Page 215 - DMTH504_DIFFERENTIAL_AND_INTEGRAL_EQUATION
P. 215

Differential and Integral Equation




                    Notes          13.4 Orthogonality of Solutions of Some Equations


                                   (a) Orthogonality of Bessel’s Functions
                                   We know that J (x ) is the solution of Bessel’s equation
                                               n
                                              2
                                            2 d J n ( )  d J n ( )  2  2
                                                 x
                                                          x
                                           x         x        (x  n  ) J n ( ) = 0
                                                                        x
                                               dx  2    dx
                                   where n is a positive integer. Putting  x  , x  we have
                                                                       d J n  =  1 d J n
                                                                       dx        dx
                                                                                  2
                                                                       2
                                                                      d J n    1 d J n
                                   and                                   2  =  2   2  ,
                                                                      dx         dx
                                   where   is a constant,
                                            2
                                          2 d J  n ( x )  d J  n ( x )  2 2  2
                                         x     2    x       (  x  n  ) ( x ) = 0                           ...(i)
                                                                     J
                                                                      n
                                             dx        dx
                                   which may be rewritten as
                                                d   d J n (x  )  2  n 2
                                                   x           x     J  n ( x ) = 0
                                                dx    dx          x
                                   which is Sturm Liouville equation for each fixed n i.e.
                                               d     d
                                                 p ( )  J n ( x )  [ ( )  1  r ( )]y  = 0
                                                               q
                                                   x
                                                                x
                                                                       x
                                              dx     dx
                                                                                      n 2
                                   with                                p(x) = x , ( )     and  ( )r x  x  and   1  2  .
                                                                                q
                                                                                  x
                                                                                       x
                                   Since p(x) = 0 for x = 0, it follows that the solution of (i) on an interval 0  x  a satisfying the
                                   boundary conditions
                                                                     J n ( a ) = 0                         ...(ii)
                                   form an orthogonal set with respect to the weight p(x) = x.
                                   Let   <    <    ... denote the positive zeros of J (x ), therefore (ii) holds for
                                       1n   2n  3n                         n  1

                                                               a =    or  =    =   mn          (m = 1, 2, ... n fixed)
                                                                    mn      mn    a
                                            d
                                               x
                                   and since   J n ( )  is continuous also at x = 0, therefore for each fixed n = 0, 1, 2, .... , the Bessel’s
                                           dx
                                                                       mn
                                   function  J  (  ) x  (m = 1, 2, ...) with   ,  form a orthogonal set on an interval 0  x  a
                                           n  mn                  mn   a
                                   with respect to weight function p(x) = x,

                                                          a
                                                           x J  n  mn x J n  pn x  = 0                   if p   m
                                                          0




          208                               LOVELY PROFESSIONAL UNIVERSITY
   210   211   212   213   214   215   216   217   218   219   220