A concatenated LDPC-marker code for channels with correlated insertion and deletion errors in bit-patterned media...

Area of Science:

• Data storage systems
• Information theory
• Error correction codes

Background:

• Existing error correction codes struggle with correlated synchronization errors common in advanced storage like bit-patterned media recording (BPMR).
• A need exists for models and codes that account for data-dependent errors in BPMR systems.

Purpose of the Study:

• To propose a probabilistic channel model for correlated insertion and deletion (CID) errors relevant to BPMR.
• To evaluate the performance and complexity of a concatenated LDPC-marker code over this CID channel.
• To adapt decoding algorithms for improved efficiency on the CID channel.

Main Methods:

• Development of a novel probabilistic channel model for correlated insertion and deletion (CID) errors.
• Investigation of a concatenated Low-Density Parity-Check (LDPC)-marker code for error correction.
• Modification of the forward-backward decoding algorithm using a two-dimensional state transition diagram for CID channels.

Main Results:

• The concatenated LDPC-marker code demonstrates superior error performance compared to conventional methods for random errors.
• Bit error rate (BER) performance is improved by over 50% on average.
• Decoding time is reduced by approximately 35% using the modified decoding algorithm.

Conclusions:

• The proposed CID channel model accurately captures synchronization error dependencies in BPMR.
• The concatenated LDPC-marker code offers a robust solution for error correction in BPMR systems.
• The adapted decoding algorithm enhances efficiency, making the system more practical.