科研之友
微信
新浪微博
Facebook
分享链接
摘要
Shingled writing on bit patterned media (BPM) is a promising technology to further increase the areal density beyond 1 Tbit/in(2) without drastic changes in writer design. In this paper, a shingled writer is modeled for writing a 2-D sector on BPM, and in particular, we investigate the potential benefit of leaving the last track in the 2-D sector untrimmed, making it fatter than the other tracks in that 2-D sector. During detection, the more reliable information from the last untrimmed (i.e., the fat) track is used as a priori information for detecting the data in the other tracks in that 2-D sector. The simulation results indicate that for 2-D sector with ten tracks (at 3.64 Tbits/in(2) channel bit density) and 5% media noise and at a target bit error rate of 10(-3), using the fat track information, 2-D equalizer +2-D target + symbol-based pattern-dependent noise-predictive detector detecting a single track can provide about 2.9 dB SNR gain compared with that with the same joint detection method using uniform tracks at the same areal density. However, the latency increases due to the decision feedback detection. We can reduce this latency penalty by detecting three tracks at a time, but the SNR gain then reduces to 0.9 dB.
- 出版日期2016-7
