CNOT-Patel Synthesis for linear reversible functions

Header: tweedledum/algorithms/synthesis/cnot_patel.hpp

Parameters

struct cnot_patel_params

Parameters for cnot_patel.

Public Members

bool allow_rewiring = false

Allow rewiring.

bool best_partition_size = false

Search for the best parition size.

uint32_t partition_size = 1u

Partition size.

Algorithm

template <class Network, class Matrix>
Network tweedledum::cnot_patel(Matrix const &matrix, cnot_patel_params params = {})

CNOT Patel synthesis for linear circuits.

This algorithm is based on the work in [PMH08].

The following code shows how to apply the algorithm to the example in the original paper.

std::vector<uint32_t> rows = {0b000011,
                              0b011001,
                         0b010010,
                         0b111111,
                         0b111011,
                         0b011100};
bit_matrix_rm matrix(6, rows);
cnot_patel_params parameters;
parameters.allow_rewiring = false;
parameters.best_partition_size = false;
parameters.partition_size = 2u;
auto network = cnot_patel<netlist<mcst_gate>>(matrix, parameters);

Parameters

• matrix: The square matrix representing a linear reversible circuit.
• params: The parameters that configure the synthesis process. See cnot_patel_params for details.

Warning

doxygenfunction: Unable to resolve multiple matches for function “tweedledum::cnot_patel” with arguments (Network&, std::vector<uint32_t> const&, Matrix const&, cnot_patel_params) in doxygen xml output for project “tweedledum” from directory: doxyxml/xml. Potential matches:

- template <class Network, class Matrix>
  Network tweedledum::cnot_patel(Matrix const&, cnot_patel_params)
- template <class Network, class Matrix>
  void tweedledum::cnot_patel(Network&, std::vector<qubit_id> const&, Matrix const&, cnot_patel_params)