In other words, compared with the BPDN algorithm, it improves the performance at the cost of multiple weighted [L.
Then the proposed algorithm degrades to the standard BPDN algorithm.
Compared with the BPDN algorithm, the proposed algorithm improves performance by utilizing the prior information of DOA changing scope.
In Figure 2, the proposed algorithm is compared with the BPDN algorithm where enough DOA separation among all DOAs is assumed.
In Figure 3, the proposed algorithm is compared with the BPDN algorithm where the BPDN algorithm cannot resolve the little DOA separation.
On the other hand, even if we have no a priori information about the DOA changing scopes, the proposed algorithm will degrade to the conventional BPDN algorithm where the weight matrix is set as a unit matrix.