This action might not be possible to undo. Are you sure you want to continue?

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

/

**Basic Electromagnetic Theory
**

Electromagnetic analysis has been an indispensable part of many engineering and scientific studies since J. C. Maxwell completed the electromagnetic theory in 1873 [1]. This is due primarily to the predictive power of Maxwell's equations as proven over the years and the pervasiveness of electromagnetic phenomena in modern technologies. Examples of these technologies are radar, remote sensing, geoelectromagnetics, bioelectromagnetics, antennas, wireless communication, optics, high-frequency/highspeed circuits, and so on. Moreover, electromagnetic theory is valid from the static to optical regimes and from subatomic to intergalactic length scales. The problem of electromagnetic analysis is actually a problem of solving a set of Maxwell's equations subject to given boundary conditions. In this chapter we review briefly some basic concepts and equations of electromagnetic theory that are used frequently in this book. Our emphasis is on the presentation of various differential equations and boundary conditions that define boundary-value problems to be solved by finite element analysis. The solution of Maxwell's equations in free space is also given in the form of an integral expression that relates the field to its source, followed by the description of Huygens's principle for calculating the exterior fields from the field on a closed surface. For a complete presentation of electromagnetic theory, the reader is referred to textbooks such as [2-7]. This chapter may be skipped if the reader is familiar with the theory. Since the entire treatment of electromagnetic theory depends on vector analysis, we first review briefly the basic concepts and theorems of vector calculus.

1

-^0 A f If (itdsxf (1. a quantity whose magnitude and direction vary as functions of space. The gradient of the scalar function / is defined as V/ : lim — Is fda\ (1. Assume that f is a vector function.— At.2) is known as the divergence theorem or Gauss's theorem.6) Ai.5) (1.V/=£.f = lim -*-\(H)f-ds\ 1 A. and gradient. h • V/ = dn (1.5) >c if the vector f and its first derivative are continuous on surface S as well as along contour C that bounds S. we can show that £* fdv fa (1.^O At' At' s [Jfa fa'*] J •ds (1.| $ d s x f | lim . Now let / be a scalar function of space..7) . From the definition of curl. From the definition of divergence.7) (1. curl.2 BASIC ELECTROMAGNETIC THEORY 1.2) if the vector f and its first derivative are continuous in volume V as well as on its surface S.4) v . Equation (1.4) h • (V x f) lim — ff f -dl — * • dl (1. it can be shown that (I (V x ff) -•dds== I f --d\ (Vx ) s f dl Js (1.1) (1. Equation (1. As—0 As A*-O As [Jc \ Jc J where c is the contour bounding surface As and n denotes the unit vector normal to As. The divergence of the vector function f is defined by the limit V .1 BRIEF REVIEW OF VECTOR ANALYSIS Perhaps the most useful concepts in vector analysis are those of divergence. The directions of h and I are related by the right-hand rule.3) whose magnitude in the direction of n is given by h 1 1 [ff 1 h • (V x f) = lim — (1.1) where s is the surface enclosing volume At' and ds is normal to s and points outward.5) is known as Stokes's theorem. In this section we present definitions and related theorems for these quantities. The curl of the vector function f is defined as X V x ff := lim —^ .3) (1.->0 At' whose magnitude in the direction of n is given by df .

2). If we substitute f = aVb into (1. H (a x b) • n dS.11) where V is known as the Laplacian. we obtain the second vector Green's theorem £ (b-(VxVxa)-a-(VxVxbpy (axVxb-bxVxa)-y5.15) .2).10) where / and f are any functions that are continuous with a continuous first derivative. From the divergence theorem.14) Switching the positions of a and b and subtracting the resulting equation from (1.b^ ) dS.1. Exchanging the positions of a and b and subtracting the resulting equation from (1.bV2a) dV=H (a^.13) If we substitute f = a x V x b into (1.1 BRIEF REVIEW OF VECTOR ANALYSIS 3 From the definition of gradient. / Is v jfs\\ dn (1. we can show that fff wid IILVfdv=sidH' fds "§ (1. we obtain the first scalar Green's theorem [[[ {aV2b + Va-Vb)dV= H a~ dS (1.. one can derive some integral theorems that are used frequently in the formulation of the finite element method. s (1.14). we obtain the first vector Green's theorem fff [(V x a) • (V x b) . Other useful vector identities are given in Appendix A.12).0 V-(Vxf) = 0 (1. (1.12) where V 2 / = V • V / .8) if / and its first derivative are continuous in volume V as well as on its surface S. dn on on.8) is known as the gradient theorem. Equation (1. we obtain the second scalar Green's theorem Hi db 7 da (aV2b . Both identities can be proven with the aid of the divergence and curl theorems and are easily verified in Cartesian coordinates.9) (1.a • (V x V x b)] dV -= (B{8LXx V xb)-hdS. Two very useful identities involving the three basic vector differential operators are V x (V/) . Another useful identity is Vx V x f = V V f -V2f 2 (1.

• dl = -— [[/ BBds • ds Jc dt JJS '£> Jc dt JJs (Faraday's law) (Faraday's law) (1. 1. Maxwell's equations in integral form are given by / E dl = -— / 0 E .6).18) (1. Derive Stokes's theorem in (1.4) from the original definition given in (1.4).17) (1.4 BASIC ELECTROMAGNETIC THEORY These are the standard Green's theorems.1). These generalized theorems are given in Appendix A and are actually more useful in the finite element formulation.1 Derive Gauss's theorem in (1.(dss= : / / / pdv Jv (1.8) from (1. Exercise 1.2 Derive the alternative definition of the curl in (1.5) from (1. and here we present both to illustrate applications of some of the integral theorems discussed in the preceding section.1 The General Integral Form For general time-vary ing fields.3). .19) Pdv (Gauss's law) (Gauss's law—magnetic) (Gauss's law—magnetic) B • ds = 0 ds Js where E = electric field intensity (volts/meter) D = electric flux density (coulombs/meter2) H = magnetic field intensity (amperes/meter) B = magnetic flux density (webers/meter2) J = electric current density (amperes/meter2) p = electric charge density (coulombs/meter3).2 MAXWELL'S EQUATIONS Maxwell's equations are a set of fundamental equations that govern all macroscopic electromagnetic phenomena.2) from the definition of the divergence in (1. Derive the gradient theorem in (1.3 Derive the alternative definition of the gradient in (1. Exercise 1. Exercise 1. 1.i = §B.7) from the original definition given in (1. They can be generalized slightly to contain another function or parameter.6). The equations can be written in both differential and integral forms.16) / H • dl = -f / / D • ds + [[ J • ds (Maxwell-Ampere law) (Maxwell-Ampere law) r Jc dt JJs JJ5 S Js OD.2.

4 dt at do equation of continuity).23).17) and (1.28) . Another fundamental equation.2 MAXWELLS EQUATIONS 5 In (1. (1. Consider a point in space where all thefieldquantities and their derivatives are continuous.27) (1.20) by using Gauss's and Stokes's theorems.16)-( 1.3 Electro. (1.2.17). (1.16) and (1.24) (1. Equations (1. which can be derived from (1. can be derived from the independent equations and thus are called auxiliary or dependent equations.18). J .16)—(1.2. or the first two equations. In this case. is the mathematical form of the law of the conservation of charge. whereas in (1. surface.20) yields v—fdt dt at V D =p V-B .26) (1.22).25) can be written as VxE =0 VxE =0 VxH =J VxH =J V-J = 0 (1. The other two equations. 1. Either the first three equations.1.20) are valid in all circumstances regardless of the medium and the shape of the integration volume.16)—(1.and Magnetostatic Fields When the field quantities do not vary with time.19). with (1.23).21).25) can be chosen as such independent equations.25) V x H =8T> J f+ V .21)-(1.24) and (1. (1. and contour. #'•*-"I#/-'*• JV a20) This equation.24) and (1.22) (1.18) and (1.21) and (1. known as the equation of continuity.23) (1. S is an arbitrary open surface bounded by contour C.2 The General Differential Form Maxwell's equations in differential form can be derived from (1. 1. is given by d dt.0 8B (Faraday's law) (Maxwell-Ampere law) (Gauss's law) (Gauss's law—magnetic) (1.22). the field is called static. and (1. (1. 5 is a closed surface enclosing volume V. Among the five equations above.25) or (1. Application of Gauss's and Stokes's theorems to (1. They can be considered as the fundamental equations governing the behavior of electromagnetic fields. only three are independent for the case of time-varying fields and thus are called independent equations.21) (1.

33) does not depend on Efo) for £2 > ^i1. if a time-harmonic field is known for any LU.27). Using the complex phasor notation [3].27). and it is also evident that the static case is the limiting case of the harmonic fields as the frequency u approaches zero.28) being a natural consequence of (1. (1. The use of time-harmonic fields is not as restrictive as it first appears.32). Maxwell's equations become definite when constitutive relations between the field quantities are specified.30) (1.D + J V V-J J = -jup -jup where the time convention e3ujt is used and suppressed and UJ is angular frequency.32) using E(UJ) given by (1.21). the field is referred to as time-harmonic.33) EM = — / E{t)e~3UJt dt.32) (1.24) remain the same. Using Fourier analysis.31) (1-31) x H juuB J V x H = ju. The constitutive relations describe the macroscopic .22). and (1. It is evident that in this case there is no interaction between electric and magnetic fields. with (1. E(CJ) = 2W-OO EMe-^dt. ^ J-oo 1 f°° Therefore.33) do not violate causality.2.4 Show that the Fourier transforms defined in (1. therefore. In other words. (1.23) and (1. its counterpart in the time domain can be obtained by evaluating (1.23) and (L26) or a magnetostatic case described by (1. Exercise 1. It is evident that in this case.33) (1.29) (1. 1.'Wwf duj E(u)ejuJt du (1.6 BASIC ELECTROMAGNETIC THEORY whereas (1.2.25) can be written in a simplified form as V x E = -JUJB VxE=-jo.B (1.32) and (1. we can have separately either an electrostatic case described by (1.4 Time-Harmonic Fields When field quantities in Maxwell's equations are harmonically oscillating functions with a single frequency.5 Constitutive Relations The three independent equations among the five Maxwell's equations described above are in indefinite form since the number of equations is less than the number of unknowns. any time-varying field can be expressed in terms of time-harmonic components via the the Fourier transforms /»OO E(t) = / J — oc — oo E(u.24) and (1. the electric and magnetic fields must exist simultaneously and they interact with each other. show that E(ti) as evaluated from (1.

23) with the aid of (1. This is demonstrated here by considering the electro.37) into (1.39) into (1. /i. one obtains -V-(eV<£) =p (1. Equation (1. the electrostatic field is governed by (1.37) where <fi is called the electric scalar potential.38) is the well-known Poisson equation. 1 3 SCALAR AND VECTOR POTENTIALS .24) and (1.34). Equation (1.3. (1. respectively. the permittivity (farads/meter). and conductivity (siemens/meter) of the medium.1. they are D = eE .27).24) can be satisfied by representing the magnetic flux density B as B = VxA (1. These parameters are tensors for anisotropic media and scalars for isotropic media. 1. Substituting (1.40) .38) which is the second-order differential equation governing (p.and magnetostatic cases.26). one may first convert the first-order differential equations involving two field quantities into second-order differential equations involving only one field quantity. they are functions of position.2 Vector Potential for Magnetostatic Field The magnetostatic field is governed by (1. To solve Maxwell's equations.1 Scalar Potential for Electrostatic Field As we mentioned above. For a simple medium.39) where A is called the magnetic vector potential.23) and (1. The latter can be satisfied by representing the electric field E as E = -V0 (1.3.36) where the constitutive parameters e.crE J =aE (1.35) yields the second-order differential equation V x (-V XA)=J. and a denote.3 SCALAR AND VECTOR POTENTIALS 7 properties of the medium being considered.27) with the aid of (1.34) (1. 1. For inhomogeneous media. whereas for homogeneous media they are not. permeability (henrys/meter). Substituting (1.fill B = fjH .35) (1.

5 Formulate time-harmonic fields in terms of a vector and a scalar potential and discuss possible choices for a gauge condition.4. whenever possible we simplify problems by using a two-dimensional model to approximate a three-dimensional problem.30) with the aid of the constitutive relations (1. V x f~j"\ V HJ V These equations are called inhomogeneous vector wave equations. the governing differential equations involving only either field. Therefore. it is not necessary to impose the gauge condition.36). in a manner similar to the above [2]. does not determine A uniquely because. Vx v /.43) also satisfies (1. Exercise 1. -juJ. 1.-A .2 Scalar Wave Equations 7) (1. ^ eV v H ) u2 2/JH. 1. Assume that .42) /I J V x f -1X7 x T ^ .34)—(1. (1. In the time-harmonic case. For this. however. and a natural choice for this condition is V .43) In electromagnetic analysis.UJi2H = V x [ . It is evident that the solution of (1. which involve both electric and magnetic fields. one can eliminate E to find the equation for H as /I (1. to determine A uniquely. if our objective is to calculate B. it is necessary to derive from Maxwell's equations. we will deal with the time-harmonic case directly in terms of the electric and magnetic fields. Such a condition is called a gauge condition. one obtains V x ( .V x E J ~cu22eE = -juJ.A = 0.4. and the solution of (1. the electric and magnetic fields can also be represented by introducing a scalar and a vector potential.1 Vector Wave Equations The differential equation for E can be obtained by eliminating H from (1.29) and (1.42) also satisfies (1. if A is a solution to (1.24).41) It is important to understand that B is always unique even if A is not. However. any function that can be written as A' = A + V / is also a solution regardless of the form of / .V x E . Doing this.uo eE i Similarly. one must impose a condition on its divergence. this will not be discussed here because in this book we will work directly with the electric and magnetic fields for time-harmonic problems. Thus.23).40). 1. The discussions above are pertinent to the static case.4 WAVE EQUATIONS As just mentioned.8 BASIC ELECTROMAGNETIC THEORY This equation.

49) (1-49) h . In other words. the fields must satisfy four conditions. It can then be shown that the z-components of (1. Zo ( y/fio/eo ) — is the intrinsic impedance of free space. .5. only one of them is the real solution to the problem. These four equations are also known as field continuity conditions. given by h x (Ei .43) become T" ( — 4~) dx \(j. Bi B2 = 0 f i • ((Bi-.48) (1. Equations of the type of (1.45) respectively.rdxj and + T" ( — 4-)+ko€r] Ez=jk0ZQJz dy \firdyj J d ( \ d + ukliir\Hz J dy \er dy J dy. a complete description of an electromagnetic problem should include information about both differential equations and boundary conditions.44) and (1. A (1. 1. 1.H 22) = 0 H ) (1.45) (1.E 2 ) = 0 (Ei h. one must know the boundary conditions associated with the domain. respectively. which are assumed here to be complex scalar functions of position.44) \ d ( 1 d\ dx K r r dx) [dx '\eedxj = ~~TT~ I d fi dx \er dx \er \ J J "I" 7T~ I •JxJx I ) 0 / i dy \e r J dy \er (1.16)—(1.1). 1. say medium 1 and medium 2.42) and (1.47) (1. where er (= e/e0) and jir (= /x//io) denote the relative permittivity and relative permeability.5 BOUNDARY CONDITIONS While there are many functions that satisfy the differential equations given above in a domain of interest.5 BOUNDARY CONDITIONS 9 the fields and the associated medium have no variation with respect to one Cartesian coordinate. say the z-coordinate.1 At the Interface between Two Media At a source-free interface between two media.854 x 1CT12 farad/meter) and /io (= 4TT x 10~ 7 henry/meter) are the permittivity and permeability of free space.47) (1.1. and e0 (= 8. In this section we present some boundary conditions that apply to many practical problems.46) (1.45) are called inhomogeneous scalar wave equations.B)2 )= 0 where h is the unit vector normal to the interface pointing from medium 2 into medium 1 (Fig.19). These can be derived from Maxwell's equations in integral form (1. To determine this solution. k0 (= cj^/eo/io ) *s t n e wavenumber in free space.(Di --D 22)) = 0 (Di D =0 h x (Hi .

49) reduces to n -•B = 0 B = 0 (1.48).1 Interface between two media. Fig.48) it is assumed that neither surface currents nor surface charges exist at the interface.5. say medium 2.54) .3 At an Imperfectly Conducting Surface When medium 2 is an imperfect conductor. becomes a perfect conductor. Among these four conditions.50) and (1.(h • E)n = 7]Zon x H (n r/Z ofi H (1.51) Derive the boundary conditions (1. it can be shown [8] that the electric and magnetic fields at the surface of the conductor are approximately related by E .18) and At a Perfectly Conducting Surface The boundary conditions can be simplified when one of the media. Since a perfect conductor cannot sustain internal fields.46) becomes nxE = 0 xE and (1. i J Exercise 1. Note that in this case the boundary can always support a surface current (Js = n x H) and a surface charge (ps = h • D).1 Interface between two media.51) from (1. 1. 1.46) and (1. (1. and the other from (1. these two equations must be modified to read h.47) and (1.(Di ~ D ) =p8 n .5.47) and (1.2 (1. If indeed there exists a surface electric current density J s and a surface charge density p s .49).( D i . Medium 2 th>*2 Fig.10 BASIC ELECTROMAGNETIC THEORY n Medium 1 /^e. 1.D 22) = p3 nx hx ( H ! . Note that in (1. only two are independent: one from (1. 1.52) (1.54) (1.53) where E and B are the fields exterior to the conductor and h is the normal pointing away from the conductor.17).50) (1-50) (1.6 (1.H 2 ) ==JsS.

n (Hi (1.1.5 BOUNDARY CONDITIONS 11 or alternatively. In the two-dimensional case.= jkor]H ^T-=jkoriHz z dn on (1.4 Across a Resistive and Conductive Sheet An electrically resistive sheet is a thin sheet of electric current with density proportional to the tangential electric field at its surface [9].54) or (1.62) into (1. we obtain h x (h x E) = -Rn x (Hi .60) (1.3—-h x (n x E) = 0 x V x E) ft Mrl V or 1 —n — n x (V x H) . the boundary conditions across its surface are n x ( E i --EE 2 ) .62) (1.H)n . Based on (1.58) for the case of E = zEz (usually referred to as the £^-polarization case). It can be written in a more standard form as 1 jh 0 —h n x ((V x E ) . Equation (1.H 2 ). and 1 — eri er\ dHz ".46) and (1.jkonn x (n x H ))= 0 -jkonn xH =0 (1.55) (1.59) for the case of H = zHz (often referred to as the Hz-polarization case). the impedance boundary condition can be written as 1 dEz jkor ±dE±_JkoE-E } /jiri dn jJLrl On 7] n (1.00 x (Ei 2) = n x (Hi2).61).55) where q = y/JJhvJ&rt is the normalized intrinsic impedance of medium 2.63) .51). 1. Substituting (1.H 2 ) = J ss =J ftx(Hi-H with hx (fixE) = -RJa n x (h x E) = -RJS (1.55) is called an impedance boundary condition.H H] h 0 \(n • (1.56) (1. ft x E = nZ0 [(n .57) ei e rrl erl which is known as a mixed boundary condition or a boundary condition of the third kind.5.61) where R is the resistivity in ohms per square.

l ) r 3ko(nr . A condition must be specified at this outer boundary to obtain a unique solution for the problem. In general.66) where G denotes the conductivity in Siemens per square. A conductive sheet is useful because it can be used to approximate a thin magnetic layer with G = — — ^ — jko{lJ.E 2 ) (n -Gh E2) (1.67) where Yo = 1/ZO and fir is the relative permeability of the layer. The boundary conditions across its surface are h x (Ht .65) (1. Assuming that all sources and objects are immersed in free space and located within a finite distance from the origin of a coordinate system. the domain is called unbounded or open.6 RADIATION CONDITIONS When the outer boundary of a domain recedes to infinity. For two-dimensional fields. the Sommerfeld radiation condition becomes ^^[|(S) ^(S)]=° lim ^fp dp' p—ocwhere p = \Jx2 + y2.l ) o r (L64) jko{er . o Hy = 0 (L68) (1.n R=^r-A-rrr jk {e . The electromagnetic dual of a resistive sheet is a magnetically conductive sheet supporting only a magnetic current. the electric and magnetic fields are required to satisfy lim V x ) + jk x ™rr[VX{H)+ikofXr(l)](=° v /E1 r—>-oo E VH.i ) r >o (1.12 BASIC ELECTROMAGNETIC THEORY Resistive sheets are useful in numerical simulations because they can be used to approximate thin dielectric layers with 7.H 22)) = 0 n (Hi H h x (ft x H) = -Gh x (Ei . = y/x 2 + z HZJ d 'Ez + + jko 'E \HZ.68) is usually referred to as the Sommerfeld radiation condition for general three-dimensional fields.68) where r = y/x2 + y2 + z2.r . Equation (1.6) i6 9 - . = 0 ((1.l)r where er is the relative permittivity and r is the thickness of the dielectric layer. a resistive and a conductive sheet can be combined to model a thin material layer whose permittivity and permeability both differ from those of the surrounding medium [9]. 1. Such a condition is referred to as a radiation condition.

Better accuracy can be achieved by developing higher-order conditions (Chapter 9).42) and (1.70) (1.7 FIELDS IN AN INFINITE HOMOGENEOUS MEDIUM 13 Equations (1.76) = V x jjj J(r')Go(r. r')] dV = (-JULL + ~^VV^J jjj J(r')G0(r.71) as E(r) = JJJ [-j^J(r') .1. In numerical analysis.r')d^' . known as the scalar Green's function (see Appendix C). (1. one can find the solution to (1. 4TT: (1-74) Equations (1. given by e-jk\r-r'\ Go(r^) = —-—. (1.69) can be regarded as the lowest-order radiation conditions with limited accuracy. r') denotes the point-source response.72) (1. It is important to recognize that (1. the problem at hand involves an infinite homogeneous medium. (1.24). 1.V x J (1.70) is equivalent to (1.r') + %(r')V'G0(r. When applied at such a finite boundary.r')dV" (1. closed-form solutions may be obtained.68) and (1. similarly.73) where Go(r.r')dV" (1.69) are exactly valid at infinity.r')dV' (1. In such a medium. where k = ujy/ejl. however.71) is equivalent to (1. Using the principle of linear superposition.43) only when it is solved together with (1.75) H(r) .V p V 2 H-h& 2 H = . it is often desirable to reduce the size of a computational domain by using a finite boundary to truncate the infinite domain. where e and /i are constant.Vp(r')] G0(r.r')dV" H(r) = jjj [V x J(r')]G0(r.23).JJJ/^) x V'G0(r.72) and (1.73) can also be written as E(r) = JJJ [-jc^J(r')Go(r. (1.43) do not have a closed-form solution. and a numerical method is often required to seek their approximate solution.70) and (1.23) and (1.43) become V 2 E + /c2E = jujfiJ + .42) only when it is solved in conjunction with (1.24).71) with the use of (1.7 FIELDS IN AN INFINITE HOMOGENEOUS MEDIUM In a general inhomogeneous medium.68) and (1. If.42) and (1.

The results above are also applicable to surface or line sources provided that the volume integrals are replaced by the corresponding surface or line integrals.70) and (1.71) is reduced to V 2 H + k2H = 0. it would be useful to have expressions to calculate the fields everywhere based on knowledge of the fields surrounding an object. Exercise 1. Exercise 1.jfe2H = 0 and (1. r')} dS' (1.p') (1. r7) + [rV x H(r')] x V'G 0 (r.72) and (1.14 BASIC ELECTROMAGNETIC THEORY with the use of vector identities and integral theorems.p') = ±:H™(k\p-p'\) (1. r') + [rY • E(r')] V'G 0 (r.24) either. r')} <iS' H(r) =(JJ){jooe [hf x E(r')] G 0 (r.77) where G0(p. Therefore. Show that H = xHoe-jkx does not satisfy the former equation.8 HUYGENS'S PRINCIPLE The preceding section provided expressions to calculate the fields radiated by a given current source.76) are equivalent to (1. (1.2).80) .r')dz' = G0(p. Therefore. They can also be reduced for two-dimensional problems by using the relation i: J — oc G0{r.78) in which p = xx + yy and HQ \k\p — p'\) is the zeroth-order Hankel function of the second kind.79) + [nf x E(r')] x V'G 0 (r. if a source is known in an infinite homogeneous medium. source-free region. the fields can readily be calculated by evaluating appropriate integrals. However. Show that this field does not satisfy (1. the fields outside the surface are given by E(r) = H {-jw [n! x H(r')] G 0 (r.73). 1.75) and (1. r') (1. but satisfies the latter.43) is reduced to V x V x H .8 Show that (1. Consider a surface S enclosing the source of radiation and any other objects so that the infinite space exterior to the surface is homogeneous (Fig. If the fields on the surface are known. 1. This exercise demonstrates that the solution to (1.71) does not necessarily satisfy Maxwell's equations. in the finite element method we usually solve for the fields surrounding an object instead of the currents induced on the object. r') + [hf • H(r')] V 7 G 0 (r.7 In a homogeneous. Huygens's principle provides such relations.

p') HP) dn> dri Jr I where <j> = Ez for £ 2 -polarization.and #2-polarizations.p>) d4>(p') (1.r') dS' H(r): = jj^ [j s(r') x V'Go^r') . dimensions using the relation in Again.77). and pe and p m are equivalent to surface electric and magnetic charges.80) using (1.79) and (1.r / )ldS / .81) and (1. (1.r') + ± (r')V'G (r.1.2 Huygens's principle. 1 G p *)^r-«w^ dT' (1-83) *<"> = i {^-^tr ~ ^ '^\ and F is a contour where <j> = Ez for £2-polarization.75) and (1. all of the results above can be reduced to two dimensions using the relation in (1. In particular. These results can be derivedrigorouslyusing Green's theorems (see Chapters 10 and 14). where h' denotes the unit vector normal to S at r' and pointing toward the exterior region.rf)\dS' -MsB (r')xV'Go(r. (1.r') E(r) a e 0 Js -M (r')xV'G0(r. we have dG0(p.77). <fi = Hz for Hz -polarization.9 Derive (1.r') H(r) JJs I Js (r') x V'G0(r. Rewrite the right-hand side of (1.80) can be written as .r') E ( r ) := fj> [-ja^ (r')Go(r.76) indicates that J s and M s are equivalent to surface electric and magnetic currents. and n' • //H(r') = pm(r').j^M s (r')G 0 (r. (1. (1. Exercise 1. Because of this interpretation.76) indicates that J s and M s and A comparison of these two equations with (1. If we let n' x H(r') = J s (r').82) are also known as the surface equivalence principle.82) + M m (r / (r')VGo(r.r') . .83) in terms of equivalent surface currents for both Ez.81) 1 + ip -pm)V / G (r.r') + %Pe(r>)VG0(r. and F is a contour that encloses the source of the field as well as all possible objects.83) from (1.82) 0 M A comparison of these two equations with (1.75) J (1.79) and (1. v! • cE(r') = p e (r').81) (1. E(r') x n' = M a (r').r')\dS'.8 HUYGENS'S PRINCIPLE 15 h S Sources Exterior medium &e Objects Fig.j WM J s (r')Go(r.and Hz-polarization cases.jwcM^rOGoCr. 1. for both Ez.!/) (1.83) dT' -Go(p.

9 RADAR CROSS SECTIONS One of emphases of this book is the treatment of open-region scattering problems.r.82).<p)\ c 2 (1.^nc) ff(^.<p incc)| 22 inc E V m in r—>oo '^' '^ ' r->oo |E ((9 .83) can be simplified to > .82) become E(r)» E(r) « e-^— H {jiOfif x [f x J sa(r')] + jkr x M8s(r')} } ejk*'r'dSf [jufir (r')] +JAX M ( r ' ) jkf" 'dS' Us Js e-jkr e -ifcr (1.87) (1. a is called the monostatic or backscatter radar cross section.(p )| 6M = lm 47rr2 ^2' lim = r—>oo W Uinc(0inc)(/?inc)| i L inc(#inc^inc)|2 r—>oo H W (r.6»inc.ef x [f x M a (r')] x J ( r ) } e? *' 'dS'. An important parameter that characterizes scattering by an object is called radar cross section [10].v5. Under the condition p — oo.87) H(r)» w H(r) € Aivr ^ r S {juer x [f x Ms(rf))-jkf jfcr x J aa( r ..p T = llim Z7T/C |Hngq^Jl r i c inc 2 P-+00 P |H |H mc ((^mc)|2 (^ )| V (1.<p )\ d-85) where p and q represent either 8 or {p.^)| = i m 2.88) } which is also referred to as echo width. {jo. the radar cross section is defined by E (r.p)| |E s c^4 2 lim 4nr CT(6».81) and (1. otherwise. The scattered field in the far zone can be evaluated using expressions such as (1.i ^ :iac1 . the radar cross section is defined by oW 1 0 ) V .H SC ) denote the scattered field observed in the direction (#. It is often normalized to A with a unit of dBw or m with a unit of dBm. under this condition.P".0.V. ipinc). The unit for a/\2 is dBsw and for <j/m2 is dBsm.<S°*) = .) } e?kkf'rr'dS'.y)| 2 \E ™{6 . This parameter is defined for plane-wave incidence.0. l i m ^ .84) (1-84) where (E SC . These expressions » can be simplified greatly for r — oo. To evaluate the radar cross section. = lim p—oo p—>oo 2TI P IE-(p ) ( p)| 2 | E l n c ((<^ inc))|2 |E inc ( ^ m c | 2 sc 2 (p.83).16 BASIC ELECTROMAGNETIC THEORY 1. we have to compute the scattered field in the far zone with r — oo using expressions such as (1. it is referred to as the bistatic radar cross section.81) and > (1. For example. To incorporate information about polarization.g. (1.88) (1.</?) and (E i n c . When the incident and observation directions are the same.r) =iimwJm c (0 incc. The radar cross section is often normalized to either A2 or m2. inc ^ . the radar cross section can also be defined by *n(0. 1 |J5^(r. In the three-dimensional case.86) (1. H i n c ) denote the incident field from the direction (<9inc. (1. In the two-dimensional case.

Exercise 1. We then presented Maxwell's equations in integral form as the fundamental equations.11 Derive (1. However. uniqueness theorem. The reader is encouraged to consult [2-7] for this purpose. reciprocity theorem. Although this chapter has not reviewed some other important concepts. New York: McGraw-Hill. New York. These will all be used in the finite element analysis of electromagnetics problems.89) from (1.86) and (1. the time-harmonic problem was formulated directly in terms of the electric or magnetic field. . This chapter also reviewed the integral expressions that relate the field to its source and Huygens's principle for calculating exterior fields from the field on a closed surface.10 Derive (1. Reprinted by Dover Publications. J. and equivalence principle.10 SUMMARY In this chapter we first reviewed the basic concepts and integral theorems in vector analysis.81) and (1. Maxwell. 1. 2. The electrostatic problem was formulated in terms of a scalar potential. and the magnetostatic problem was formulated in terms of a vector potential.REFERENCES 17 read «*)'•{£?-"'I </>(p) 8np \p-n'Hp') i d<t>(p'yikp-P' e jk an' dr. C. A. 1954. Finally.82) when r -> oo.89) by using the asymptotic expression for the Hankel function. from which Maxwell's equations in differential form and boundary conditions were derived. J. 1873. This was followed by the presentation of a variety of boundary conditions and radiation conditions. (1. Exercise 1. A Treatise on Electricity and Magnetism.87) from (1. 1941. we presented the definition of radar cross section and its calculation because it is used often to present the results of application of the finite element method to scattering problems. Stratton. Helmholtz. Electromagnetic Theory. a good comprehension of these concepts can help greatly in computational electromagnetics research. REFERENCES 1. The corresponding partial differential equations were derived in the form of the Poisson.83) when p -> oo. These will be used frequently in the chapters to follow. and vector wave equations. although it was also possible to use scalar and vector potentials for formulation. Oxford: Oxford University Press. such as the duality principle.

Van Bladel. New York: McGrawHill. vol. 1985. 6." IEEE Trans.. Electromagnetic Fields. Balanis. J. D. Senior. "Combined resistive and conductive sheets. L. Kraus. B.18 BASIC ELECTROMAGNETIC THEORY 3. 1989. J. 1987. 1975. New York: McGraw-Hill. 1964. New York: McGraw-Hill. "Impedance boundary conditions for imperfectly conducting surface. J. New York.. E. J. Electromagnetics (4th edition). A. 1961. B. Harrington. vol. Revised printing by Hemisphere Publishing. F. C." AppL Sci. Senior. R. New York: Wiley. Bowman. A. Revised printing by Hemisphere Publishing. Res. Classical Electrodynamics (2nd edition). 4. 5. . D. 1969. New York. May 1985. T. Uslenghi. 577-579. 9. Advanced Engineering Electromagnetics. Jackson. A. 10. pp. Time-Harmonic Electromagnetic Fields. 8. 1960. Antennas Propagat. Amsterdam: North-Holland. T. 418-436. 8. 7. AP-33. and P. J. New York: Wiley. Eds. 1973. pp. A. T. 5. Senior. B. Electromagnetic and Acoustic Scattering by Simple Shapes.

- electromagnetic filed theory
- shuvo
- [ QUA] Quantum Field Theory for Mathematicians_ Hamiltonian Mechanics & Symplectic Geometry
- Review of Vector Analysis
- Vector II 2015
- Multivariable Calculus Review Sheet
- Ch1 Co Ordinate System
- Advanced FT
- Rotation
- Dyadics Identities
- Dyadic Notations
- 67985_02
- lec44.pdf
- Fourier Series
- Chap1_JQPMaster
- zEMFT Thidé Book Exercises Eletromagnetic Field Magnetic
- Noether Talk
- Chap15_Sec6
- Linear System and Background
- PhysRev.159.1251
- Ece Signals and Systems
- Finite Element Analysis
- IGCSE Edexcel Further Pure Mathematics Formula Note
- Three Introductory Lectures on Fourier
- Fundamental Solutions in the Theory of Microfluids
- Curious Space Time Singularities
- hill en
- Aero 7sem Ae2403nol
- Fourier Transform - Wikipedia, The Free Encyclopedia
- euler savary equation for kinematic synthesis

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue listening from where you left off, or restart the preview.

scribd