If the integral ∫ 0 ∞ ( e 2 π x + 1 ) ( n 2 + 4 x 2 ) e π x d x = a n 1 ( ψ 0 ( c n + b ) − ψ 0 ( c n + d ) ) where a , b , c , d are positive integers, then find a + b + c + d .
Notation: ψ 0 ( ⋅ ) denotes the digamma function .
This is an original problem and shared as proposed problem in Romanian Mathematical Magazine for proving the closed form .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Thank you for the solution, sir. I did it with real method. Can you please let me know books for beginners in complex analysis? I would be grateful to you, Thank you.
The variant of this problem is here ,I hope you will like it.
Problem Loading...
Note Loading...
Set Loading...
Let us define X ( n ) = ∫ 0 ∞ ( e 2 π x + 1 ) ( n 2 + 4 x 2 ) e π x d x n > 0 noting that X ( n ) = ∫ 0 ∞ ( e π x + e − π x ) ( n 2 + 4 x 2 ) 1 d x = 2 1 ∫ 0 ∞ n 2 + 4 x 2 s e c h ( π x ) d x = 4 1 ∫ R n 2 + 4 x 2 s e c h ( π x ) d x If, for any positive integer N , C N is the square with vertices N + N i , − N + N i , − N − N i , N − N i , then it is standard that ∣ cosh ( π z ) ∣ ≥ 1 for all z ∈ C N . Thus, if R N is the positively oriented rectangle with vertices N , N + N i , − N + N i − N , then we deduce that N → ∞ lim ∫ R N n 2 + 4 z 2 s e c h ( π z ) d z = ∫ R n 2 + 4 x 2 s e c h ( π x ) d x = 4 X ( n ) Suppose for the moment that n > 0 is not an odd positive integer. Then n 2 + 4 z 2 s e c h ( π z ) is meromorphic in the upper half plane, with simple poles at 2 1 n i and ( m − 2 1 ) i for all m ∈ N . Moreover R e s z = 2 1 n i n 2 + 4 z 2 s e c h ( π z ) R e s z = ( m − 2 1 ) i n 2 + 4 z 2 s e c h ( π z ) = 4 × i n s e c h ( 2 1 n π i ) = 4 i n 1 sec ( 2 1 n π ) = π sinh ( π ( m − 2 1 ) i ) [ n 2 − ( 2 m − 1 ) 2 ] 1 = π i ( n + 2 m − 1 ) ( n − 2 m + 1 ) ( − 1 ) m − 1 and since the residue calculus tells us that ∫ R N n 2 + 4 z 2 s e c h ( π z ) d z = 2 π i ( R e s z = 2 1 n i + m = 1 ∑ N R e s z = ( m − 2 1 ) i ) n 2 + 4 z 2 s e c h ( π z ) = 2 n π sec ( 2 1 π n ) + 2 m = 1 ∑ N ( n + 2 m − 1 ) ( n − 2 m + 1 ) ( − 1 ) m − 1 we deduce that X ( n ) = 8 n π sec ( 2 1 n π ) + 2 1 m = 1 ∑ ∞ ( n + 2 m − 1 ) ( n − 2 m + 1 ) ( − 1 ) m − 1 Now 8 n m = 1 ∑ ∞ ( n − 2 m + 1 ) ( n + 2 m − 1 ) ( − 1 ) m − 1 = 4 m = 1 ∑ ∞ ( n + 2 m − 1 ( − 1 ) m − 1 + n − 2 m + 1 ( − 1 ) m − 1 ) = 4 m = 1 ∑ ∞ ( n + 4 m − 3 1 − n + 4 m − 1 1 + n − 4 m + 3 1 − n − 4 m + 1 1 ) = m = 1 ∑ ∞ ( m + 4 n − 3 1 − m + 4 n − 1 1 − m − 4 n + 3 1 + m − 4 n + 1 1 ) = − ψ ( 4 n + 1 ) + ψ ( 4 n + 3 ) + ψ ( 4 1 − n ) − ψ ( 4 3 − n ) = − ψ ( 4 n + 1 ) + ψ ( 4 n + 3 ) + ψ ( 1 − 4 n + 3 ) − ψ ( 1 − 4 n + 1 ) = 2 [ ψ ( 4 n + 3 ) − ψ ( 4 n + 1 ) ] + π [ cot ( 4 1 ( n + 3 ) π ) − cot ( 4 1 ( n + 1 ) π ) ] = 2 [ ψ ( 4 n + 3 ) − ψ ( 4 n + 1 ) ] − 2 π sec ( 2 1 n π ) so we deduce that X ( n ) = 8 n 1 [ ψ ( 4 n + 3 ) − ψ ( 4 n + 1 ) ] making the answer 8 + 3 + 4 + 1 = 1 6 .
We have proved this formula for all n > 0 , except when n is an odd positive integer (in that case the pole at 2 1 n i is a double pole, not simple, which would change the details of the above calculations. However this final formula for X ( n ) can be extended to all n > 0 by continuity.