An entirely new approach for iterative signal reconstruction using partial information and reconstruction of the signals which are embedded in noise is presented. It is shown that the new algorithms are able to maintain their performance under extreme noise conditions where the existing algorithms completely fail. Convergence of these algorithms are also extremely fast but computation time taken is longer than the existing algorithms. More powerful algorithmic structures can be developed by simply convolving the new approaches with higher order cumulants in order to stduy signal reconstruction problems where the cost function is highly nonlinear.
|Number of pages||5|
|Journal||IEE Conference Publication|
|Publication status||Published - 1 Jan 1999|
|Event||Proceedings of the 1999 IEE National Conference on Antennas and Propagation - York, UK|
Duration: 31 Mar 1999 → 1 Apr 1999