Performance of TCPR and FDTS/DF for high-density magnetic recording

We compared the performance of trellis-coded partial response (TCPR) and fixed delay tree search with decision feedback (FDTS/DF) in a recording system with both electronic and medium noise characterized by Thurlings´ model. To distinguish influences of the noise correlation on the Viterbi algorithm and the tree search algorithm, the bit error rate with and without noise correlation was calculated by using a noise correlation coefficient after equalization. We obtained the signal-to-noise ratio required for a bit-error-rate of 10^-9, versus a normalized recording density for recording system with 50% electronic noise and 50% medium noise. The results indicate that TCPR gives the best performance for low linear recording density