Constrained Codes for Single-Reader/Two-Track Reading Technique in Bit-Patterned Magnetic Recording Systems

Abstract

A reduction in the track width of magnetic recording systems results in a welcome increase in areal density (AD), but also in the unfortunate appearance of extreme inter-track interference (ITI) that can severely deteriorate system performance [1]. To deal with this problem, we propose to use a single-reader/two-track reading (SRTR) technique in a bit-patterned magnetic recording (BPMR) system where a single reader was used to read the desired parallel-track that was arranged as in the staggered structure [2]. Then, the received readback signal was sampled using the over-sampling technique as shown in Fig. 1. The data sequences that were obtained from the over-sampling process is then easily operated using one-dimensional (1D) equalization and detection schemes [3]. Moreover, we also present the rate-3/4 and -5/6 constrained codes for the SRTR BPMR system with the new criteria for encoding the user data bit. The error percentages of all considered data patterns i.e., 24 = 16 and 26 = 64 possible patterns for the rate-3/4 and -5/6 constrained codes, respectively, are first investigated as shown in Fig. 2. Then, the selected patterns which provide the lowest error percentage will be then given as the codeword. For instance, the first 32 patterns that give the lowest error percentage will be select as codewords for the rate-5/6 coded system as depicted in Fig. 2. Furthermore, the specified constraints at the boundaries between each codeword can be mitigated using the proposed sub-codeword which depends on its prior transmitted codeword. Besides, the decoding process can be easily operated under both the Euclidean distance and max-min based on Boolean logic functions [4] methods. The computer simulation results show that the effect of severe ITI can be mitigated through the use of our proposed coding schemes when compared with uncoded-, coded-SRTR, and conventional recording systems at the same considered user densities.