Comparison of Hardware Complexity between Convolutional Guided Scrambling and Galois-Field-Addition Guided Scrambling for Digital Recording

Abstract

The guided scrambling (GS) coding technique is used to control the runlength within code blocks, such as d or k, as well as to suppress the DC component. It achieves a code rate very close to the channel capacity. In a GS encoder, a selection set of channel bit streams is constructed by subjecting a source bit stream to several kinds of scramblings and to runlength limited (RLL) coding if necessary. Then one that satisfies the imposed channel constraints best is selected. Typically, the GS technique uses a convolutional operation or an addition operation in the Galois field (GF); these two schemes can generate channel bit streams that meet the channel constraints better than conventional coding techniques. The latter scheme is better in terms of symbol-error probability when both schemes are concatenated with Reed-Solomon (RS) codes. However, nothing is known about the hardware complexity of either scheme. We describe their hardware configurations using ROM tables and compare the table sizes. From this comparison, we conclude that the hardware complexity of the latter scheme is smaller under certain conditions.