Abstract
The wavefield is investigated in a two-dimensional problem of nontangential incidence of high-frequency wave from a point source on an impedance contour with a jump in curvature. The field in the vicinity of the limit ray is described in detail. It is shown that expressions found within Kirchhoff’s method agree with results of the boundary layer method. Suitability areas of the expressions obtained are described, which is based upon a detailed study of geometry of the problem.
Similar content being viewed by others
Notes
Here, we use expressions (A.4) derived in Appendix for coordinates of a point lying on the contour.
This is seen from the obvious equivalence of the inequality \({1 \mathord{\left/ {\vphantom {1 {k{{\rho }}}}} \right. \kern-0em} {k{{\rho }}}} = {1 \mathord{\left/ {\vphantom {1 {k{{r}_{0}}}}} \right. \kern-0em} {k{{r}_{0}}}} + {1 \mathord{\left/ {\vphantom {1 {k{{r}_{1}}}}} \right. \kern-0em} {k{{r}_{1}}}} \ll 1\) to the pair of inequalities \({1 \mathord{\left/ {\vphantom {1 {k{{r}_{0}}}}} \right. \kern-0em} {k{{r}_{0}}}} \ll 1\) and \({1 \mathord{\left/ {\vphantom {1 {k{{r}_{1}}}}} \right. \kern-0em} {k{{r}_{1}}}} \ll 1\).
It was shown in [27] that in the case of a jump in curvature, upon substitution of expression (A.1) into the Helmholtz equation, a delta function and its derivative appear at the limit ray.
Expressions similar to (A.11) were obtained in the case of reflection of a plane wave from a contour with a nonsmooth curvature (both discontinuous [8] and continuous [20, 21]). The range of their applicability was described in these papers by inequalities similar to (57), but the fulfillment of (48) was not required.
REFERENCES
V. H. Weston, IRE Trans. Antennas Propag. 10 (6), 775 (1962).
V. H. Weston, IEEE Trans. Antennas Propag. 13 (4), 611 (1965).
L. Kaminetzky and J. B. Keller, SIAM J. Appl. Math. 22 (1), 109 (1972).
T. B. A. Senior, IEEE Trans. Antennas Propag. 20 (3), 323 (1972).
D. Bouche, Ann. Télécommun. 47 (9–10), 391 (1992).
Z. M. Rogoff and A. P. Kiselev, Wave Motion 33 (2), 183 (2001).
D. P. Bouche, P. Aguilera, and S. Loillier, in Proc. Int. Conf. on Electromagn. in Adv. Appl. (ICEAA), Verona, Sep. 11–15, 2017 (IEEE, New York, 2017), p 531.
E. A. Zlobina and A. P. Kiselev, Wave Motion 96, 102671 (2020).
A. V. Popov, in Proc. All-Union Simp. on Diffraction and Propagation of Waves, Leningrad, June 13–17, 1970 (Nauka, Leningrad, 1971), p. 171.
A. Michaeli, IEEE Trans. Antennas Propag. 38 (6), 929 (1990).
N. Ya. Kirpichnikova, V. B. Filippov, and A. S. Kirpichnikova, J. Math. Sci. 108 (5), 689 (2002).
A. S. Kirpichnikova and V. B. Philippov, IEEE Trans. Antennas Propag. 49 (12), 1618 (2001).
G. L. James, Geometrical Theory of Diffraction for Electromagnetic Waves (IEEE Electromagnetic. Ser. 1, London, 1986).
J. B. Keller, J. Opt. Soc. Am. 52 (5), 116 (1962).
V. A. Borovikov and B. Ye. Kinber, Geometrical Theory of Diffraction (Svyaz’, Moscow, 1978; Institute of Electrical Engineers, London, 1994).
A. S. Kryukovskii, D. S. Lukin, E. A. Palkin, and D. V. Rastyagaev, J. Commun. Technol. Electron. 51 (10), 1087 (2006).
A. S. Kryukovskii, Uniform Asymptotic Theory of Regional and Angular Wave Accidents (Ros. Novyi Univ., Moscow, 2013) [in Russian].
V. M. Babich and N. Ya. Kirpichnikova, Boundary Layer Method in Diffraction Problems (Leningr. Gos. Univ., Leningrad, 1974; Springer, Berlin, 1979) .
V. M. Babich and V. S. Buldyrev, Asymptotic Methods in Short-Wavelength Diffraction Theory (Nauka, Moscow, 1971; Alpha Science, Oxford, 2009).
E. A. Zlobina and A. P. Kiselev, St. Petersburg Math. J. 33 (2), 207 (2022).
E. A. Zlobina, Zap. Nauchn. Sem. POMI 493, 169 (2020).
L. M. Brekhovskikh and O. A. Godin, Acoustics of Layered Media I: Plane and Quasi-Plane Waves (Nauka, Moscow, 1989; Springer, Berlin, 1998).
Handbook of Mathematical Functions, Ed. by M. Abramovitz and I. Stegun (Dover, New York, 1971; Nauka, Moscow, 1979).
P. Ya. Ufimtsev, Theory of Edge Diffraction in Electromagnetics (Tech Science Press, Encino, 2003; BINOM. Laboratoriya Znanii, Moscow, 2012).
A. V. Popov, Sov. Phys. Acoust. 19 (4), 375 (1974).
A. Erdélyi, Asymptotic Expansions (Dover, New York, 1956; Fizmatlit, Moscow, 1962).
E. A. Zlobina and A. P. Kiselev, J. Math. Sci. 243 (5), 707 (2019).
ACKNOWLEDGMENTS
Authors are indebted to A.V. Popov for a helpful discussion.
Funding
The work was supported by the Russian Foundation for Basic Research, grant no. 20-01-00627.
Author information
Authors and Affiliations
Corresponding authors
Ethics declarations
The authors declare that they have no conflicts of interest.
APPENDIX
APPENDIX
ANALYSIS OF THE RAY FORMULA FOR REFLECTED WAVE
Let us analyze the classical (see, e.g., [19]) ray formula for the reflected wave \({{u}^{{{\text{ref}}}}}\) in the vicinity of the limit ray at moderate (\(\kappa {{\rho }} \sim 1\)) and small (\(\kappa {{\rho }} \ll 1\)) distances from the nonsmoothness point \(O\). This will allow us to determine the applicability areas of formulas obtained above.
A.1. Ray Method Formula
In the case of a smooth boundary, the ray method [19] gives the following formula for a specularly reflected wave:
Here, \(l_{{0,1}}^{{\text{*}}}\) are the distances from the source point \({{M}_{0}}\) and from the observation point \({{M}_{1}}\) to the point of specular reflection \(N{\text{*}}\), respectively, \({{\tau {\text{*}}}} = l_{0}^{{\text{*}}} + l_{1}^{{\text{*}}}\) is the value of the eikonal at \({{M}_{1}}\) (that is, the geometrical travel time), \({{\alpha {\text{*}}}}\) is the angle of incidence at \(N{\text{*}}\), is the value of curvature at \(N{\text{*}}\), \(R{\text{*}}\) is reflection coefficient at \(N{\text{*}}\) (see Fig. 3). For a contour whose curvature has a jump at the point \(O\), expression (A.1) correctly describes waves specularly reflected from smooth parts of the contour to the right and left of \(O\). However, expression (A.1) has a discontinuity on the limit ray and therefore is unsuitable in its narrow vicinityFootnote 4.
Let us simplify the ray formula (A.1) for observation points \({{M}_{1}}\) close to the limit ray, \(\left| {{{\delta \varphi }}} \right| \ll 1\), similarly to how it was done in [8, 20] for the case of a plane wave incidence. We will separately consider moderate and small distances.
A.2. Moderate Distance
First, let \(\kappa {{\rho }} \sim 1\). We will obtain an expression for the arc length \(s{\text{*}}\) corresponding to the reflection point \(N{\text{*}}\) (see Fig. 3). It is clear that for observation points from a narrow neighborhood of the limit ray (\(\left| {{{\delta \varphi }}} \right| \ll 1\)) reflection points lie near \(O\), that is, \(s{\text{*}}\) is small.
The law of specular reflection reads
where \({{\varphi }}_{{0,1}}^{{\text{*}}} = {{{{\varphi }}}_{{0,1}}}(s*)\) are the angles between the axis \(Ox\) and lines \({{N}_{{\text{*}}}}{{M}_{{0,1}}}\), \({{\psi{\text{*}}}} = {{\psi }}\left( {s{\kern 1pt}{\text{*}}} \right)\) is the tangential angle of the contour at the point \(N{\text{*}}\). It follows from (A.2) that \({{\tan\varphi }}_{0}^{{\text{*}}} = {\text{tan}}\left( {{{\varphi }}_{1}^{{\text{*}}} - 2{{\psi*{ *}}}} \right)\). This is written in terms of coordinates as
Here, \(X{\text{*}} = X\left( {s{\text{*}}} \right)\) and \(Y{\text{*}} = Y\left( {s{\text{*}}} \right)\) are Cartesian coordinates of the point \(N{\text{*}}\).
The contour consists of two arcs of circles (see (3)), hence
Substituting (A.4) into (A.3) and using the smallness of \(s{\text{*}}\), we straightforwardly derive
Considering \(\left| {{{\delta \varphi }}} \right| \ll 1\) and employing (44), after some algebra, from (A.5) we get
The expression (3) for the curvature \({\text{of}}\) the contour is used here.
Now let us simplify the phase of exponential function in the ray formula (A.1). For points \(N{\text{*}}\) close to \(O\), the value of eikonal \({{\tau{\text{*}}}} = l_{0}^{{\text{*}}} + l_{1}^{{\text{*}}}\) at the point \({{M}_{1}}\) is written quite similarly to (27):
Substituting here (A.6) and again taking into account the smallness of \({{\delta \varphi }}\), we obtain an expression for the phase in (A.1):
The correction term is small under condition (48).
Similarly, we transform the expression in the denominator of the formula (A.1):
We transfer the last term in (A.8) to the amplitude and obtain a simplified form of the ray formula for \(\kappa {{\rho }} \sim 1\) and \(\left| {{{\delta \varphi }}} \right| \ll 1\):
see (46). Formula (45) agrees with (A.9) in the area where inequalities (47) and (48) are satisfied.
A.3. Small Distance
Address now the case where the observation point \({{M}_{1}}\) is close to \(O\), \(\kappa {{\rho }} \ll 1\), but \(k{{\rho }} \gg 1\) (see (29)). Let us transform the formula (A.9). Obviously,
and then expression (A.9) takes the form
Formula (A.11) was obtained under the assumption that inequalities (48) and (57) hold.Footnote 5 In the main approximation (with respect to the parameter \(\kappa {{\rho }} \ll 1\)) the reflected wave coincides with a wave from an imaginary source corresponding to a flat boundary. The terms linear in curvature are of smaller order.
From (A.11) it is clear that the terms in (53) corresponding to \({{U}_{0}}\) and \({{U}_{1}}\), in the area of their suitability, match with the specularly reflected wave.
Rights and permissions
About this article
Cite this article
Zlobina, E.A., Kiselev, A.P. Transition Zone in High-Frequency Diffraction on Impedance Contour with Jumping Curvature. Kirchhoff’s Method and Boundary Layer Method. J. Commun. Technol. Electron. 67, 130–139 (2022). https://doi.org/10.1134/S1064226922020164
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1064226922020164