1
$\begingroup$

We try to decompose 8x8 unitary matrices \begin{pmatrix} e^{ia} & & & & & & \\ & e^{ib} & & & & & \\ & & e^{ic} & & & & \\ & & & e^{id} & & & \\ & & & & e^{ie} & & \\ & & & & & e^{if} & \\ & & & & & & e^{ig} \\ & & & & & & & e^{ih} \end{pmatrix} using the methods answered in here,by taking the tensor product of three phase gates formed $P_1 = \begin{pmatrix} 1 & 0 \\ 0 & e^{ii} \end{pmatrix}, \quad P_2 = \begin{pmatrix} 1 & 0 \\ 0 & e^{ij} \end{pmatrix}, \quad P_3 = \begin{pmatrix} 1 & 0 \\ 0 & e^{ik} \end{pmatrix}$,

and element wise with the 4 controled-controled-phased gate: \begin{pmatrix} 1& 0& 0& 0& 0& 0& 0&0 \\ 0& 1& 0& 0& 0& 0& 0&0 \\ 0& 0& 1& 0& 0& 0& 0&0 \\ 0& 0& 0& 1& 0& 0& 0&0 \\ 0& 0& 0& 0& 1& 0& 0&0 \\ 0& 0& 0& 0& 0& 1& 0&0 \\ 0& 0& 0& 0& 0& 0& 1&0 \\ 0& 0& 0& 0& 0& 0& 0&e^{i(\lambda)} \end{pmatrix}

We have the result of tensor product of three phased gate $P_1 \otimes P_2 \otimes P_3$: \begin{pmatrix} 1& 0& 0& 0& 0& 0& 0&0 \\ 0& e^{ik}& 0& 0& 0& 0& 0&0 \\ 0& 0& e^{ij}& 0& 0& 0& 0&0 \\ 0& 0& 0& e^{i(k+j)}& 0& 0& 0&0 \\ 0& 0& 0& 0& e^{ii}& 0& 0&0 \\ 0& 0& 0& 0& 0& e^{i(k+i)}& 0&0 \\ 0& 0& 0& 0& 0& 0& e^{i(i+j)}&0 \\ 0& 0& 0& 0& 0& 0& 0&e^{i(k+j+i)} \end{pmatrix}

Taking global phased factor $e^{il}$: \begin{pmatrix} e^{il}& 0& 0& 0& 0& 0& 0&0 \\ 0& e^{i(k+l)}& 0& 0& 0& 0& 0&0 \\ 0& 0& e^{i(l+j)}& 0& 0& 0& 0&0 \\ 0& 0& 0& e^{i(k+j+l)}& 0& 0& 0&0 \\ 0& 0& 0& 0& e^{i(i+l)}& 0& 0&0 \\ 0& 0& 0& 0& 0& e^{i(k+i+l)}& 0&0 \\ 0& 0& 0& 0& 0& 0& e^{i(i+j+l)}&0 \\ 0& 0& 0& 0& 0& 0& 0&e^{i(k+j+i+l)} \end{pmatrix}

We obtain eight nontrivial conditions, one for each diagonal entry:

  • a = l
  • b = l+k -> k = b - a
  • c = l+j -> j = c - a
  • d = l+k+j -> d = b + c - a
  • e = l+i -> i = e - a
  • f = l+k+i -> f = b + e - a
  • g = l+i+j -> g = e + c - a
  • h = l+i+j+k -> h = e + c + b - 2a

And then we put value of i,j,k,l back into the matrix when we taking the global phases factor. \begin{pmatrix} e^{ia}& 0& 0& 0& 0& 0& 0&0 \\ 0& e^{ib}& 0& 0& 0& 0& 0&0 \\ 0& 0& e^{ic}& 0& 0& 0& 0&0 \\ 0& 0& 0& e^{i(b+c-a)}& 0& 0& 0&0 \\ 0& 0& 0& 0& e^{ie}& 0& 0&0 \\ 0& 0& 0& 0& 0& e^{i(b+e-a)}& 0&0 \\ 0& 0& 0& 0& 0& 0& e^{i(e+c-a)}&0 \\ 0& 0& 0& 0& 0& 0& 0&e^{i(e+c+b-2a)} \end{pmatrix}

We wonder how we can construct the controled-controled-phase gates to satisfy and achieve the original unitary matrice or did we make any mistake through our work. We would love to hear your response. \begin{pmatrix} e^{ia}& 0& 0& 0& 0& 0& 0&0 \\ 0& e^{ib}& 0& 0& 0& 0& 0&0 \\ 0& 0& e^{ic}& 0& 0& 0& 0&0 \\ 0& 0& 0& e^{id}& 0& 0& 0&0 \\ 0& 0& 0& 0& e^{ie}& 0& 0&0 \\ 0& 0& 0& 0& 0& e^{if}& 0&0 \\ 0& 0& 0& 0& 0& 0& e^{ig}&0 \\ 0& 0& 0& 0& 0& 0& 0&e^{ih} \end{pmatrix}

$\endgroup$

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.