A comparison of two coding schemes for generating DC-free runlength-limited sequences

Abstract

We report on the performance assessment of two methods for generating DC-free runlength-limited sequences. Redundant bits used to optimize the low-frequency properties are multiplexed with user data or encoded data. In the first method, the redundant bits are multiplexed with the channel bits generated by a (d, k) encoder. In the second method, on the other hand, the redundant bits are multiplexed with the user data. i.e. prior to application of the (d, k) encoder. The second method can only be used if the (d, k) encoder satisfies the parity preserving condition, i.e., data words and code words of that code must have the same parity. In the article we describe in detail an example of a parity preserving (1, 8) code. © 2000 Publication Board, Japanese Journal of Applied Physics.