Professional Documents
Culture Documents
Electromagnetic Fields and Waves in Fractional Dimensional Space PDF
Electromagnetic Fields and Waves in Fractional Dimensional Space PDF
and Technology
Electromagnetic Fields
and Waves in Fractional
Dimensional Space
123
Muhammad Zubair Qaisar Abbas Naqvi
Faculty of Electronic Engineering Department of Electronics
GIK Institute of Engineering Quaid-e-Azam University
Sciences and Technology Islamabad
Topi Pakistan
Pakistan e-mail: nqaisar@yahoo.com
e-mail: zubair_wah@hotmail.com
To my father
Mr. Abdul Ghafoor Mughal
for his love and kindness when he was alive
and his beautiful memories when he is no
longer with us
M. J. Mughal
To my parents
Q. A. Naqvi
Preface
ix
x Preface
This book has been divided into six chapters. In Chap. 2, a novel generalization
of differential electromagnetic equations in fractional space is provided on the
basis of modified vector differential operators for fractional space. A new form of
vector differential operator Del, written as rD, and its related differential operators
is formulated in fractional space. Using these modified vector differential opera-
tors, the classical Maxwells electromagnetic equations have been worked out. The
Laplaces, Poissons and Helmholtzs equations in fractional space are also
derived by using modified vector differential operators. Also a new fractional
space generalization of potentials for static and time-varying fields is presented.
Most of the work in later chapters is related to the solution of the established
differential electromagnetic equations in fractional space.
In Chap. 3, a fractional space generalization of potentials for static and time-
varying fields is presented by solving Laplaces equation and inhomogeneous
vector wave equation, respectively, in fractional space.
In Chap. 4, the phenomenon of wave propagation in fractional space is
investigated by solving Helmholtzs equation in different coordinate systems.
General plane wave solutions, in source-free and lossless as well as lossy media, in
fractional space are presented by solving vector wave equation in D-dimensional
fractional space. An exact solution of cylindrical as well as spherical wave
equation, for electromagnetic field in D-dimensional fractional space, is also
presented. All these investigated solutions of vector wave equation provide a basis
for the application of the concept of fractional space to the wave propagation
phenomenon in fractal media. For all investigated cases when integer dimension is
considered, the classical results were recovered to validate obtained results.
Chapter 5 deals with the solution procedure for radiation problems in fractional
space.The proposed solution procedure can be used to study the radiation phe-
nomenon in any non-integer dimensional fractal media. As an application, the
fields radiated by a Hertzian dipole in fractional space have been worked out.
Finally, conclusions are drawn in Chap. 6.
In summary, the subject covered in this book is relatively new and emerging
area of research in the field of electromagnetics. The concept of fractional
dimensional space has potential to make a significant impact on future directions
in fractional electromagnetics research. We highly recommend this book to
graduate students, researchers, and professionals working in the areas of electro-
magnetic-wave propagation, radiation, scattering, diffraction, and other related
fields of applied mathematics. The topics in this book can also be covered in any
graduate course on Advanced Engineering Electromagnetics.
xi
Contents
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Fractional Dimensional Space . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Axiomatic Basis for Fractional Dimensional Space . . . . . . . . . . 3
1.3 Differential Geometry of Fractional Dimensional Space . . . . . . . 4
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
xiii
xiv Contents
6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
About the Authors
Mr. Muhammad Zubair did his BS in Electronic Engineering with Gold Medal
from International Islamic University, Islamabad, Pakistan in 2009. Recently, he
has completed his MS in Electronic Engineering with Highest Distinction from
GIK Institute of Engineering Sciences and Technology, Topi, Pakistan in 2011 and
joined the same institute as Research Associate. Mr. Zubairs research interests are
in the field of Analytical Electromagnetics. He has applied the concept of frac-
tional dimensional space in the study of electromagnetic wave propagation,
radiation and scattering in fractal media. He is also member of Pakistan Engi-
neering Council.
Dr. M. Junaid Mughal did his M.Sc and M.Phil in Electronics from Quaid-e-
Azam university, Islamabad in 1993 and 1995, respectively. He did his PhD from
the University of Birmingham, UK in 2001. He worked as Director of Engineering
in Nuonics Inc., Orlando, Fl, USA form 2001 to 2003. He is presently working as
Associate Professor in the Faculty of Electronic Engineering in GIK Institute.
Dr. Mughals research interests are primarily in the field of communications and
particularly in RF and Optical Communications. He has worked in antennas, EM
scattering, propagation modeling for mobile applications and fiber optics. In the
field of optical communication Dr Mughal is coinventor of high dynamic range
variable optical attenuators based on Acousto-Optic and MEMS technology, high
speed fiber-optic switches, fiber optic tunable filters and laser beam profiling
systems. Currently he is working in the area of tunable metamaterials, wave
propagation in fractal media and focusing systems embedded in Chiral medium.
Dr. Qaisar Abbas Naqvi completed his M.Sc., M.Phil., and Ph.D., all in
Electronics, from Department of Electronics, Quaid-i-Azam University, Islama-
bad, Pakistan in 1991, 1993, and 1998 respectively. He joined Department of
Electronics as Assistant Professor in 1998. Uptill now, he has successfully
supervised more than thirty M.Phil and nine PhD students. He is now Associate
Professor and Chairman of Department of Electronics. He is author of more than
xvii
xviii About the Authors
100 papers in international refereed journals. He is also serving as referee for more
than 10 international journals. His research interests include fractional paradigm in
electromagnetics, bi-isotropic and chiral mediums, high frequency techniques and
Kobayashi potential method.
Chapter 1
Introduction
Every one of us has learnt that the lines and curves are one-dimensional, planes and
surfaces are two-dimensional, solids and volumes are three dimensional, and so on.
In a formal way, we say that a set is n-dimensional if we have n independent variables
to describe a neighborhood of any point. Such a notion of dimension is called the
topological dimension of a set. Also we observe that if we take the union of infinite
many sets of n dimension, the overall dimension of new set can grow to n+1 e.g., the
union of infinite number of (one-dimensional) lines give rise to a (two-dimensional)
plane.
Now, we think about the dimension in another way. We may break a line segment
into 4 self-similar intervals, each with the same length, and each of which can be
magnified by a factor of 4 to yield the original segment. We can also break a line
segment into 7 self-similar pieces, each with magnification factor 7 can yield the
original segment, or 20 self-similar pieces with magnification factor 20 to yield the
original segment. In general, we can break a line segment into N self-similar pieces,
each with magnification factor N to yield the original segment. If, we decompose a
square into 4 self-similar sub-squares, and the magnification factor here will be 2 to
yield in the original square. Alternatively, we can break the square into 9 self-similar
pieces with magnification factor 3, or 25 self-similar pieces with magnification factor
5. Clearly, the square may be broken into N 2 self-similar copies of itself, each of
which must be magnified by a factor of N to yield the original figure. Finally, we
can decompose a cube into N 3 self-similar pieces, each of which has magnification
factor N. Following above discussion we can say that the dimension is simply the
exponent of the number of self-similar pieces with magnification factor N into which
the figure may be broken.
So what is the dimension of the Sierpinski triangle? How do we find the exponent
in this case? For this, we need logarithms.
Note that, for the square, we have N 2 self-similar pieces, each with magnification
factor N. So we can write
In general, the Sierpinski triangle breaks into 3N self-similar pieces with magnifica-
tion factors 2N , so we again have
This estimates that the Sierpinski triangle sits somewhere in between lines and planes.
Similarly, many fractal structures are known in literature that possess a fractal dimen-
sion. Roughly speaking, we state that the space embedding such fractal curves or
surfaces is known as fractional dimensional space.
There has been much interest to study different physical phenomenon in fractional
dimensional space [128] during the last few decades. The concept of fractional space
is used to replace the real anisotropic confining structure with an isotropic fractional
1.1 Fractional Dimensional Space 3
This work is based on the Stillingers [6] axiomatic basis for spaces with non-integer
dimension. Here, we briefly describe these axioms. In [6], four topological axioms
are proposed which generate a space with non-integer dimension D. Let SD denote
the fractional space which contains points x, y, . . . and has topological structure
specified by the following axioms:
Axiom 1 SD is a metric space.
4 1 Introduction
The full implication of Axiom 4 is that any two points in SD are connected by a
continuous line embedded in that space so SD is connected. So any convex or star
domain in SD will be contractible. Based on these axioms Stillinger [6] as well as
recently Muslih [18], provided a dimensional regularization technique in order to
convert any integral of a function from fractional dimensional space to a regular
dimensional space. The fractional space generalization of the Laplacian operator,
provided in later chapters, is based on the same dimensional regularization technique
according to which a fractional space is related to fractional integrals and derivatives.
i /2
d i xi = |x|i 1 dxi , i = 1, 2, . . . , N
(i /2)
This space is called fractional dimensional space or simply a fractional space, and
D = 1 + 2 +, . . . , +N is the dimension of the fractional space. Following
the above discussion, any volume element in D-dimensional fractional space can
be defined as:
d(V )D = d 1 x1 , d 2 x1 , . . . , d N xN
References
1. M. Zubair, M.J. Mughal, Q.A. Naqvi, The wave equation and general plane wave solutions in
fractional space. Prog. Electromagnet. Res. Lett. 19, 137146 (2010)
2. M. Zubair, M.J. Mughal, Q.A. Naqvi, On electromagnetic wave propagation in fractional space.
Non-linear Anal. B: Real World App. 12(5), 28442850 (2011)
3. M. Zubair, M.J. Mughal, Q.A. Naqvi, A.A. Rizvi, Differential electromagnetic equations in
fractional space. Prog. Electromagnet. Res. 114, 255269 (2011)
4. M. Zubair, M.J. Mughal, Q.A. Naqvi, An exact solution of cylindrical wave equation for
electromagnetic field in fractional dimensional space. Prog. Electromagnet. Res. 114, 443455
(2011)
5. M. Zubair, M.J. Mughal, Q.A. Naqvi, An exact solution of spherical wave in D-dimensional
fractional space. J. Electromagn. Waves App. 25, 14811491 (2011)
6. F.H. Stillinger, Axiomatic basis for spaces with noninteger dimension. J. Math. Phys. 18(6),
12241234 (1977)
7. X. He, Anisotropy and isotropy: a model of fraction-dimensional space. PSolid State Commun.
75, 111114 (1990)
8. C. Palmer, P.N. Stavrinou, Equations of motion in a noninteger-dimension space. J. Phys. A
37, 69877003 (2004)
9. K.G. Willson, Quantum field-theory, models in less than 4 dimensions. Phys. Rev. D 7(10),
29112926 (1973)
10. B. Mandelbrot, The Fractal Geometry of Nature. (W.H. Freeman, New York, 1983)
11. C.G. Bollini, J.J. Giambiagi, Dimensional renormalization: The number of dimensions as a
regularizing parameter. Nuovo Cimento B 12, 2026 (1972)
12. J.F. Ashmore, On renormalization and complex space-time dimensions. Commun. Math. Phys.
29, 177187 (1973)
13. O.P. Agrawal, Formulation of Euler-Lagrange equations for fractional variational problems. J.
Math. Anal. Appl. 271(1), 368379 (2002)
14. D. Baleanu, S. Muslih, Lagrangian formulation of classical fields within Riemann-Liouville
fractional derivatives. Phys. Scripta 72(23), 119121 (2005)
6 1 Introduction
15. V.E. Tarasov, Electromagnetic fields on fractals. Modern Phys. Lett. A 21(20), 15871600
(2006)
16. V.E. Tarasov, Continuous medium model for fractal media. Phys. Lett. A 336(23), 167174
(2005)
17. S. Muslih, D. Baleanu, Fractional multipoles in fractional space. Nonlinear Anal: Real World
App. 8, 198203 (2007)
18. S.I. Muslih, O.P. Agrawal, A scaling method and its applications to problems in fractional
dimensional space. J. Math. Phys. 50(12):123501123511 (2009)
19. D. Baleanu, A.K. Golmankhaneh, A.K. Golmankhaneh, On electromagnetic field in fractional
space. Nonlinear Anal: Real World App. 11(1):288292 (2010)
20. Z. Wang, B. Lu, The scattering of electromagnetic waves in fractal media. Waves Random
Complex Media 4(1), 97103 (1994)
21. C.M. Bender, K.A. Milton, Scalar casimir effect for a D-dimensional sphere. Phys. Rev. D 50,
65476555 (1994)
22. S. Muslih, D. Baleanu, Mandelbrot scaling and parametrization invariant theories. Romanian
Rep. Phys. 62(4), 689696 (2010)
23. S. Muslih, M. Saddallah, D. Baleanu, E. Rabe, Lagrangian formulation of maxwells field in
fractional D dimensional space-time. Romanian Rep. Phys. 55(78), 659663 (2010)
24. S. Muslih, O.P. Agrawal, Riesz fractional derivatives and fractional dimensional space. Int. J.
Theor. Phys. 49(2):270275 (2010)
25. S. Muslih, Solutions of a particle with fractional [delta]-potential in a fractional dimensional
space. Int. J. Theor. Phys. 49(9), 20952104 (2010)
26. E. Rajeh, S.I. Muslih, B. Dumitru, E. Rabei, On fractional Schrodinger equation in [alpha]-
dimensional fractional space. Nonlinear Anal.: Real World App. 10(3):12991304 (2009)
27. M. Sadallah, S.I. Muslih, Solution of the equations of motion for einsteins field in fractional
D dimensional space-time. Int. J. Theor. Phys. 48(12):33123318 (2009)
28. S. Muslih, D. Baleanu, E.M. Rabe, Solutions of massless conformal scalar field in an
n-dimensional einstein space. Acta Phys. Pol. Ser. B 39(4):887892 (2008)
29. K.B. Oldham, J. Spanier, The Fractional Calculus. (Academic Press, New York, 1974)
30. A. Hussain, Q.A. Naqvi, Fractional rectangular impedance waveguide. Prog. Electromagnet.
Res. 96, 101116 (2009)
31. Q.A. Naqvi, Planar slab of chiral nihility metamaterial backed by fractional dual/PEMC inter-
face. Prog. Electromagnet. Res. 85, 381391 (2008)
32. Q.A. Naqvi, Fractional dual interface in chiral nihility medium. Prog. Electromagnet. Res. Lett.
8, 135142 (2009)
33. Q.A. Naqvi, Fractional dual solutions in grounded chiral nihility slab and their effect on outside
fields. J. Electromagn. Waves App. 23(56), 773784(12) (2009)
34. A. Naqvi, S. Ahmed, Q.A. Naqvi, Perfect electromagnetic conductor and fractional dual inter-
face placed in a chiral nihility medium. J. Electromagn. Waves App. 24(1415), 19911999(9)
(2010)
35. A. Naqvi, A. Hussain, Q.A. Naqvi, Waves in fractional dual planar waveguides containing
chiral nihility metamaterial. J. Electromagn. Waves App. 24(1112), 15751586(12) (2010)
36. E.I. Veliev, M.V. Ivakhnychenko, T.M. Ahmedov, Fractional boundary conditions in plane
waves diffraction on a strip. Prog. Electromagn. Res. 79, 443462 (2008)
37. S.A. Naqvi, M. Faryad, Q.A. Naqvi, M. Abbas, Fractional duality in homogeneous bi-isotropic
medium. Prog. Electromagn. Res. 78, 159172 (2008)
Chapter 2
Differential Electromagnetic Equations
in Fractional Space
where r (x0 , x1 ) is the distance between two points x0 and x1 , and weight W (r )
given by
W (r ) = (D)r D1 (2.2)
with
2 D/2
(D) = (2.3)
(D/2)
In Eq. 2.4 and throughout the discussion, the subscript D is used to emphasize the
dimension of space in which this operator is defined. An extension of formalism in
Eq. 2.1 to two variable integration yields an expression for a two-coordinate Laplacian
operator in fractional space.
2 2 D2
D
2
= + + , 0<D2 (2.5)
x 2 y 2 y y
In [2] the results from [1] are generalized to n orthogonal coordinates and Laplacian
operator in D-dimensional fractional space in three-spatial coordinates is given as:
2 1 1 2 2 1
D
2
= + + 2+
x 2 x x y y y
(2.6)
2 3 1
+ 2+
z z z
where, three parameters (0 < 1 1, 0 < 2 1 and 0 < 3 1) are used to
describe the measure distribution of space where each one is acting independently
on a single coordinate and the total dimension of the system is D = 1 + 2 + 3 .
It is obvious that for three dimensional space (D = 3), if we set 1 = 2 = 3 = 1
in (2.6), the fractional Laplacian operator D2 reduces to the classical Laplacian
From Eq. 2.6, we consider single variable Laplacian operator in fractional space:
2.2 Fractional Space Generalization of Del Operator and Related Differential Operators 9
2 D1
D
2
= + , 0<D1 (2.7)
x2 x x
We wish to find an expression for Del operator D in fractional space. As
D = | D | D (2.8)
In single variable case we assume D = x also | D | = D
2 , because . =
D D
2 , where 2 is given in (2.7):
D D
2 D1
| D | = + (2.9)
x 2 x x
Expansion of (2.9) using Binomial series expansion [3] for | x |>> 1, ignoring
terms involving second or higher degree of x in denominator, leads to the following
form:
1 D1
| D | = + (2.10)
x 2 x
From (2.8) and (2.10), Del operator in single variable x with fractional dimension D
is given by:
1 D1
D = + x (2.11)
x 2 x
The gradient of a scalar field in fractional space is a vector that represents both the
magnitude and the direction of maximum space rate of increase of in fractional
space. Using (2.12) the modified form of the gradient of scalar field , written as
grad D , in far-field region in the fractional space is given as:
1 (1 1) 1 (2 1)
grad D = D = + x + + y
x 2 x y 2 y
1 (3 1)
+ + z (2.13)
z 2 z
Fx 1 (1 1)Fx Fy 1 (2 1)Fy
div D F = D F = + + +
x 2 x y 2 y
Fz 1 (3 1)Fz
+ + (2.14)
z 2 z
curl D F = D F
Fz 1 (2 1)Fz Fy 1 (3 1)Fy
= + + x
y 2 y z 2 z
Fx 1 (3 1)Fx Fz 1 (1 1)Fz
+ + + y
z 2 z x 2 x
Fy 1 (1 1)Fy Fx 1 (2 1)Fx
+ + + z (2.15)
x 2 x y 2 y
or
x y z
1 1 1 1 2 1 1 3 1
curl D F = D F= + + + (2.16)
x 2 x y 2 y z 2 z
Fx Fy Fz
2.3 Fractional Space Generalization of Differential Maxwells Equations 11
The Maxwells equations are the fundamental equations describing the behavior of
electric and magnetic fields. In classical electromagnetic theory following quantities
are dealt with:
E = electric field intensity (V /m)
B = magnetic field intensity (A/m)
D = electric flux density (C/m 2 )
B = magnetic flux density (W/m 2 )
J = electric current density (A/m 2 )
v = electric charge density (C/m 3 )
with B = H and D = E, where and are permeability and permittivity of
the medium, respectively.
All of these quantities are functions of space variables x, y, z and tim t. The basic
classical Maxwells equations in differential form in Euclidean space are [4]:
D = v (2.17)
B=0 (2.18)
B
E= (2.19)
t
D
H=J+ (2.20)
t
v
J= (2.21)
t
div D D = v (2.22)
div D B = 0 (2.23)
B
curl D E = (2.24)
t
D
curl D H = J + (2.25)
t
12 2 Differential Electromagnetic Equations in Fractional Space
v
div D J = (2.26)
t
where, div D and curl D are defined in Eqs. 2.142.16. Equations 2.222.25 provide
generalization of classical Maxwells equations form integer dimensional Euclidean
space to a non-integer dimensional fractional space. For D = 3, these fractional
equations can be reduced to classical Maxwells equations in Euclidean space.
In phasor form, assuming a time factor e jt , Maxwells equations in fractional
space are given by replacing t with j [4] as below:
div D Ds = vs (2.27)
div D Bs = 0 (2.28)
From Maxwells equations in previous section, it is shown that the behavior of elec-
trostatic field in fractional space can be described by two differential equations:
v
div D E = (2.32)
0
curl D E = 0 (2.33)
where, 0 is permittivity of free space. Equation 2.33 being equivalent to the statement
that E is the gradient of a scalar function, the scalar potential for electric field .
Because
so,
E = grad D (2.35)
A detailed proof of Eq. 2.34 is provided in Appendix A. Equations 2.32 and 2.35 can
be combined into one partial differential equation for the single function (x, y, z)
as follows:
v
div D grad D = (2.36)
0
As div D grad D = D
2 , so finally we get
v
D
2
= (2.37)
0
2.37 is called Poissons equation in fractional space. In regions of space that lack a
charge density, the scalar potential satisfies the Laplaces equation given by:
D
2
=0 (2.38)
div D B = 0 (2.39)
curl D H = J (2.40)
From Eq. 2.40 we say that in problems concerned with finding the magnetic fields
in a current free region, the curl D of magnetic field H is zero. Any vector with zero
curl D may be represented as the grad D of a scalar (see e.g., Eq. 2.34). Thus, the
magnetic field for points in such regions can be expressed as
H = grad D m (2.41)
where, m (in amperes) is the magnetic scalar potential and the minus sign is taken
to complete the analogy with electrostatic field in (2.35).
From (2.39), the divergence of B is zero everywhere, so using (2.39) and (2.41)
Thus for a homogenous medium in fractional space the magnetic scalar potential m
satisfies the Laplace equation:
14 2 Differential Electromagnetic Equations in Fractional Space
D
2
m = 0 (2.43)
From (2.39) we know that for magnetostatic field div D B = 0. Also we know that
In order to satisfy (2.39) and (2.44) simultaneously, we can define vector magnetic
potential A (in webers/meter) such that
B = curl D A (2.45)
with
div D A = 0 (2.48)
in (2.46) we get
D
2
A = J (2.49)
A we have seen, in Maxwells equations fields are related to each other and sources
as well. But sometimes it is helpful to introduce some intermediate functions, known
as potential functions, which are directly related to sources and from which we can
drive fields [4]. Such function are found useful for static fields as well (see e.g., Eqs.
2.35, 2.41, 2.45).
From (2.45) we have B = curl D A. This relation may now be substituted into
Maxwells equation (2.24) to get
A
curl D E + =0 (2.50)
t
2.5 Fractional Space Generalization of Potentials for Static Fields 15
Equation 2.50 states that curl D of a certain quantity is zero. But this condition allows
a vector to be derived as a grad D of a scalar .
A
E+ = grad D (2.51)
t
A
E = grad D (2.52)
t
Equation 2.45 and 2.52 are the valid relationships between fields and potential func-
tions A and . We substitute (2.52) into (2.22), to obtain
(div D A) v
D
2
= (2.53)
t
Then by substituting (2.45) and (2.52) into (2.53), we get
2A
curl D curl D A = J + grad D 2 (2.54)
t t
div D A = (2.55)
t
Equation 2.53 and 2.54 can be reduced to
2 v
D
2
= (2.56)
t 2
2A
D
2
A = J (2.57)
t 2
Thus the potential functions A and , defined in terms of sources J and v by the
Eqs. 2.56 and 2.57 in fractional space, may be used to drive electric and magnetic
fields using (2.45) and (2.52).
H
curl D E = (2.58)
t
E
curl D H = J + (2.59)
t
Taking curl D of Eq. 2.58 on both sides and using (2.59) gives
16 2 Differential Electromagnetic Equations in Fractional Space
E
curl D curl D E = J+ (2.60)
t t
This result can be simplified using (2.47) and (2.26) in (2.60) as :
1 J 2E
D
2
E= grad D v + + 2 (2.61)
t t
For source-free region (v = 0, J = 0) (2.61) becomes
2E
D
2
E =0 (2.62)
t 2
Equation 2.62 is the Helmholtzs equation, or wave equation, for E in fractional
space. An identical equation for H in fractional space can also be derived in the same
manner:
2H
D
2
H =0 (2.63)
t 2
2.7 Summary
References
1. F.H. Stillinger, Axiomatic basis for spaces with noninteger dimension. J. Math. Phys. 18(6),
12241234 (1977)
2. C. Palmer, P.N. Stavrinou, Equations of motion in a noninteger-dimension space. J Phys A 37,
69877003 (2004)
3. M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions with Formulas Graphs and
Mathematical Tables. (Department of Commerce, U.S., 1972)
4. C.A. Balanis, Advanced Engineering Electromagnetics. (Wiley, New York, 1989)
Chapter 3
Potentials for Static and Time-Varying
Fields in Fractional Space
In this chapter, a fractional space generalization of potentials for static and time-
varying fields is discussed. The fractional space generalization of static and
time-varying potentials, provided in this chapter, can be used to study electrosta-
tic problems in fractal media.
In Sect. 3.1, electrostatic potential in D-dimensional fractional space is studied.
In Sect. 3.2, time-varying auxiliary potential is studied in fractional space.
In this section, an exact solution of the Laplaces equation for electrostatic potential
in D-dimensional fractional space is presented. As an application, the electrostatic
potential inside the rectangular box with surfaces held at constant potentials is
obtained in fractional pace. It is also shown that for integer value of dimension
D, the classical results are recovered. The obtained solution can be used to study
complex electrostatic problems in fractal media.
In Sect. 3.1.1, we investigate full analytical solution to the Laplaces equation in
D-dimensional fractional space, where the parameter D is used to describe the mea-
sure distribution of space. In Sect. 3.1.2, the electrostatic potential inside the rectan-
gular box with surfaces held at constant potentials in fractional space is obtained.
Finally, in Sect. 3.1.3, results are summarized.
2 1 1 2 2 1 2 3 1
D
2
= + + + + + (3.1.2)
x2 x x y2 y y z 2 z z
where, in addition,
2 + 2 = 2 (3.1.6)
Equation (3.1.5a) through (3.1.5c) are all of the same form; solution for any one of
them can be replicated for others by inspection.We choose to work first with f (x).
We write (3.1.5a) as
d2 d
x 2 +a + 2 x f = 0 (3.1.7)
dx dx
where, Jv1 (x) is referred to as Bessel function of the first kind of order v1 , Yv1 (x)
as the Bessel function of the second kind of order v1 . Finally the solution of (3.1.5a)
becomes
1
f (x) = xv1 C1 Jv1 (x) + C2 Yv1 (x) , v1 = 1 (3.1.10)
2
Similarly, the solutions to (3.1.5b) and (3.1.5c) are obtained as
2
g(y) = yv2 C3 Jv2 (y) + C4 Yv2 (y) , v2 = 1 (3.1.11)
2
h(z) = z v3 C5 Jv3 (j z) + C6 Yv3 (j z)
3
= z v3 C5 Jv3 (j 2 + 2 z) + C6 Yv3 (j 2 + 2 z) , v3 = 1
2
(3.1.12)
From (3.1.4) and (3.1.10) through (3.1.13), the solution of (3.1.3) have the form
(x, y, z) = xv1 yv2 z v3 C1 Jv1 (x) + C2 Yv1 (x) C3 Jv2 (y) + C4 Yv2 (y)
C5 Jv3 (j 2 + 2 z) + C6 Yv3 (j 2 + 2 z)
(3.1.13)
where, C1 through C6 are constant coefficients. Also and can be determined by
imposing boundary conditions on potential. This solution can be used to study the
electrostatic field in a non-integer dimensional fractal media.
The partial potential mn satisfying all boundary conditions except one, can be
written as:
And the potential can expanded in terms of these mn with initial arbitrary coefficients
(to be chosen to satisfy final boundary condition):
V (x, y) = Amn xv1 yv2 cv3 Jv1 (n x)Jv2 (m y)Jv3 (jmn c) (3.1.19)
n,m=1
Equation (3.1.18) provides the required solution in fractional space. Now, if we take
D = 3 i.e., 1 = 2 = 3 = 1 in (3.1.18) and use Bessel functions of fractional order
then the classical results given by Jackson [3], for the same problem in Euclidean
space, can be recovered.
3.1 Electrostatic Potential in Fractional Space 21
3.1.3 Summary
The procedure for analysis of radiation problems is to specify sources and get the
fields radiated by the sources. For analysis of radiation problems in fractional space
we have to introduce auxiliary potential function A (magnetic vector potential) and F
(electric vector potential). The fractional space generalization of the relation between
auxiliary potential functions and sources is given by inhomogeneous vector potential
wave equations. In this Section, a novel exact solution of the inhomogeneous vector
potential wave equations in D-dimensional fractional space is presented. It is also
shown that for integer values of dimension D, the classical results are recovered.
The solution of inhomogeneous vector potential wave equation in fractional space is
useful to study the radiation phenomenon in fractal media.
In Sect 3.2.1, we investigate full analytical solution of the inhomogeneous vector
potential wave equation in D-dimensional fractional space, where the parameter D
is used to describe the measure distribution of space, also the solution of inhomoge-
neous vector potential wave equation in integer-dimensional space is justified from
the results obtained. Finally, results are summarized drawn in Sect. 3.2.2.
The procedure for analysis of radiation problems is to specify sources and get the
fields radiated by the sources. For analysis of radiation problems in fractional space
we have to introduce auxiliary potential function A (magnetic vector potential) and F
(electric vector potential). The fractional space generalization of the relation between
auxiliary potential functions and sources is given by inhomogeneous vector potential
wave equations as below:
D
2
A + k 2 A = J (3.2.1)
D
2
F + k 2 F = M (3.2.2)
22 3 Potentials for Static and Time-Varying Fields
fractional space defined in Equation (3.1.2) and J, M are harmonic electric and
magnetic currents. In spherical coordinates fractional Laplacian operator becomes:
2
2 D1 1 D2
D
2
= 2+ + +
r r r r2 2 tan
2 (3.2.3)
1 D3
+ 2 +
r sin 2 tan
where, 2 < D 3. In this section we present a novel exact solution of the inho-
mogeneous vector potential wave equations in (3.2.1) and (3.2.2) in D-dimensional
fractional space. Once the solution to any one of equation (3.2.1) and (3.2.2) in frac-
tional space is known, the solution to the other can be written by duality principle.
To drive the solution to (3.2.1), we assume a source with current density Jz , which
in limit is an infinitesimal point source, is placed at origin of a x, y, z coordinate
system. Since the current density is directed along z-axis, only an Az component will
exist. Thus, using (3.2.1)
D
2
Az + k 2 Az = Jz (3.2.4)
At point removed from the source (Jz = 0), the wave equation reduces to
D
2
Az + k 2 Az = 0 (3.2.5)
where, 2 < D 3. The partial derivatives are replaced with ordinary derivative
because Az is a function of r only. Equation (3.2.6) is reducible to Bessels equation
under substitution Az (r) = xn as follows:
d2 d D
r2 2
+ r + (k 2 2
r n 2
) = 0, n = 1 (3.2.7)
dr dr 2
(1) (2)
where, Hn (kr) is referred to as Hankel function of the first kind of order n, Hn (kr)
as the Hankel function of the second kind of order n. Finally, the solution of (3.2.6)
becomes
3.2 Time-Varying Potentials in Fractional Space 23
D
Az = rn C1 Hn(1) (kr) + C2 Hn(2) (kr) , n = 1 (3.2.9)
2
Equations (3.2.10a, b) represent inward and outward going waves (assuming a time
dependency ejt ). For this problem source is located at origin and the waves are
going outward. So, we choose the solution of (3.2.6) as
Thus, at points removed from the source, the time-varying and static solutions
of (3.2.11) and (3.2.12) differ only by the argument of Hankel function from kr
to zero; or the time-varying solution can be found by multiplying static form with
(2)
rHn (kr).
In the presence of the source (Jz = 0) an (k = 0) the wave equation (2.2.4) is
reduced to
D
2
Az = Jz (3.2.13)
This equation is known as Poissons equation in fractional space. The well known
form of Poissons equation relating the scalar field to the electric charge density
is given by
D
2
= (3.2.14)
using analogy between (3.2.11) and (3.2.12), the time-varying solution of (3.2.4) is
given by
24 3 Potentials for Static and Time-Varying Fields
(2)
23D (3/2) Hn (kr)
Az = Jz dV (3.2.17)
4 (D/2) rD5/2
If the current were in the x- and y-directions (Jx ,Jy ), the wave equations for each
would reduce to have same form as (3.2.4) and will possess the same solutions as in
equation (3.2.17).
Finally, we have the solution of vector wave equation in (3.2.4) as:
(2)
23D (3/2) Hn (kr)
A= J D5/2 d V (3.2.18)
4 (D/2) r
If the source is removed from origin and placed at position represented by the primed
coordinates (x , y , z ), (3.2.18) can be written as:
23D (3/2) Hn(2) (kR)
A= J(x , y , z ) D5/2 d V (3.2.19)
4 (D/2) R
where, R is the distance between any point in the source to observation point, (x)
is the gamma function and Hn(2) (kR) denotes the Hankel function of second kind of
order n representing outward waves from source point.
Now, for validation of our provided solution in (3.2.19), we get vector potential A
from our solution by substituting D = 3 in (3.2.19). For D = 3, we have n = 1/2.
Using following Hankel function of fractional order [5]
2 jx
H 1 (x) =
2
e (3.2.20)
2 x
Equation (3.2.19) gets reduced to
ejkR
A = C J(x , y , z ) dV (3.2.21)
4 R
where, C is a constant term. Equation (3.2.21) is in exact agreement with the solution
provided in [6] for Euclidean space.
In a similar fashion, we can show the solution of (3.2.2) as
23D (3/2) Hn(2) (kR)
F= M(x , y , z ) D5/2 d V (3.2.22)
4 (D/2) R
For D = 3, using Hankel function of fractional order given by (3.2.20), finally
equation (3.2.22) gets reduced to
ejkR
F=C M(x , y , z ) dV (3.2.23)
4 R
Equation (3.2.23) is in exact agreement with the solution provided in [6] for Euclid-
ean space. The fractional space solutions of inhomogeneous vector potential wave
equation, given by (3.2.19) and (3.2.22), can be used to solve complex radiation
problems in fractional space.
3.2 Time-Varying Potentials in Fractional Space 25
3.2.2 Summary
References
The wave equation has very important role in many areas of physics. It has a funda-
mental meaning in classical as well as quantum field theory. With this view, one is
strongly motivated to discuss solutions of the wave equation in all possible situations.
The wave equation in fractional space can effectively describe the wave propagation
phenomenon in fractal media. In this chapter, exact solutions of different forms of
wave equation in D-dimensional fractional space are provided, which describe the
phenomenon of electromagnetic wave propagation in fractional space.
In Sect. 4.1, the general plane wave solutions in fractional space are provided for
lossless medium case. In Sect. 4.2, the general plane wave solutions in fractional
space are provided for lossy medium case. In Sect. 4.3, the general cylindrical wave
solutions in fractional space are provided by solving cylindrical wave equation in
D-dimensional fractional space. In Sect. 4.4, the general spherical wave solutions in
fractional space are provided by solving spherical wave equation in D-dimensional
fractional space.
For source-free and lossless media, the vector wave equations for the complex electric
and magnetic field intensities are given by the Helmholtzs equation as follows [1].
D
2
E + 2E = 0 (4.1.1a)
D
2
H + H =0
2
(4.1.1b)
2 1 1 2 2 1 2 3 1
D
2
= + + + + + (4.1.2)
x 2 x x y 2 y y z 2 z z
D
2
(ax E x + a y E y + az E z ) + 2 (ax E x + a y E y + az E z ) = 0 (4.1.4)
D
2
E x (x, y, z) + 2 E x (x, y, z) = 0 (4.1.5a)
D
2
E y (x, y, z) + E y (x, y, z) = 0
2
(4.1.5b)
D
2
E z (x, y, z) + 2 E z (x, y, z) = 0 (4.1.5c)
Eq. (4.1.5a) through (4.1.5c) are all of the same form; solution for any one of them
in fractional space can be replicated for others by inspection.We choose to work first
with E x as given by (4.1.5a). In expanded form (4.1.5a) can be written as
2 Ex 1 1 E x 2 Ex 2 1 E x 2 Ex
+ + + +
x 2 x x y 2 y y z 2
3 1 E x
+ + 2 Ex = 0 (4.1.6)
z z
Equation (4.1.6) is separable using separation of variables. We consider
where, in addition,
x2 + y2 + z2 = 2 (4.1.9)
= C1 Jn (x x) + C2 Yn (x x) (4.1.12)
From (4.1.7) and (4.1.13) through (4.1.15), the solution of (4.1.5a) have the form
30 4 Electromagnetic Wave Propagation in Fractional Space
E x (x, y, z) = x n 1 y n 2 z n 3 C1 Jn 1 (x x) + C2 Yn 1 (x x)
C3 Jn 2 ( y y) + C4 Yn 2 ( y y)
C5 Jn 3 (z z) + C6 Yn 3 (z z) (4.1.16)
and
E z (x, y, z) = x n 1 y n 2 z n 3 F1 Jn 1 (x x) + F2 Yn 1 (x x)
F3 Jn 2 ( y y) + F4 Yn 2 ( y y)
F5 Jn 3 (z z) + F6 Yn 3 (z z) (4.1.18)
where E x (x, y, z), E y (x, y, z) and E z (x, y, z) are given by (4.1.16) through (4.1.18).
Equation (4.1.19) provides a general plane wave solution in fractional space. This
solution can be used to study the phenomenon of electromagnetic wave propagation
in any non-integer dimensional space.
2 Ex 2 Ex 2 Ex D 3 Ex
+ + + + 2 Ex = 0 (4.1.26)
x2 y2 z 2 z z
Solving this equation by separation of variables leads to the following result
where, n = 2 D
2. Here if we set D = 3, and using (4.1.21a) and (b), we get
2
E x (x, y, z) = [G 1 cos(x x) + G 2 sin(x x)]
z
G 3 cos( y y) + G 4 sin( y y)
G 5 sin(z z) + G 6 cos(z z) (4.1.28)
Fig. 4.1 Usual wave propagation (D = 3). [This figure was originally published in [5] , reproduced
courtesy of The Electromagnetics Academy]
d2 Ex D 3 d Ex
+ + 2 Ex = 0 (4.1.29a)
dz 2 z dz
d 2 Hy D 3 d Hy
2
+ + 2 Hy = 0 (4.1.29b)
dz z dz
Solution of (4.1.29a) and (4.1.29b) takes the similar form as (4.1.27) and under
above mentioned boundary conditions the fields will have the following form:
Js0 Js
E1 = x z n Jn (z z), H1 = yz n 0 Jn (z z); z < 0 (4.1.30a)
2 20
n Js0 n Js0
E2 = x z Yn (z z), H2 = yz Yn (z z); z > 0 (4.1.30b)
2 20
4.1 General Plane Wave Solutions in Fractional Space: Lossless Medium Case 33
Fig. 4.2 Wave propagation in fractional space (D = 2.5). [This figure was originally published in
[5] , reproduced courtesy of The Electromagnetics Academy]
where, 0 is wave impedance in free space. Assuming a time dependency e jwt and
Js0 = 2 A/m, the solution for the usual wave for z > 0 with D = 3 is shown in
Fig. 4.1, which is comparable to well known plane wave solutions in 3-dimensional
space [1]. Similarly, for D = 2.5 we have fractal medium wave for z > 0 as shown
in Fig. 4.2, where amplitude variations are described in terms of Bessel functions.
4.1.3 Summary
General plane wave solution in source-free and lossless media in fractional space is
presented by solving vector wave equation in D-dimensional fractional space. When
the wave propagates in fractional space, the amplitude variations are described by
Bessel functions. The obtained general plane wave solution is a generalization of
integer-dimensional solution to a non-integer dimensional space. For all investigated
cases when D is an integer-dimension, the classical results are recovered.
In this section, an extension of previous work to the case of plane wave propagation
in lossy medium in fractional space is presented. The generalized analytical solu-
tion investigated in this section have potential applications in electromagnetic wave
propagation problems in lossy media present in fractional space.
34 4 Electromagnetic Wave Propagation in Fractional Space
For source-free and lossy media, the vector wave equations for the complex electric
and magnetic field intensities are given by the Helmholtzs equation as follows [1].
D
2
E 2E = 0 (4.2.1)
D H 2H = 0
2
(4.2.2)
where
2 = j( + j) (4.2.3)
= + j (4.2.4)
2
=
1+ 1 (4.2.5)
2
2
=
1+ +1 (4.2.6)
2
2 1 1 2 2 1 2 3 1
D
2
= + + + + + (4.2.7)
x2 x x y2 y y z 2 z z
4.2 General Plane Wave Solutions in Fractional Space: Lossy Medium Case 35
D
2
(ax E x + a y E y + az E z ) 2 (ax E x + a y E y + az E z ) = 0 (4.2.9)
D
2
E x (x, y, z) 2 E x (x, y, z) = 0 (4.2.10)
D
2
E y (x, y, z) 2 E y (x, y, z) = 0 (4.2.11)
D
2
E z (x, y, z) 2 E z (x, y, z) = 0 (4.2.12)
Equation (4.2.10) through (4.2.12) are all of the same form; solution for any one of
them in fractional space can be replicated for others by inspection.We choose to work
first with E x as given by (4.2.10). In expanded form (4.2.10) can be written as
2 Ex 1 1 E x 2 Ex 2 1 E x
+ + +
x2 x x y2 y y
(4.2.13)
Ex
2 3 1 E x
+ + 2 Ex = 0
z 2 z z
Equation (4.2.13) is separable using separation of variables. We consider
where, in addition,
36 4 Electromagnetic Wave Propagation in Fractional Space
x2 + y2 + z2 = 2 (4.2.18)
= C1 Jn ( jx x) + C2 Yn ( jx x) (4.2.21)
From (4.2.14) and (4.2.22) through (4.2.24), the solution of (4.2.10) for E x (x, y, z)
in D-dimensional fractional space have the form
E x (x, y, z) = x n 1 y n 2 z n 3 C1 Jn 1 ( jx x) + C2 Yn 1 ( jx x)
C3 Jn 2 ( j y y) + C4 Yn 2 ( j y y)
C5 Jn 3 ( jz z) + C6 Yn 3 ( jz z) (4.2.25)
and
E z (x, y, z) = x n 1 y n 2 z n 3 F1 Jn 1 ( jx x) + F2 Yn 1 ( jx x)
F3 Jn 2 ( j y y) + F4 Yn 2 ( j y y)
F5 Jn 3 ( jz z) + F6 Yn 3 ( jz z) (4.2.27)
where, E x (x, y, z), E y (x, y, z) and E z (x, y, z) are given by (4.2.25) through (4.2.27).
Equation (4.2.28) provides a general plane wave solution for lossy media in frac-
tional space. This solution can be used to study the phenomenon of electromagnetic
wave propagation in any non-integer dimensional space.
2j
f (x) = C1 sinh(x x) + C2 cosh(x x) (4.2.32)
x
2j
g(y) = C3 sinh( y y) + C4 cosh( y y) (4.2.33)
y
2j
h(z) = C5 sinh(z z) + C6 cosh(z z) (4.2.34)
z
8
E x (x, y, z) = C1 sinh(x x) + C2 cosh(x x)
j 3 x y z
C3 sinh( y y) + C4 cosh( y y)
C5 sinh(z z) + C6 cosh(z z) (4.2.35)
8
E y (x, y, z) = D1 sinh(x x) + D2 cosh(x x)
j x y z
3
D3 sinh( y y) + D4 cosh( y y)
D5 sinh(z z) + D6 cosh(z z) (4.2.36)
8
E z (x, y, z) = F1 sinh(x x) + F2 cosh(x x)
j 3 x y z
F3 sinh( y y) + F4 cosh( y y)
F5 sinh(z z) + F6 cosh(z z) (4.2.37)
Now, solution obtained in (4.2.35) through (4.2.37) is comparable to the well known
solution of Helmholtzs equation in lossy media present in integer dimensional space
obtained by Balanis [1].
As a special case, if we choose a single parameter for non-integer dimension D
where 2 < D 3, i.e., we take 1 = 2 = 1 so D = 3 + 2. In this case from
Eq. (4.2.13) we obtain
2 Ex 2 Ex 2 Ex D 3 Ex
+ + + 2 Ex = 0 (4.2.38)
x2 y2 z 2 z z
4.2 General Plane Wave Solutions in Fractional Space: Lossy Medium Case 39
or
E x (x, y, z) = z n G 1 cosh(x x) + G 2 sinh(x x)
G 3 cosh( y y) + G 4 sinh( y y)
G 5 Hn(1) ( jz z) + G 6 Hn(2) ( jz z) (4.2.40)
2j
E x (x, y, z) = G 1 cosh(x x) + G 2 sinh(x x)
z
G 3 cosh( y y) + G 4 sinh( y y)
G 5 sinh(z z) + G 6 cosh(z z) (4.2.41)
E y (x, y, z) = x n 1 y n 2 z n 3 D1 Jn 1 (x x) + D2 Yn 1 (x x)
D3 Jn 2 ( y y) + D4 Yn 2 ( y y)
D5 Jn 3 (z z) + D6 Yn 3 (z z) (4.2.43)
40 4 Electromagnetic Wave Propagation in Fractional Space
E z (x, y, z) = x n 1 y n 2 z n 3 F1 Jn 1 (x x) + F2 Yn 1 (x x)
F3 Jn 2 ( y y) + F4 Yn 2 ( y y)
F5 Jn 3 (z z) + F6 Yn 3 (z z) (4.2.44)
Fractional space solution for lossless medium obtained in (4.2.42) through (4.2.44)
is in exact agreement with that obtained in previous section.
d2 Ex D 3 d Ex
2
+ 2 Ex = 0 (4.2.45)
dz z dz
d 2 Hy D 3 d Hy
2
+ 2 Hy = 0 (4.2.46)
dz z dz
Solution of (4.2.45) and (4.2.46) takes the similar form as (4.2.40) and under
above mentioned boundary conditions the fields will have the following form:
Js0 n (1)
E1 = x z Hn ( j1 z); z<0 (4.2.47)
2
Js
H1 = y 0 z n Hn(1) ( j1 z); z<0 (4.2.48)
21
Js
E2 = x 0 z n Hn(2) ( j2 z); z>0 (4.2.49)
2
Js0 n (2)
H2 = y z Hn ( j2 z); z>0 (4.2.50)
22
where, 12 = j1 (1 + j1 ), 22 = j2 (2 + j2 ), 1 = j j2
1 , 2 = 2 .
1
Assuming a time dependency e jwt and Js0 = 2 A/m, the solution for the usual
wave for z > 0 with D = 3 is shown in Fig. 4.4, which is comparable to well known
plane wave solutions for 3-dimensional space in [1]. Similarly, for D = 2.5 we have
fractal wave propagating in fractional space for z > 0 as shown in Fig. 4.5, where
amplitude variations are described in terms of Hankel functions.
42 4 Electromagnetic Wave Propagation in Fractional Space
Fig. 4.5 Wave propagation in lossy medium present in fractional space (D = 2.5)
4.2.4 Summary
The problems that exhibit cylindrical geometries are needed to be solved using cylin-
drical coordinate system. As for the case of rectangular geometries, the electric and
magnetic fields of cylindrical geometry boundary-value problem must satisfy cor-
responding cylindrical wave equation [1]. Let us assume that the space in which
fields must be solved is fractional dimensional and source-free. For source-free and
lossless media, the vector wave equations for the complex electric and magnetic field
intensities are given by the Helmholtzs equation as follows [1].
D
2
E + 2E = 0 (4.3.1)
D
2
H + 2H = 0 (4.3.2)
2 1 1 2 2 1 2 3 1
D
2
= + + + + + (4.3.3)
x 2 x x y 2 y y z 2 z z
where x, y and z are rectangular coordinates. Equation (4.3.3) uses three parameters
(0 < 1 1, 0 < 2 1 and 0 < 3 1) to describe the measure distribution
of space where each one is acting independently on a single coordinate and the total
dimension of the system is D = 1 + 2 + 3 . To find cylindrical wave solutions
of wave equation in D-dimensional fractional space, it is likely that a cylindrical
coordinate system (, , z) will be used. In cylindrical coordinate system (4.3.3)
becomes
2 1
D
2
= + (1 + 2 1)
2
2
1
+ 2 {(1 1) tan (2 1) cot }
2
2 3 1
+ + (4.3.4)
z 2 z z
Once the solution to any one of Eqs. (4.3.1) and (4.3.2) in fractional space is known,
the solution to the other can be written by an interchange of E with H or H with E
due to duality [1]. We will examine the solution for E.
In cylindrical coordinates, a general solution for E can be written as
E(, , z) = a E (, , z) + a E (, , z) + az E z (, , z) (4.3.5)
D
2
(a E + a E + az E z ) + 2 (a E + a E + az E z ) = 0 (4.3.6)
Since,
D
2
(a E ) = a D
2
E (4.3.7)
D
2
(a E ) = a D
2
E (4.3.8)
D
2
(az E z ) = az D
2
Ez (4.3.9)
So, Eq. (4.3.6) cannot be reduced to simple scalar wave equations, but it can be
reduced to coupled scalar partial differential equations. However for simplicity, the
wave mode solution can be formed in cylindrical coordinates that must satisfy the
following scalar wave equation:
D
2
(, , z) + 2 (, , z) = 0 (4.3.10)
2 1
+ (1 + 2 1)
2
2
1
+ 2 {(1 1) tan ( 2 1) cot }
2
2 3 1
+ + + 2 = 0 (4.3.11)
z 2 z z
Equation (4.3.11) is separable using method of separation of variables. We con-
sider
(, , z) = f ()g()h(z) (4.3.12)
2 + z2 = 2 (4.3.16)
or
1 +2
f 2 () = 1 2 D1 Hv(1) ( ) + D2 Hv(2) ( ) (4.3.20)
where, v = 21 (2 1 2 )2 + 4m 2 . In (4.3.19) Jv ( ) is referred to as Bessel
function of the first kind of order v and Yv (r ) as the Bessel function of the second
kind of order v. They are used to represent standing waves. In (4.3.20) Hv(1) (r )
(2)
is referred to as Hankel function of the first kind of order v and Hv ( ) as the
Hankel function of the second kind of order v, and are used to represent traveling
waves.
Now, we find the solution of equation (4.3.14) for g(). Equation (4.3.14) can be
reduced to following Gaussian hypergeometric equation after proper mathematical
steps under substitution = sin2 () [3]:
d 2 g() dg()
(1 ) 2
+ {(A + B + 1) C} + ABg() = 0 (4.3.21)
d d
where,
1
A+ B+1= (2 2 + 1 ) (4.3.22)
2
m2
AB = (4.3.23)
4
46 4 Electromagnetic Wave Propagation in Fractional Space
1
C= (2 2 ) (4.3.24)
2
where,
(A)k (B)k k
F(A, B, C; ) = 1 + (4.3.26)
(C)k k!
k=1
with,
(z) = C5 Jn (z z) + C6 Yn (z z) (4.3.30)
[z {C5 Jn (z z) + C6 Yn (z z)}]
n
(4.3.32)
4.3 Cylindrical Wave Propagation in Fractional Space 47
f 1 () = C1 Jm ( ) + C2 Ym ( ) (4.3.33)
and
which is comparable to the cylindrical wave solutions of the wave equation in integer
dimensional space obtained by Balanis [1].
As an example, the fields inside a circular waveguide filled with fractal media
of dimension D can be obtained by assuming a D-dimensional fractional space
inside the circular waveguide. Within such circular waveguide of radius a (see
Fig. 4.6), standing waves are created in the radial() direction, periodic waves in
the -direction, and traveling waves in the z-direction.
For the fields to be finite at = 0 where Y v( ) possesses a singularity, (4.3.32)
reduces to
1 +2
1 (, , z) =[ 1 2 {C1 Jv ( )}] [{C3 F(A, B, C; )
+ C4 1C
F(A C + 1, B C + 1, 2 C; )}]
[z n {C5 Hn(2) (z z) + C6 Hn(1) (z z)}] (4.3.43)
4.3 Cylindrical Wave Propagation in Fractional Space 49
Fig. 4.7 Cylindrical wave propagation in Euclidean space (D = 3). [This figure was originally pub-
lished in [6] , reproduced courtesy of The Electromagnetics Academy]
Fig. 4.8 Cylindrical wave propagation in fractional space (D = 2.5). [ This figure was originally
published in [6] , reproduced courtesy of The Electromagnetics Academy]
To represent the fields in the region outside the cylinder, where three dimensional
space is assumed because there is no fractal media outside the cylinder, a typical
solution for (, , z) would take the form
50 4 Electromagnetic Wave Propagation in Fractional Space
Fig. 4.9 Cylindrical wave propagation in fractional space (D = 2.1). [ This figure was originally
published in [6] , reproduced courtesy of The Electromagnetics Academy]
2 (, , z) = C2 Hm(2) ( ) [C3 cos(m) + C4 sin(m)]
[C5 sin(z z) + C6 cos(z z)] (4.3.44)
In the region outside the cylinder, outward traveling waves are formed, in contract to
standing waves inside the cylinder. In this way, the general cylindrical wave solution
in fractional space can be used to study the wave propagation in the cylindrical
geometries containing fractal media.
Now, as another example we assume that a cylindrical wave exists in a fractional
space due to some infinite line source. Since the source do not vary with z, the fields
will not vary with z but will propagate away from the source in -direction. Also for
simplicity, we choose to visualize only the radial amplitude variations of scalar field
in fractional space which is given by (4.3.32) as:
1 +2
() = A 1 2 Hv(2) ( ) (4.3.45)
From (4.3.47),
4.3 Cylindrical Wave Propagation in Fractional Space 51
for D = 3,() 1
for D = 2.5, () 1
0.25
for D = 2.1, () 1
0.05
Assuming a time dependency e jwt , the radial amplitude variations of scalar field
are shown for different values of dimension D in Figs 4.7, 4.8, 4.9. It is seen that
the amplitude of cylindrical wave propagating in higher dimensional space decays
rapidly.
4.3.3 Summary
The problems that exhibit spherical geometries are needed to be solved using spher-
ical coordinates. As for the case of rectangular geometries, the electric and magnetic
fields of spherical geometry boundary-value problem must satisfy corresponding
spherical wave equation [1]. Let us assume that the space in which fields must be
solved is fractional dimensional, source-free and lossless. For source-free and loss-
less media, the vector wave equations for the complex electric and magnetic field
intensities are given by the Helmholtzs equation as follows:
52 4 Electromagnetic Wave Propagation in Fractional Space
D
2
E + 2E = 0 (4.4.1)
D
2
H + 2H = 0 (4.4.2)
where, 2 < D 3. Once the solution to any one of Eq. (4.4.1) and (4.4.2) in fractional
space is known, the solution to the other can be written by an interchange of E with
H or H with E due to duality. We will examine the solution for E.
In spherical coordinates, a general solution for E can be written as
D
2
(ar Er + a E + a E ) + 2 (ar Er + a E + a E ) = 0 (4.4.5)
Since,
D
2
(ar Er ) = ar D
2
Er (4.4.6)
D
2
(a E ) = a D
2
E (4.4.7)
D
2
(a E ) = a D
2
E (4.4.8)
So, Eq. (4.4.5) cannot be reduced to three simple scalar wave equations, but it can be
reduced to three coupled scalar partial differential equations. However for simplicity,
the wave mode solution can be formed in spherical coordinates that must satisfy the
following scalar wave equation:
D
2
(r, , ) + 2 (r, , ) = 0 (4.4.9)
where, (r, , ) is a scalar function that can represent a field or vector potential
component. In expanded form (4.4.9) can be written as
2 D 1 1 2 D 2
+ + +
r 2 r r r 2 2 tan
2
1 D 3
+ 2 + + 2 = 0
r sin 2 tan (4.4.10)
4.4 Spherical Wave Propagation in Fractional Space 53
m 2
d2 D2 d
+ + n(n + 1) g=0 (4.4.13)
d 2 tan d sin
2
d D3 d
+ + m 2
h=0 (4.4.14)
d 2 tan d
or
D
f 2 = r 1 2 D1 Hv(1) (r ) + D2 Hv(2) (r ) (4.4.18)
where, v = 21 (2 D)2 + 4n(n + 1). In (4.4.17) Jv (r ) is referred to as Bessel
function of the first kind of order v and Yv (r ) as the Bessel function of the second
(1)
kind of order v. They are used to represent radial standing waves. In (4.4.18) Hv (r )
(2)
is referred to as Hankel function of the first kind of order v and Hv (r ) as the Hankel
function of the second kind of order v, and are used to represent radial traveling waves.
54 4 Electromagnetic Wave Propagation in Fractional Space
Now, we find the solution of Eq. (4.4.13) for g( ). Equation (4.4.13) can be
reduced to following Gaussian hypergeometric equation after proper mathematical
steps under substitution g = w sinm and z 1 = 1+cos
2 :
d 2w dw
z 1 (1 z 1 ) 2
+ {(1 + 1 + 1)z 1 1 } + 1 1 w = 0 (4.4.19)
dz 1 dz 1
where,
1 1 = (n m)(m + n 1) (4.4.21)
1 = (D 2)(n + 1) (4.4.22)
w = C3 F(1 , 1 , 1 ; z 1 )
11 (4.4.23)
+ C4 z 1 F(1 1 + 1, 1 1 + 1, 2 1 ; z 1 )
where,
(1 )k (1 )k z 1k
F(1 , 1 , 1 ; z 1 ) = 1 + (4.4.24)
(1 )k k!
k=1
with,
(1 )k = 1 (1 + 1)...(1 + k + 1) (4.4.25)
g = [C3 F(1 , 1 , 1 ; z 1 )
11
+ C4 z 1 F(1 1 + 1, 1 1 + 1, 2 1 ; z 1 )]sin m (4.4.26)
Now, we find the solution of equation (4.4.14) for h(). Equation (4.4.14) can be
written as:
2
d d
+ pcot +q h =0 (4.4.27)
d 2 d
4.4 Spherical Wave Propagation in Fractional Space 55
d2h dh
z 2 (1 z 2 ) 2
+ {(2 + 2 + 1)z 2 2 } + 2 2 h = 0 (4.4.28)
dz 2 dz 2
where,
2 + 2 = D 5 (4.4.29)
2 2 = m 2 (4.4.30)
1
2 = (4 D) (4.4.31)
2
h = C5 F(2 , 2 , 2 ; z 2 )
12
+ C6 z 2 F(2 2 + 1, 2 2 + 1, 2 2 ; z 2 ) (4.4.32)
a special case, for three-dimensional space, this problem reduces to classical wave
propagation concept; i.e., as a special case, if we set D = 3 in Eq. (4.4.17) then
v = n + 21 and it gives
1
f 1 = C1 Jn+ 1 (r ) + C2 Yn+ 1 (r ) (4.4.34)
r 2 2
here, regular Bessel functions in (4.4.34) are related to spherical Bessel function
by [4]
jn (r ) = J 1 (r ) (4.4.35)
2r n+ 2
yn (r ) = Y 1 (r ) (4.4.36)
2r n+ 2
here, regular Hankel functions in (4.4.38) are related to spherical Hankel function
by [4]
(1) (1)
h n (r ) = H (r ) (4.4.39)
2r n+ 21
h (2)
n (r ) = H (2) (r ) (4.4.40)
2r n+ 21
m d m Pn ( )
Pnm ( ) = (1 2 ) 2 (4.4.43)
d m
m d m Q n ( )
n ( ) = (1 )
Qm 2 2
(4.4.44)
d m
g = C3 Pnm (cos ) + C4 Q m
n (cos ) (4.4.45)
From (4.4.11), (4.4.37), (4.4.45) and (4.4.46), a typical solution in three dimen-
sional space ( a special case of fractional space) for (r, , ) to represent the fields
within a spherical geometry will take the form
2
(r, , ) = [C1 jn (r ) + C2 yn (r )]
[C3 Pnm (cos ) + C4 Q m
n (cos )]
[C5 cos(m) + C6 sin(m)] (4.4.47)
From (4.4.49),
for D = 3,(r ) r1
for D = 2.5, (r ) r 0.75
1
Fig. 4.10 Spherical wave propagation in Euclidean space (D = 3). [ This figure was originally pub-
lished in [7] , reproduced courtesy of The BRILL]
Fig. 4.11 Spherical wave propagation in fractional space (D = 2.5). [ This figure was originally
published in [7] , reproduced courtesy of The BRILL]
Assuming a time dependency e jwt , the radial amplitude variations of scalar field
are shown for different values of dimension D in Figs. 4.10, 4.11, 4.12. In Fig. 4.10
a spherical wave propagating in Euclidean space of dimension D = 3 is shown,
where the amplitude decays by r1 . In Figs. 4.3, 4.4, 4.5 a spherical wave propagating
4.4 Spherical Wave Propagation in Fractional Space 59
Fig. 4.12 Spherical wave propagation in fractional space (D = 2.1). [ This figure was originally
published in [7] , reproduced courtesy of The BRILL]
4.4.3 Summary
References
4. M. Abramowitz, I.A. Stegun, Handbook of mathematical functions with formulas, Graphs, and
Mathematical Tables.U.S. Department of Commerce (1972)
5. M. Zubair, M.J. Mughal, Q.A. Naqvi, The wave equation and general plane wave solutions in
fractional space. Prog Electromn Res Lett 19, 137146 (2010)
6. M. Zubair, M.J. Mughal, Q.A. Naqvi, An exact solution of cylindrical wave equation for electro-
magnetic field in fractional dimensional space. Prog Electromn Res Lett 114, 443455 (2011)
7. M. Zubair, M.J. Mughal, Q.A. Naqvi, An exact solution of spherical wave in D-dimensional
fractional space. J Electromn Waves Appl 25, 14811491 (2011)
Chapter 5
Electromagnetic Radiations from Sources
in Fractional Space
In analysis of radiation problems, the procedure is to specify sources and get the
fields radiated by the sources. A common practice in analysis of radiation problems
in fractional space is to introduce auxiliary potential function AD (magnetic vector
potential) and FD (electric vector potential) [1]. An overview of steps involved in
solving typical radiation problems in fractional space are shown in Fig. 5.1 .
D
2
AD + k 2 AD = J (5.1.1)
Fig. 5.1 Block diagram for computing radiated fields in fractional space
Using the solution of Poissons equation in fractional space [2] and considering
analogy of (5.1.1) with vector wave equation in fractional space solved in Sect. 4.1,
we solve (5.1.1) for AD and get
(2)
23D (3/2) Hn (k R)
AD = J(x , y , z ) dV (5.1.2)
4 (D/2) R D5/2
where, C is a constant term. Equation 5.1.4 is in exact agreement with the solution
provided in [1] for Euclidean space.
D
2
FD + k 2 FD = M (5.1.5)
5.1 Solution Procedure for Radiation Problems in Fractional Space 63
(3.2.3), and k 2 = 2 .
Using the analogy to the solution provided for AD in (5.1.2), we get
23D (3/2) Hn(2) (k R)
FD = M(x , y , z ) dV (5.1.6)
4 (D/2) R D5/2
In the previous Sections, we have developed equations that can be used for electric
and magnetic fields generated by and electric current source J and a magnetic current
source M in fractional space. The procedure for radiation analysis in fractional space
requires that the potential functions AD and FD are generated, respectively, by J and
M. In turn the corresponding electric and magnetic fields are then determined in far
zone. The fields radiated in far zone (k R >> 1) due to AD are EAD and HAD and
are given by [1]
j
HAD = ar AD (5.1.9)
where is wave impedance. The fields radiated due to FD are EFD and HFD and are
given in the same form as (5.1.8) and (5.1.9):
j
EFD = ar FD (5.1.11)
64 5 Electromagnetic Radiations from Sources in Fractional Space
And finally, the total fields are given by superposition of the individual fields due to
AD and FD as:
I(z ) = az I0 (5.2.1)
where I0 is constant.
Using the procedure defined in Sect. 5.1, first we find AD and FD . As no magnetic
source is assumed so M = 0 and hence, FD = 0. We find AD using (5.1.2) as:
(2)
23D (3/2) Hn (k R)
AD = az I(x , y , z ) dl (5.2.2)
4 (D/2) L R D5/2
The next step of procedure will be to find EAD and EFD . To do this, it will be more
convenient to transform (5.2.3) in spherical coordinates. Here, A x D = A y D = 0 so
in spherical coordinates (5.2.3) becomes
(2)
I0 l 23D (3/2) Hn (kr )
Ar D = A z D cos( ) = cos (5.2.4)
4 (D/2) r D5/2
5.2 Elementary Hertzian Dipole in Fractional Space 65
(2)
I0 l 23D (3/2) Hn (kr )
A D = A z D sin( ) = sin (5.2.5)
4 (D/2) r D5/2
A D = 0 (5.2.6)
Er D = E D = Hr D = H D = 0 (5.2.8)
(2)
jk I0 l 23D (3/2) Hn (kr )
H D = sin (5.2.9)
4 (D/2) r D5/2
66 5 Electromagnetic Radiations from Sources in Fractional Space
The ratio of E D to H D is
E D
Zw = = (5.2.10)
H D
jkI0 l e jkr
E = sin (5.2.11)
4 r
Er D = E D = Hr D = H D = 0 (5.2.12)
jk I0 l e jkr
H = sin (5.2.13)
4 r
The field expressions (5.2.115.2.13) are in exact agreement with those obtained
in [1] for far-field region. This validates our fractional solutions in (5.2.75.2.9) in
far-field region. In (5.2.75.2.9), E D and H D are perpendicular to each other and
transverse to the direction of propagation and r is separable for and . The shape
of pattern is not function of r and fields form a Transverse Electromagnetic (TEM)
wave with wave impedance equal to the intrinsic impedance. This relation is applica-
ble in the far-zone of all antennas of finite dimensions in fractional space. Comparing
our fractional space results in (5.2.75.2.9) with the results for Euclidean space in
(5.2.115.2.13), we state that in radiated fields of fractional space, amplitude varia-
tions are described by Hankel functions of order as a function of space dimension.
5.2.2 Directivity
1
WD = Re(E D H D ) = ar | E |2 (5.2.14)
2 2
For Hertzian dipole this gives
k I0 l 23D (3/2) 2 sin2
W D = ar (5.2.15)
2 4 (D/2) r 2D4
and finally, the maximum directivity D0D of Hertzian dipole in fractional space is
given by
U0D U0D 3
D0D = 4 = 4 = (5.2.19)
PD U0D 8
3
2
5.3 Summary
References
1. C.A. Balanis, Antenna Theory: Analysis and Design (Wiley, New York, 1982)
2. S. Muslih, D. Baleanu, Fractional multipoles in fractional space. Nonlinear Anal.: Real World
Appl. 8, 198203 (2007)
3. M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions with Formulas Graphs and
Mathematical Tables (U.S. Department of Commerce, U.S., 1972)
4. C.A. Balanis, Advanced Engineering Electromagnetics (Wiley, New York, 1989)
Chapter 6
Conclusions
D-dimensional fractional space, was also presented. The obtained exact solution
of cylindrical and spherical wave equation is a generalization of classical integer-
dimensional solutions to a non-integer dimensional space. All these investigated
solutions of vector wave equation provide a basis for the application of the concept
of fractional space to the wave propagation phenomenon in fractal media.
We have also discussed a novel solution procedure for radiation problems in frac-
tional dimensional space. The proposed solution procedure can be used to study the
radiation phenomenon in any non-integer dimensional fractal media. As an applica-
tion, the fields radiated by a Hertzian dipole in fractional space were worked out.
It was investigated that the amplitude variations of the radiated fields in fractional
space are described by Hankel functions. For all investigated cases when integer
dimension was considered, the classical results were recovered.