Optimized Generalized LDPC Convolutional Codes

Area of Science:

• Coding Theory
• Information Theory
• Digital Communications

Background:

• Generalized LDPC convolutional codes (GLDPC-CCs) are crucial for reliable data transmission.
• Existing encoding and decoding schemes for GLDPC-CCs have limitations in performance and complexity.
• Optical communication systems demand highly efficient error correction codes.

Purpose of the Study:

• To propose novel optimized encoding and decoding schemes for GLDPC-CCs.
• To enhance the performance and reduce the complexity of GLDPC-CCs.
• To validate the effectiveness of the proposed schemes for optical communication applications.

Main Methods:

• Proposed a flexible doping method replacing single parity check (SPC) nodes with generalized check (GC) nodes (e.g., BCH codes).
• Introduced a hybrid layered normalized min-sum (HLNMS) decoding algorithm combining LNMS for SPC nodes and Chase-Pyndiah for GC nodes.
• Developed an adaptive weight factor for GC nodes and an early stop decoding strategy based on mutual information.

Main Results:

• The proposed flexible doping method allows for selecting different types of BCH codes as GC nodes.
• The HLNMS decoding algorithm with adaptive weighting and early stopping improves decoding performance and reduces complexity.
• Simulation results demonstrate the superiority of the proposed GLDPC-CC schemes over prior art.

Conclusions:

• The developed encoding and decoding schemes offer significant improvements for GLDPC-CCs.
• The proposed methods enhance error correction capabilities, making GLDPC-CCs more suitable for demanding applications.
• GLDPC-CCs with the proposed schemes show great application potential in optical communication systems.