In this paper, we introduce a new class of coders called minimum noise ladder-based biorthogonal (MINLAB) coders. The MINLAB coders belong to a subclass of two-channel biorthogonal filter banks (FB's), but its quantizers are placed at different locations. The minimum noise structure ensures that the quantization error has a unity noise gain, even though the FB is biorthogonal. The coder has a very low design and implementation cost. The perfect reconstruction property is structurally preserved. Moreover, the coding gain of the optimal MINLAB coder is always greater than or equal to unity, and it can be expressed in closed form. For both the AR(1) and MA(1) processes, the MINLAB coder with two taps has a higher coding gain than the optimal orthonormal coder with an infinite number of taps. In addition to its superior coding performance, it has many other desired features that make it a valuable and attractive alternative to the orthonormal coder.
|Number of pages||1|
|Journal||IEEE Transactions on Signal Processing|
|State||Published - 1 Dec 1999|