 2(k) and Figs. 3(b), 3(e), 3(h), 3(k) ; and theoretical results are Figs. 2(c), 2(f), 2(i), 2(l) and Figs. 3(c), 3(f), 3(i), 3(l) . Each row of images in Figs. 2 and 3 is a Stokes parameter— $S_{0}$ are Figs. 2(a)-2(c) and Figs. 3(a)-3(c); $S_{1}$ are Figs. 2(d)-2(f) and Figs. 3(d)-3(f); $S_{2}$ are Figs. 2(g)-2(i) and Figs. 3(g)-3(i); and $S_{3}$ are Figs. 2(j)-2(l) and Figs. 3(j)-3(l)—and on the same color scale specified by the color bar in each row. Lastly, the spectral degree of coherence $η$ is shown in Figs. 2(m) and 3(m). The PS and CS statistics were computed at the simulated EGSM source plane (see Fig. 1) using the results of 20,000 simulations. The theoretical Stokes parameters and $η$ are related to the CSD matrix elements by [27],

\begin{matrix} S_{0} (ρ) = W_{x x} (ρ, ρ) + W_{y y} (ρ, ρ) \\ S_{1} (ρ) = W_{x x} (ρ, ρ) - W_{y y} (ρ, ρ) \\ S_{2} (ρ) = W_{x y} (ρ, ρ) + W_{y x} (ρ, ρ) \\ S_{3} (ρ) = j [W_{y x} (ρ, ρ) - W_{x y} (ρ, ρ)] \\ η (ρ_{1}, ρ_{2}) = η (x_{1}, y_{1}, x_{2}, y_{2}) = \frac{Tr W (ρ_{1}, ρ_{2})}{\sqrt{Tr W (ρ_{1}, ρ_{1})} \sqrt{Tr W (ρ_{2}, ρ_{2})}}, \end{matrix}

where

Tr

is the trace of the CSD matrix

W

[10].

Fig. 2 Case I PS and CS simulation results versus theory. The rows are $S_{0}$ , $S_{1}$ , $S_{2}$ , $S_{3}$ , and $η$ , respectively, while the columns are the PS, CS, and theory results, respectively. Each row of images is on the same color scale specified by the color bar in each row.

Download Full Size | PDF

Fig. 3 Case II PS and CS simulation results versus theory. The rows are $S_{0}$ , $S_{1}$ , $S_{2}$ , $S_{3}$ , and $η$ , respectively, while the columns are the PS, CS, and theory results, respectively. Each row of images is on the same color scale specified by the color bar in each row.

Download Full Size | PDF

Overall, the agreement between the simulation and the theoretical results is quite excellent. Although the Case I $S_{2}$ and $S_{3}$ simulation results initially appear to conflict with the correct theoretical $S_{2}$ and $S_{3}$ values, one must note the color scales of the $S_{2}$ and $S_{3}$ images which are between 7 and 9 and 36 and 46 times smaller than the scales for the $S_{1}$ and $S_{0}$ images, respectively. Therefore, these results validate the proposed approaches to synthesizing EGSM sources.

4. Conclusion

Two random screen methods, the PS and CS approaches, for generating EGSM sources were developed. The relationships between the desired source parameters and the random screen parameters were derived and discussed. For the CS approach, these relations were easily inverted. Eight of the nine desired EGSM source parameters could be produced exactly—any two of $δ_{x x}$ , $δ_{y y}$ , or $δ_{x y}$ could be produced exactly (the remaining parameter’s value is set by the values of the other two). The CS approach is well suited for simulation purposes; however, it is difficult to implement in the laboratory because field amplitude (in addition to phase) must be controlled. A major advantage of this method is its ability to easily simulate non-Gaussian electromagnetic Schell-model sources.

For the PS method, the relations between the desired EGSM source parameters and the screen parameters formed a system of coupled nonlinear equations which could not be analytically inverted. Constrained nonlinear optimization was used to find the best solution. In theory, all nine EGSM source parameters could be produced exactly because the nonlinear system was underdetermined, i.e., there are more screen parameters than desired EGSM source parameters. However, because of the complexity of the inverse problem, the optimal parameters were (generally) slightly different than the desired EGSM parameters. The PS approach is well suited for both simulation and laboratory experiments. Future work is neededto generalize this approach to non-Gaussian electromagnetic Schell-model sources.

Both the PS and CS approaches were tested through numerical wave optics simulations. The simulation results showed excellent agreement with published theory, thus validating the proposed approaches. Future work will include implementation of the approaches in the laboratory.

Acknowledgments

This research was supported in part by an appointment to the Postgraduate Research Participation Program at the Air Force Institute of Technology administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and AFIT.

O. Korotkova’s research is supported by AFOSR (FA9550-12-1-0449) and ONR (N00189-12-T-0136).

D. Voelz's research is supported by the Air Force Office of Scientific Research (AFOSR) Multidisciplinary Research Program of the University Research Initiative (MURI) Grant FA9550-12-1-0449.

The views expressed in this paper are those of the authors and do not reflect the official policy or position of the U.S. Air Force, the Department of Defense, or the U.S. Government.

References and links

1. D. F. James, “Change of polarization of light beams on propagation in free space,” J. Opt. Soc. Am. A 11(5), 1641–1649 (1994). [CrossRef]

2. F. Gori, M. Santarsiero, G. Piquero, R. Borghi, A. Mondello, and R. Simon, “Partially polarized Gaussian Schell-model beams,” J. Opt. A, Pure Appl. Opt. 3(1), 1–9 (2001). [CrossRef]

3. S. Zhu, Y. Cai, and O. Korotkova, “Propagation factor of a stochastic electromagnetic Gaussian Schell-model beam,” Opt. Express 18(12), 12587–12598 (2010). [CrossRef] [PubMed]

4. J. C. G. de Sande, G. Piquero, M. Santarsiero, and F. Gori, “Partially coherent electromagnetic beams propagating through double-wedge depolarizers,” J. Opt. 16(3), 035708 (2014). [CrossRef]

5. Y. Zhu, D. Zhao, and X. Du, “Propagation of stochastic Gaussian-Schell model array beams in turbulent atmosphere,” Opt. Express 16(22), 18437–18442 (2008). [CrossRef] [PubMed]

6. O. Korotkova, M. Salem, and E. Wolf, “The far-zone behavior of the degree of polarization of electromagnetic beams propagating through atmospheric turbulence,” Opt. Commun. 233(4-6), 225–230 (2004). [CrossRef]

7. M. Salem, O. Korotkova, A. Dogariu, and E. Wolf, “Polarization changes in partially coherent electromagnetic beams propagating through turbulent atmosphere,” Waves Random Media 14(4), 513–523 (2004). [CrossRef]

8. O. Korotkova, “Scintillation index of a stochastic electromagnetic beam propagating in random media,” Opt. Commun. 281(9), 2342–2348 (2008). [CrossRef]

9. O. Korotkova, Random Beams: Theory and Applications (CRC, 2013).

10. E. Wolf, Introduction to the Theory of Coherence and Polarization of Light (Cambridge, 2007).

11. F. Gori, “Matrix treatment for partially polarized, partially coherent beams,” Opt. Lett. 23(4), 241–243 (1998). [CrossRef] [PubMed]

12. O. Korotkova, M. Salem, and E. Wolf, “Beam conditions for radiation generated by an electromagnetic Gaussian Schell-model source,” Opt. Lett. 29(11), 1173–1175 (2004). [CrossRef] [PubMed]

13. J. Tervo, T. Setälä, and A. T. Friberg, “Theory of partially coherent electromagnetic fields in the space-frequency domain,” J. Opt. Soc. Am. A 21(11), 2205–2215 (2004). [CrossRef] [PubMed]

14. H. Roychowdhury and O. Korotkova, “Realizability conditions for electromagnetic Gaussian Schell-model sources,” Opt. Commun. 249(4-6), 379–385 (2005). [CrossRef]

15. F. Gori, M. Santarsiero, R. Borghi, and V. Ramirez-Sanchez, “Realizability condition for electromagnetic Schell-model souces,” J. Opt. Soc. Am. A 25(5), 1016–1021 (2008). [CrossRef]

16. A. S. Ostrovsky, G. Martínez-Niconoff, V. Arrizón, P. Martínez-Vara, M. A. Olvera-Santamaría, and C. Rickenstorff-Parrao, “Modulation of coherence and polarization using liquid crystal spatial light modulators,” Opt. Express 17(7), 5257–5264 (2009). [CrossRef] [PubMed]

17. X. Xiao and D. Voelz, “Wave optics simulation of partially coherent and partially polarized beam propagation in turbulence,” Proc. SPIE 7464, 74640T (2009). [CrossRef]

18. G. Piquero, F. Gori, P. Romanini, M. Santarsiero, R. Borghi, and A. Mondello, “Synthesis of partially polarized Gaussian Schell-model sources,” Opt. Commun. 208(1-3), 9–16 (2002). [CrossRef]

19. F. Wang, G. Wu, X. Liu, S. Zhu, and Y. Cai, “Experimental measurement of the beam parameters of an electromagnetic Gaussian Schell-model source,” Opt. Lett. 36(14), 2722–2724 (2011). [CrossRef] [PubMed]

20. T. Shirai, O. Korotkova, and E. Wolf, “A method of generating electromagnetic Gaussian Schell-model beams,” J. Opt. A, Pure Appl. Opt. 7(5), 232–237 (2005). [CrossRef]

21. A. S. Ostrovsky, G. Rodríguez-Zurita, C. Meneses-Fabián, M. A. Olvera-Santamaría, and C. Rickenstorff-Parrao, “Experimental generating the partially coherent and partially polarized electromagnetic source,” Opt. Express 18(12), 12864–12871 (2010). [CrossRef] [PubMed]

22. P. Meemon, M. Salem, K. S. Lee, M. Chopra, and J. P. Rolland, “Determination of the coherency matrix of a broadband stochastic electromagnetic light beam,” J. Mod. Opt. 55(17), 2765–2776 (2008). [CrossRef]

23. S. Avramov-Zamurovic, C. Nelson, R. Malek-Madani, and O. Korotkova, “Polarization-induced reduction in scintillation of optical beams propagating in simulated turbulent atmospheric channels,” Waves Complex Random Media. in press.

24. S. Avramov-Zamurovic, C. Nelson, R. Malek-Madani, and O. Korotkova, “The dependence of the intensity PDF of a random beam propagating in the maritime atmosphere on source coherence,” Waves Complex Random Media 24(1), 69–82 (2014). [CrossRef]

25. J. W. Goodman, Statistical Optics (Wiley, 2000).

26. Boulder Nonlinear Systems, Inc., Spatial Light Modulators—XY Series (Retrieved November 13, 2014 from http://www.meadowlark.com/store/data_sheet/Datasheet_XYseries_SLM.pdf).

27. S. Sahin, Z. Tong, and O. Korotkova, “Sensing of semi-rough targets embedded in atmospheric turbulence by means of stochastic electromagnetic beams,” Opt. Commun. 283(22), 4512–4518 (2010). [CrossRef]

Case I ( $W_{x y} = W_{y x} = 0$ )
	$A_{x}$	$A_{y}$	$∠ B_{x y}$	$σ_{x}$ (mm)	$σ_{y}$ (mm)	$δ_{x x}$ (mm)	$δ_{y y}$ (mm)	$δ_{x y}$ (mm)	$\| B_{x y} \|$
Desired	1.2	1	0	0.4286	0.4286	0.1071	0.1429	0.1714	0
PS	1.2	1	0	0.4286	0.4286	*0.1071*	*0.1429*	*0.1714*	*5.2 ‰ 10⁻¹¹*
CS	1.2	1	0	0.4286	0.4286	*0.1071*	*0.1429*	*0.1263*	0

Case II (Fully-Populated CSD Matrix)
	$A_{x}$	$A_{y}$	$∠ B_{x y}$	$σ_{x}$ (mm)	$σ_{y}$ (mm)	$δ_{x x}$ (mm)	$δ_{y y}$ (mm)	$δ_{x y}$ (mm)	$\| B_{x y} \|$
Desired	1.3	1	$- π / 6$	0.4286	0.3750	0.1500	0.1607	0.1714	0.1500
PS	1.3	1	$- π / 6$	0.4286	0.3750	*0.1501*	*0.1608*	*0.1713*	*0.1500*
CS	1.3	1	$- π / 6$	0.4286	0.3750	*0.1500*	*0.1607*	*0.1554*	*0.1500*

Computational approaches for generating electromagnetic Gaussian Schell-model sources

Abstract

1. Introduction

2. Methodology

2.1 PS approach

2.2. CS approach

2.3. Generating phase screens (PS approach)

2.4. Generating complex screens (CS approach)

3. Validation

3.1 Simulation description

3.2 Simulation results

4. Conclusion

Acknowledgments

References and links

Cited By

Figures (3)

Tables (1)

Equations (39)

Optics Express