Codes for Symbol-Pair Read Channels Yuval Cassuto EPFL – ALGO Formerly: Hitachi GST Research November 3, 2010 IPG Seminar.

Slides:



Advertisements
Similar presentations
Ulams Game and Universal Communications Using Feedback Ofer Shayevitz June 2006.
Advertisements

Cyclic Code.
Applied Algorithmics - week7
Error Control Code.
Locally Decodable Codes from Nice Subsets of Finite Fields and Prime Factors of Mersenne Numbers Kiran Kedlaya Sergey Yekhanin MIT Microsoft Research.
Information and Coding Theory
Information Theory Introduction to Channel Coding Jalal Al Roumy.
Achilleas Anastasopoulos (joint work with Lihua Weng and Sandeep Pradhan) April A Framework for Heterogeneous Quality-of-Service Guarantees in.
Cellular Communications
Lihua Weng Dept. of EECS, Univ. of Michigan Error Exponent Regions for Multi-User Channels.
Codes for Deletion and Insertion Channels with Segmented Errors Zhenming Liu Michael Mitzenmacher Harvard University, School of Engineering and Applied.
Asymptotic Enumerators of Protograph LDPCC Ensembles Jeremy Thorpe Joint work with Bob McEliece, Sarah Fogal.
DIGITAL COMMUNICATION Coding
Probabilistic Methods in Coding Theory: Asymmetric Covering Codes Joshua N. Cooper UCSD Dept. of Mathematics Robert B. Ellis Texas A&M Dept. of Mathematics.
EXPANDER GRAPHS Properties & Applications. Things to cover ! Definitions Properties Combinatorial, Spectral properties Constructions “Explicit” constructions.
Vladimir V. Ufimtsev Adviser: Dr. V. Rykov A Mathematical Theory of Communication C.E. Shannon Main result: Entropy function - average value of information.
Theta Function Lecture 24: Apr 18. Error Detection Code Given a noisy channel, and a finite alphabet V, and certain pairs that can be confounded, the.
7/2/2015Errors1 Transmission errors are a way of life. In the digital world an error means that a bit value is flipped. An error can be isolated to a single.
Ger man Aerospace Center Gothenburg, April, 2007 Coding Schemes for Crisscross Error Patterns Simon Plass, Gerd Richter, and A.J. Han Vinck.
Channel Polarization and Polar Codes
Hamming Code Rachel Ah Chuen. Basic concepts Networks must be able to transfer data from one device to another with complete accuracy. Data can be corrupted.
exercise in the previous class (1)
Hamming Codes 11/17/04. History In the late 1940’s Richard Hamming recognized that the further evolution of computers required greater reliability, in.
Analysis of Iterative Decoding
©2003/04 Alessandro Bogliolo Background Information theory Probability theory Algorithms.
林茂昭 教授 台大電機系 個人專長 錯誤更正碼 數位通訊
DIGITAL COMMUNICATION Error - Correction A.J. Han Vinck.
Channel Coding Part 1: Block Coding
Information and Coding Theory Linear Block Codes. Basic definitions and some examples. Juris Viksna, 2015.
Exercise in the previous class p: the probability that symbols are delivered correctly C: 1 00 → → → → What is the threshold.
Information Coding in noisy channel error protection:-- improve tolerance of errors error detection: --- indicate occurrence of errors. Source.
Uncorrectable Errors of Weight Half the Minimum Distance for Binary Linear Codes Kenji Yasunaga * Toru Fujiwara + * Kwansei Gakuin University, Japan +
Threshold Phenomena and Fountain Codes Amin Shokrollahi EPFL Joint work with M. Luby, R. Karp, O. Etesami.
ERROR CONTROL CODING Basic concepts Classes of codes: Block Codes
MIMO continued and Error Correction Code. 2 by 2 MIMO Now consider we have two transmitting antennas and two receiving antennas. A simple scheme called.
Introduction to Coding Theory. p2. Outline [1] Introduction [2] Basic assumptions [3] Correcting and detecting error patterns [4] Information rate [5]
Wireless Mobile Communication and Transmission Lab. Theory and Technology of Error Control Coding Chapter 5 Turbo Code.
Error Control Code. Widely used in many areas, like communications, DVD, data storage… In communications, because of noise, you can never be sure that.
Coding Theory. 2 Communication System Channel encoder Source encoder Modulator Demodulator Channel Voice Image Data CRC encoder Interleaver Deinterleaver.
§6 Linear Codes § 6.1 Classification of error control system § 6.2 Channel coding conception § 6.3 The generator and parity-check matrices § 6.5 Hamming.
DIGITAL COMMUNICATIONS Linear Block Codes
Linear codes of good error control performance Tsonka Baicheva Institute of Mathematics and Informatics Bulgarian Academy of Sciences Bulgaria.
1 Private codes or Succinct random codes that are (almost) perfect Michael Langberg California Institute of Technology.
On Coding for Real-Time Streaming under Packet Erasures Derek Leong *#, Asma Qureshi *, and Tracey Ho * * California Institute of Technology, Pasadena,
Low Density Parity Check codes
Basic Concepts of Encoding Codes and Error Correction 1.
Perfect and Related Codes
Some Computation Problems in Coding Theory
Error Detection and Correction
1 Asymptotically good binary code with efficient encoding & Justesen code Tomer Levinboim Error Correcting Codes Seminar (2008)
Turbo Codes. 2 A Need for Better Codes Designing a channel code is always a tradeoff between energy efficiency and bandwidth efficiency. Lower rate Codes.
Lower bounds on data stream computations Seminar in Communication Complexity By Michael Umansky Instructor: Ronitt Rubinfeld.
Raptor Codes Amin Shokrollahi EPFL. BEC(p 1 ) BEC(p 2 ) BEC(p 3 ) BEC(p 4 ) BEC(p 5 ) BEC(p 6 ) Communication on Multiple Unknown Channels.
1 Reliability-Based SD Decoding Not applicable to only graph-based codes May even help with some algebraic structure SD alternative to trellis decoding.
Error Detecting and Error Correcting Codes
Exercise in the previous class (1) Define (one of) (15, 11) Hamming code: construct a parity check matrix, and determine the corresponding generator matrix.
1 Code design: Computer search Low rate: Represent code by its generator matrix Find one representative for each equivalence class of codes Permutation.
Institute for Experimental Mathematics Ellernstrasse Essen - Germany DATA COMMUNICATION introduction A.J. Han Vinck May 10, 2003.
Joint Decoding on the OR Channel Communication System Laboratory UCLA Graduate School of Engineering - Electrical Engineering Program Communication Systems.
The Viterbi Decoding Algorithm
Data Link Layer Objective: to achieve reliable and efficient communication between 2 adjacent machines Data link layer design issues services provided.
Error-Correcting Codes:
k-center Clustering under Perturbation Resilience
Richard Anderson Lecture 25 NP-Completeness
Data Link Layer Objective: to achieve reliable and efficient communication between 2 adjacent machines Data link layer design issues services provided.
Data Link Layer Objective: to achieve reliable and efficient communication between 2 adjacent machines Data link layer design issues services provided.
Miguel Griot, Andres I. Vila Casado, and Richard D. Wesel
Compute-and-Forward Can Buy Secrecy Cheap
Chapter 10 Error Detection and Correction
Lihua Weng Dept. of EECS, Univ. of Michigan
Presentation transcript:

Codes for Symbol-Pair Read Channels Yuval Cassuto EPFL – ALGO Formerly: Hitachi GST Research November 3, 2010 IPG Seminar

Data-Storage Systems Physical Media Read/Write Interface Error Correcting Codes User Data User Data

Error Correcting Codes Physical Media Read/Write Interface The Key Objective

Physical Media Read/Write Interface The Key Objective Error Correcting Codes “The redundancy must be introduced in the proper way to combat the particular noise structure involved” C.E Shannon, “A Mathematical Theory of Communication”, 1948

The Key Challenge Error Correcting Codes Physical Media Read/Write Interface Combinatorial Statistical Complex Physical Mechanisms Error-Control Guarantees Multiple Reliability Issues Simple Design Objectives

The Key Strategy Error Correcting Codes Physical Media Read/Write Interface ErrorModel

Conventional Magnetic Media W R bit

Patterned Magnetic Media W R bit1.Reversal stability 2.Better SNR

Background and Motivation Read resolution < Write resolution Symbol Read Head ResponseHard

Pair Read Simpler Background and Motivation Read resolution < Write resolution Head Response L R

Symbol Read Channels Channel Codeword Received word Error Models: Random, burst, symmetric, asymmetric, hard/soft decision

Symbol-Pair Read Channels Channel Codeword Received word

Example [ (2,3) (3,5) (5,1) (1,4) (4,2) ] 0 pair-errors Design objective: t pair-errors [ (2,3) (3,4) (5,1) (2,0) (4,2) ] 2 pair-errors Wrap around

Related Topics Multiple burst errors ISI channels

Notation Stored vector Received pair-vector No error if

Notation Pair error Pair distance (metric) Consistent pair-vector if or (or both)

Some Facts Proposition: Pair-distance relation to Hamming- distance Proposition: A code can correct t pair-errors if and only if the minimum pair-distance is ≥2t+1 Theorem: With consistent-only channel outputs –Sufficient: ≥2t+1 –Necessary: ≥2t

A Closer Look on D p L is the number of runs of differing indices

Construction Idea Good Bad Interleaving

Interleave vs. Direct Is interleaving optimal? Not optimal! Interleaved Hamming code (optimal)Direct Corrects 2 pair errorsCorrects 3 pair errors

Construction: Cyclic Codes Theorem: A cyclic code whose g(x) has at least d H roots satisfies –Proof: Dual of BCH bound

Proof Sketch BCH Bound [BC 1960, H 1959]: g(x) with many consecutive roots codewords with few zeros BCH Dual-Transposed: g(x) with many roots codewords without many consecutive zeros Non-zeros occupy multiple consecutive subsets (L>1) BCH Dual: codewords with many consecutive zeros g(x) with few roots

Applications of L>1 Theorem: Cyclic Hamming codes have d p ≥5 –Note: not true for non-cyclic Hamming codes –Corollary: cyclic Hamming codes are pair-perfect (with d p =5) d p =4 Codeword:

Stronger Lower Bounds on L Theorem: A cyclic code whose g(x) has at least m roots and d H ≤ min(2m-n+2,m-1) satisfies –Main proof tool: Dual of Hartmann-Tzeng bound

Decoding by Reduction Received pair-vector Symbol Error/Erasure Decoder

Decoding by Reduction Is reduction optimal? NO, example: Theorem: Optimal for interleaved codes Symbol Error/Erasure Decoder D p =1 D p =2

Bounds on Code Sizes Hamming SpherePair Sphere Symbol vector Non-consistent pair-vector

Enumerate Pair-Spheres How many symbol-words at pair-distance D p ? Pair Sphere

Count L-Subsets is the number of subsets of {0,…,n-1} of size l that fall into L runs (with wrap around). Definition: Key:

L-Run Layouts Non all-around All-around Spacer of length 1 or more Run of length 1 or more Blacks sum to l Whites sum to n - l

Closed Form Count Non all-around All-around

Pair-Sphere and Pair-Ball Pair-Sphere: Pair-Ball: Sum to h

Closed Form Bounds A code corrects t pair-errors only if There exists a code with minimum pair- distance d if Pair-sphere packing (upper) bound: Pair Gilbert-Varshamov (lower) bound:

Asymptotic Bounds Combinatorial bounds are exact for any parameter set [n,d,q] Offer little insight on asymptotic behavior Need succinct (but still tight) bounds on pair-ball volumes

Pair Gilbert-Varshamov

Asymptotic Correctability: Pair vs. Symbol dpdp d H +12d H Asymptotic pair advantage VanishingDouble

Pair vs. Hamming GV Bound Pair-error codes: Yes Symbol-error codes: (Probably) No

Graph Theoretic View Complete directed graph (with self loops) Codes with large pair-distance = Closed walks on graph with small edge overlap

Conclusion Initial study of symbol-pair coding Can correct significantly more pair-errors than symbol-errors Open problems in –Algebraic coding theory (e.g. cyclic codes) –Codes on graphs/trellises –Coding+detection (soft decoding)

To Read More [YC,Blaum], ISIT 2010 [YC,Blaum], IT-Trans Submission