Precoded Polar Codes With High Error-Correction Capability
summary
Claims low-complexity recipes to design precoded polar codes that hit both high error-correction capability and low SCL decoding complexity, with concrete results: joint optimization of frozen set plus frozen-bit expressions for short lengths (N=32,64) yields weight distributions matching or beating extended BCH and reduces required list size to approach ML; for long codes (N≈1k–4k) frozen sets are parameterized by three integers, found by evaluating ~20–100 candidates, and outperform state‑of‑the‑art codes under SCL with list sizes like 32.
AI-written summary · about
chapters (15)
- 0:00Introduction
- 0:31Polar codes overview
- 2:25Precoded polar codes
- 3:29Mathematical structure
- 6:46Polar code design
- 7:50Design criteria SCL
- 10:48Joint optimization short
- 12:20Optimization long codes
- 13:53Frozen set optimization
- 16:59Search space reduction
- 19:20Parameterizing frozen sets with three integers
- 21:35Design criteria summary
- 23:28Q&A LDPC comparison
- 25:14Q&A adaptive list
- 27:34Q&A distance CRC
related talks
15mLESS is More for I/O-Efficient Repairs in Erasure-Coded Storage
10mRevisiting Network Coding for Warm Blob Storage
44mTutorial on List Decodable Coding Schemes and Interactive Communication in Networks (Part 2 of 2)
57mCoding for Interactive Communication
28mCommunication Avoiding Algorithms for Linear Algebra and Beyond
11mExploiting Combined Locality for Wide-Stripe Erasure Coding in Distributed Storage
27mList Recovery and Local Decoding of Expander Codes
31mMulti-version Coding and Shared Memory Emulation