We report analytic formulas for the elements of the 2×2 cross-spectral density matrix of a stochastic electromagnetic anisotropic beam propagating through the turbulent atmosphere with the help of vector integration. From these formulas the changes in the spectral density (spectrum), in the spectral degree of polarization, and in the spectral degree of coherence of such a beam on propagation are determined. As an example, these quantities are calculated for a so-called anisotropic electromagnetic Gaussian Schell-model beam propagating in the isotropic and homogeneous atmosphere. In particular, it is shown numerically that for a beam of this class, unlike for an isotropic electromagnetic Gaussian Schell-model beam, its spectral degree of polarization does not return to its value in the source plane after propagating at sufficiently large distances in the atmosphere. It is also shown that the spectral degree of coherence of such a beam tends to zero with increasing distance of propagation through the turbulent atmosphere, in agreement with results previously reported for isotropic beams.
©2007 Optical Society of America
Studies of light beams propagating through the turbulent atmosphere are very important for many applications such as tracking, remote sensing, and free-space optical communications. In the past three decades, statistical properties of scalar partially coherent (stochastic) beams propagating in the atmospheric turbulence have been investigated extensively [1–9]. Quite recently a class of electromagnetic stochastic beams became of a great interest. After the unified theory of coherence and polarization had been presented by Wolf , it became possible to determine the changes in statistical properties of such beams, in particular, in the spectral density, in the spectral degree of coherence, and in their polarization properties. All aforementioned properties of electromagnetic stochastic beams can be determined from the basic quantity of the unified theory, i.e. from their 2×2 electric correlation matrix, also known as the cross-spectral density matrix [11, 12].
Until now the unified theory has been used only for the analysis of the changes in the statistical properties of isotropic electromagnetic beams propagating through the turbulent atmosphere [13–17]. In particular, it has been shown in these studies that after propagating at sufficiently long distances through the homogeneous and isotropic turbulent atmosphere all polarization properties, including the spectral degree of polarization of the beam, return to their values in the source plane. In very recent publications coherence and polarization properties of partially coherent general beams propagating in atmospheric turbulence have been investigated by Eyyuboglu et al. [18, 19]. The effects of coherence on anisotropic electromagnetic Gaussian-Schell model beams propagating through free space have also been studied .
In this paper, we derive a general analytic formula for the elements of cross-spectral density matrix of an anisotropic stochastic electromagnetic beam propagating in the homogeneous and isotropic turbulent atmosphere. The changes in the spectral density, in the spectral degree of coherence, and in the spectral degree of polarization of this class of beams on propagation in the atmosphere are investigated in details.
To illustrate the theory we consider an example illustrating the evolution of the statistical properties of a typical anisotropic electromagnetic Gaussian Schell-model beam propagating in the atmosphere with Kolmogorov power spectrum.
Let us consider a stochastic, statistically stationary, electromagnetic beam-like field generated by a source located in the plane z=0 and propagating into the positive half-space z>0, filled with the turbulent atmosphere (see Fig. 1).
The second-order correlation properties of such a beam at a pair of points r 1, r 2, can be characterized by the 2×2 electric cross-spectral density matrix 
Here the asterisk denotes the complex conjugate and the angular brackets stand for the ensemble average in the sense of coherence theory in the space-frequency domain . x and y are two mutually orthogonal directions perpendicular to the beam axis. E=(Ex, Ey) is a statistical ensemble of the fluctuating component of the transverse electric field, specified by the position vector r≡(ρ, z) (see Fig.1). The propagation of each of the two transverse components of the electromagnetic beam in the turbulent atmosphere can be treated by the use of the extended Huygens-Fresnel integral formula :
where k=2π/λ is the wave number, λ is the wavelength, ρ′ and ρ denote the two-dimensional vectors in the source plane and the output plane, respectively. ψ is the random part of the complex phase of a spherical wave propagating through the turbulent atmosphere.
where subscript m implies that the average is taken over the ensemble of statistical realizations of the turbulent medium. The term in the sharp brackets with the subscript m in Eq. (3) can be written as 
with ρ′d=ρ′ 1-ρ′ 2d, and ρ d=ρ 1-ρd. Here Dψ is the structure function of the random complex phase in Rytov’s representation, ρ0=(0 0.545C2nk2z)-3/5 is the coherence length of a spherical wave propagating in the turbulent medium and C2n is the structure parameter of the refractive index. It should be noted that Rytov’s phase structure function is accepted to be valid not only in the regime weak fluctuations atmospheric fluctautions, but also in the strong fluctuations regime. We have employed a quadratic approximation  for the Rytov’s phase structure function in order to obtain simpler and viewable analytical results.
where ρ′T 12=(ρ′T 1, ρ′T 2)=(x′1, y′1, x′2, y′2) and ρ T12=(ρ T1, ρ T2)=x1, y1, x2, y2) are four-dimensional vectors, T stands for the matrix transposition, Det is the determinant,
and I is a 2×2 unitary matrix.
It is to be noted that our results will reduce to those obtained in Ref. 24, if one lets ρ 1=ρ 2 (i.e. if ρ d=0).
3. An example
To illustrate the preceding analysis we now consider an anisotropic electromagnetic Schell-model beam. The elements of the cross-spectral density matrix of such a beam in the source plane z=0 are given by the expressions 
where S(0)i is the spectral density of the component Ei of the electric field and η(0)ij is the spectral degree of correlation between the components Ei and Ej. Further, if S(0)i and η(0)ij are both Gaussian functions then the source is called an electromagnetic Gaussian Schell-model source and the elements of its cross-spectral density matrix are then given by the expressions [13, 14]
Here the coefficients Ai, Aj, Bij and the variances σ2i, σ2j, δ2ij are independent of position but may depend on frequency (see Ref. 21, Sec. 5.3.2).
In order to use the method of vector integration, it will be convenient to rewrite Eq. (8) in the form
where ρ′12=(ρ 1′, ρ′2) and M′-1 ij is a 4×4 matrix (whose elements are complex numbers, in general) of the form
On substituting from Eq. (9) into Eq. (5), and after performing tedious vector integration, we obtain for the elements of the cross-spectral density matrix of a random electromagnetic beam propagating through the turbulent atmosphere the formula
where Ī is a the 4×4 unitary matrix. Equation (11) is the main result of this paper. From this formula we can now derive the formulas for the spectral density, the spectral degree of coherence and the spectral degree of polarization of an anisotropic electromagnetic Gaussian-Schell-model beam (see Ref. 10). The spectral density of the beam at the point (ρ, z) with ρ T 12=(ρT, ρT) is given by the formula
where Tr stands for the trace.
The spectral degree of coherence of the electric field at a pair of points r 1≡(ρ 1, z) and r 2≡(ρ 2,z) is given by the formula
where ρ T 11=(ρ T 1, ρ T 1) and ρ T 22=(ρ T 2, ρ T 2).
Finally, the spectral degree of polarization at the point (ρ, z) with ρ T 12=(ρ T, ρ T) is given by the formula
In Fig. 2 the changes in the normalized spectral density along the z-axis are shown in the limiting cases when atmospheric turbulence is locally extremely strong (C2n=10-12 m-2/3) and locally very weak (C2n=10-16 m-2/3). S0 is the spectral density in the source plane. From Fig. 2 one can see that the spectral density tends to zero after the beams propagate at sufficiently long distances from the source due to diffraction of the beam propagating through the turbulent atmosphere. We also can see that the drop in the spectral density is influenced more by the source coherence than by the values of C2n.
In Fig. 3 the changes in the spectral degree of coherence of random electromagnetic beams propagating through the turbulent atmosphere are shown for the case when the pair of field points are located symmetrically with respect to the z-axis (i. e. ρ 2=-ρ1 ; see Fig. 1). The initial spectral degrees of coherence are different in Figs. 3(a) and 3(b), but both of them tend to zero with increasing distance of propagation through the turbulent atmosphere. On the contrary, if the beam propagates in free space (i.e. C2n=0), the degree of coherence at a pair of field points tends to 1 in the far-zone. The result implies that the atmospheric turbulence destroys spatial coherence of stochastic electromagnetic beams and it is so at smaller distances from the source for larger values of C2n.
In Fig. 4 we illustrate the changes in the spectral degree of polarization along the z-axis (ρ 12=0) of electromagnetic anisotropic Gaussian Schell-model beams passing through the turbulent atmosphere with different structure parameters. In Fig. 4(a) σi and σj are assumed to be equal, the spectral degree of polarization of the beam returns to its value in the source plane after propagating a long distance that is in agreement with the previous results [13–17]. In Fig. 4(b) we show that the spectral degree of polarization of a beam generated by an anisotropic source, i.e. the source with σi≠σj, does not, in general, tend to the same value as it has in the source plane. This result is in striking difference with the one obtained previously for isotropic beams (Compare with Fig. 4(a)) and is independent of the local strength of atmospheric turbulence, i.e. of parameter C2n.
4. Concluding remarks
In summary, general analytic expressions for the elements of the cross-spectral density matrix of an anisotropic stochastic electromagnetic beam propagating in the homogeneous and isotropic turbulent atmosphere have been derived. With the help of numerical calculations we have then studied the changes in the statistical properties of an anisotropic electromagnetic beam of the Gaussian Schell-model type. In particular, we found that, unlike an isotropic beam, an anisotropic stochastic electromagnetic beam may not, in general, recover its polarization properties in the source after propagation in the turbulent atmosphere over sufficiently long propagation distance.
Of course, the method of vector integration can also be applied to study the propagation of other classes of partially coherent beams (such as pulsed beams, flattened beams, vortex beams, etc.) passing through the turbulent atmosphere or other linear isotropic and homogeneous media.
This work was supported by the National Natural Science Foundation of China under grant 60478041, and the program of the Ministry of Education of China for New Century Excellent Talents in University. One of the authors (D. Zhao now as a visiting scholar at the University of Rochester) was grateful to Professor E. Wolf and Professor M. Alonso for their useful discussions, and also grateful to the support by Pao Yu-Kong and Pao Zhao-Long Scholarship.
References and links
1. M. S. Belenkii, A. I. Kon, and V. L. Mironov, “Turbulent distortions of the spatial coherence of a laser beam,” Sov. J. Quantum Electron. 7, 287–290 (1977). [CrossRef]
2. S. C. H. Wang and M. A. Plonus, “Optical beam propagation for a partially coherent source in the turbulent atmosphere,” J. Opt. Soc. Am. 69, 1297–1304 (1979). [CrossRef]
3. Y. Baykal, M. A. Plonus, and S. J. Wang, “The scintillations for weak atmospheric turbulence using a partially coherent source beam,” Radio Sci. 18, 551–556 (1983). [CrossRef]
4. J. Wu, “Propagation of a Gaussian-Schell beam through turbulent media,” J. Mod. Opt. 37, 671–684 (1990). [CrossRef]
5. J. Wu and A. D. Boardman, “Coherence length of a Gaussian-Schell beam and atmospheric turbulence,” J. Mod. Opt. 38, 1355–1363 (1991). [CrossRef]
6. J. C. Ricklin and F. M. Davidson, “Atmospheric turbulence effects on a partially coherent Gaussian beam: implications for free-space laser communication,” J. Opt. Soc. Am. A 19, 1794–1802 (2002). [CrossRef]
7. G. Gbur and E. Wolf, “Spreading of partially coherent beams in random media,” J. Opt. Soc. Am. A 19, 1592–1598 (2002). [CrossRef]
8. T. Shirai, A. Dogariu, and E. Wolf, “Mode analysis of spreading of partially coherent beams propagating through atmospheric turbulence,” J. Opt. Soc. Am. A 20, 1094–1102 (2003). [CrossRef]
9. G. P. Berman and A. A. Chumak, “Photon distribution function for long-distance propagation of partially coherent beams through the turbulent atmosphere,” Phys. Rev. A 74, 013805 (2006). [CrossRef]
10. E. Wolf, “Unified theory of coherence and polarization of random electromagnetic beams,” Phys. Lett. A 312, 263–267 (2003). [CrossRef]
11. E. Wolf, “Correlation-induced changes in the degree of polarization, the degree of coherence, and the spectrum of random electromagnetic beams on propagation,” Opt. Lett. 28, 1078–1080 (2003). [CrossRef] [PubMed]
12. H. Roychowdhury and E. Wolf, “Determination of the electric cross-spectral density matrix of a random electromagnetic beam,” Opt. Commun. 226, 57–60 (2003). [CrossRef]
13. H. Roychowdhury, S. A. Ponomarenko, and E. Wolf, “Change in the polarization of partially coherent electromagnetic beams propagating through the turbulent atmosphere,” J. Mod. Opt. 52, 1611–1618 (2005). [CrossRef]
14. 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, 225–230 (2004). [CrossRef]
15. X. Ji, E. Zhang, and B. Lü, “Changes in the spectrum and polarization of polychromatic partially coherent electromagnetic beams in the turbulent atmosphere,” Opt. Commun. 275, 292–300 (2007). [CrossRef]
16. M. Salem, O. Korotkova, A. Dogariu, and E. Wolf, “Polarization changes in partially coherent EM beams propagating through turbulent atmosphere,” Waves in Random Media 14, 513–523 (2004). [CrossRef]
17. O. Korotkova, M. Salem, A. Dogariu, and E. Wolf, “Changes in the polarization ellipse of random electromagnetic beams propagating through turbulent atmosphere,” Waves in Random and Complex Media 15, 353–364 (2005). [CrossRef]
18. H. T. Eyyuboglu, Y. Baykal, and Y. Cai, “Complex degree of coherence for partially coherent general beams in atmospheric turbulence,” J. Opt. Soc. Am. A 24, 2891–2901 (2007). [CrossRef]
19. H. T. Eyyuboglu, Y. Baykal, and Y. Cai, “Degree of polarization for partially coherent general beams in turbulent atmosphere,” Appl. Phys. B 89, 91–97 (2007). [CrossRef]
21. L. Mandel and E. Wolf, Optical Coherence and Quantum Optics (Cambridge University Press, Cambridge, 1995).
22. L. C. Andrews and R. L. Phillips, Laser Beam Propagation through Random Media (SPIE Press, Bellingham, Wash., 1998).
24. Y. Cai and S. He, “Propagation of a partially coherent twisted anisotropic Gaussian Schell-model beam in a turbulent atmosphere,” Appl. Phys. Lett. 89, 041117 (2006). [CrossRef]