1. To propose new class of multiple splitting algorithms based on alternating directions and optimal gradient techniques that are much more efficient and requires much less memory than previous algorithm for solving the inverse problems.
2. To prove the convergent theorems for new class of multiple splitting algorithms
that proposed.
3. To reformulate the inverse problems corresponding with image restoration and to investigate approximate solutions to transformed problems.
4. To construct and improve new fixed point iterative schemes by using hybrid projection methods for solving the network bandwidth allocation problem that can be more effectively used in a wider situation.
5. To investigate the convergent theorem for the proposed network bandwidth
allocation problem
6. To present the simulation testes and the numerical examples for confirming our results.
