1
$\begingroup$

The following indefinite sum with (CC[n]=1 for even n and CC[n]=i for odd n)

CC[n_] := (1/2 + I/2) (1 + I (-1)^(-1 + n))    

Sum[Exp[-\[Pi] \[Alpha]^2 (n2^2 + n3^2)] CC[n1 + n2] Conjugate[
CC[n1]] CC[n1 + n3] Conjugate[CC[n1 + n2 + n3]], {n1, 0, 
NN - 1}, {n2, -Infinity, Infinity}, {n3, -Infinity, Infinity}]

doesn't seem to work. Surely there is a better method?

$\endgroup$
  • $\begingroup$ In the sum above, NN=2 and [Alpha] is assumed to be real. $\endgroup$ – 121 Jan 23 '18 at 11:51
1
$\begingroup$

Tell MMA additional assumptions

summand = 
   Exp[-\[Pi] \[Alpha]^2 (n2^2 + n3^2)] CC[n1 + n2] Conjugate[
   CC[n1]] CC[n1 + n3] Conjugate[CC[n1 + n2 + n3]] // 
   FullSimplify[#, 
   n1 \[Element] Integers && n2 \[Element] Integers && 
   n3 \[Element] Integers] &

Sum[summand, {n1, 0, NN - 1}, {n2, -Infinity, 
   Infinity}, {n3, -Infinity, Infinity}] /. NN -> 2

(*    1/8 (8 EllipticTheta[3, 0, E^(-\[Pi] \[Alpha]^2)]^2 + 
      16 EllipticTheta[3, 0, E^(-\[Pi] \[Alpha]^2)] EllipticTheta[4, 0, 
      E^(-\[Pi] \[Alpha]^2)] - 
     8 EllipticTheta[4, 0, E^(-\[Pi] \[Alpha]^2)]^2)    *)
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.