an improvement to the gilbert varshamov bound for
play

An improvement to the Gilbert-Varshamov bound for permutation codes - PowerPoint PPT Presentation

An improvement to the Gilbert-Varshamov bound for permutation codes An improvement to the Gilbert-Varshamov bound for permutation codes Yiting Yang Department of Mathematics Tongji University Joint work with Fei Gao and Gennian Ge May 11,


  1. An improvement to the Gilbert-Varshamov bound for permutation codes An improvement to the Gilbert-Varshamov bound for permutation codes Yiting Yang Department of Mathematics Tongji University Joint work with Fei Gao and Gennian Ge May 11, 2013

  2. An improvement to the Gilbert-Varshamov bound for permutation codes Outline Outline 1 Introduction to permutation codes 2 Under Hamming distance Upper bounds Lower bounds Our improvement 3 Under Chebyshev distance Constructions Lower and upper bounds

  3. An improvement to the Gilbert-Varshamov bound for permutation codes Introduction to permutation codes Permutation codes Definition Let S n be the set of all permutations of length n. The permutation code C is just a subset of S n . The length of C is n and each permutation in C is called a codeword . Applications: Powerline communication and Flash memories P. Frankl, M. Deza, On the maximum number of permuations with givern maximal or minimal distance, J. Combin. Theory Ser. A 22 (3) (1977), 352-360. N. Pavlidou, A. J. H. Vinck, J. Yazdani and B. Honary, Powerline communications: State of the art and future trends, IEEE Communications Magazine, (2003), 34-40. A. Jiang, R. Mateescu, M. Schwartz, and J. Bruck, Rank modulation for flash memories, in Proc. IEEE Int. Symp. Information Theory, 2008, 1736-1740.

  4. An improvement to the Gilbert-Varshamov bound for permutation codes Introduction to permutation codes Hamming and Chebyshev metrics Definition For two distinct permutations σ, π ∈ S n , their Hamming distance d H ( σ, π ) is the number of elements that they differ. Definition Let π = π 1 π 2 . . . , π n , σ = σ 1 σ 2 . . . , σ n ∈ S n . The Chebyshev distance between π and σ is d C ( π, σ ) = max {| π j − σ j || 1 ≤ j ≤ n } . P. Diaconis, Group Representations in probability and Statistics, Hayward, CA: Inst. Math. Statist., 1988. T. Kløve, T. Lin, S. Tsai, and W. Tzeng, Permutation Arrays Under the Chebyshev Distance, IEEE Tran. Inform. Theory, 56 (6), 2611-2617 (2010).

  5. An improvement to the Gilbert-Varshamov bound for permutation codes Introduction to permutation codes Permutation code of minimum distance d Example Let σ = 23451 and π = 12543 . Then d H ( σ, π ) = 5 and d C ( σ, π ) = 2 . We say a permutation code C has minimum Hamming distance d if the Hamming distance of any pair of distinct permutations in C is at least d . Similarly, C is called a permutation code with minimum Chebyshev distance d if the Chebyshev distance of any pair of distinct permutations in C is at least d. They are both called a ( n, d ) -permutation code.

  6. An improvement to the Gilbert-Varshamov bound for permutation codes Introduction to permutation codes M ( n, d ) and P ( n, d ) The maximum number of codewords in a permutation code with minimum Hamming distance d is denoted by M ( n, d ) . The maximum number of codewords in a permutation code with minimum Chebyshev distance d is denoted by P ( n, d ) . Problems: • Construct large permutation codes with some fixed minimum Hamming or Chebyshev distance. • Find M ( n, d ) and P ( n, d ) , or give some good lower or upper bounds of them.

  7. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance The permutation code under Hamming distance

  8. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Constructions • Clique search • Greedy algorithm • Automorphisms • Direct constructions from permutation polynomials • Recursive construction W. Chu, C. J. Colbourn, and P. Dukes, Constructions for Permutation Codes in Powerline Commnications, Des. Codes Cryptogr. 32 (2004), 51-64. D. H. Smith and R. Montemanin, A new table of permutation codes, Des. Codes Cryptogr. 63 (2)(2012), 241-253.

  9. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Basic results on M ( n, d ) 1 M ( n, 2) = n ! ; 2 M ( n, 3) = n ! / 2 ; 3 M ( n, n ) = n ; 4 M ( n, d ) ≤ nM ( n − 1 , d ) . P. Dukes and N. Sawchuck, bounds on permutation codes of distance four, J. Algebraic Combin. 31 (1) (2010), 143-158.

  10. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Upper bounds Sphere-packing bound Definition Let D ( n, k ) ( k = 0 , 1 , . . . , n ) denote the set of all permutations in S n which are exactly at distance k from the identity. � n � Clearly, | D ( n, k ) | = D k . k Theorem n ! M ( n, d ) ≤ � . � ⌊ d − 1 2 ⌋ � n D k k =0 k P. Dukes and N. Sawchuck, bounds on permutation codes of distance four, J. Algebraic Combin. 31 (1) (2010), 143-158.

  11. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Upper bounds The upper bound for M ( n, 4) Theorem (Frankl and Deza, 1977) M ( n, 4) ≤ ( n − 1)! . Theorem (Dukes and Sawchuck, 2010) If k 2 ≤ n ≤ k 2 + k − 2 for some integer k ≥ 2 , then n ! ( n + 1) n ( n − 1) M ( n, 4) ≥ 1+ n ( n − 1) − ( n − k 2 )(( k + 1) 2 − n )(( k + 2)( k − 1) − n ) . P. Frankl, M. Deza, On the maximum number of permuations with givern maximal or minimal distance, J. Combin. Theory Ser. A 22 (3) (1977) 352-360.

  12. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Lower bounds Gilbert-Varshamov bound Theorem n ! M ( n, d ) ≥ � . � d − 1 � n k =0 D k k P. Dukes and N. Sawchuck, bounds on permutation codes of distance four, J. Algebraic Combin. 31 (1) (2010), 143-158.

  13. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement Motivation

  14. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement EKR Theorem Theorem (Erd˝ os, Ko and Rado,1961) Let k be positive integers with n > 2 k . If F is a intersecting family of k -subsets of { 1 , 2 , . . . , n } , then � n − 1 � |F| ≤ . k − 1 � n − 1 � Moreover, |F| = if and only if F is the collection of all k − 1 k -subsets that contain a fixed i ∈ { 1 , . . . , n } . P. Erd˝ os, C. Ko, R. Rado, Intersection theorems for systenms of finite sets , Quart. J. Math. Oxford Ser. 12(2) (1961) 313-320.

  15. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement Intersecting families of permutations Definition Two permutations σ, τ ∈ S n are said to k -intersect if they agree on at least k points. A set I ⊂ S n is k -intersecting if any σ, τ ∈ I k -intersect. Conjecture (Frankl and Deza, 1977) For n sufficiently large, the size of the maximum set of permutations of an n -set that are k -intersecting is ( n − k )! . P. Frankl, M. Deza, On the maximum number of permuations with givern maximal or minimal distance, J. Combin. Theory Ser. A 22 (3) (1977) 352-360.

  16. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement The proof for k = 1 Theorem (Cameron and Ku,2003, Godsil and Meagher,2009) Let n ≥ 2 . If F ⊂ S n is an intersecting family of permutations. Then |F| ≤ ( n − 1)! , with equality holds if and only if F is a coset of a stabilizer of a point. P. J. Cameron and C. Y. Ku, Intersection families of the permutations, European J. Combin. 24 (2003) 881-890. C. Godsil and K. Meagher, A new proof of the Erd˝ os-Ko-Rado Theorem for intersecting families of permutations, European J. Combin. 30 (2009) 404-414.

  17. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement Proof of Frankl and Deza’s conjecture • k-coset: T i 1 → j 1 ,...,i k → j k = { σ ∈ S n , σ ( i 1 ) = j 1 , . . . , σi k = j k } Theorem (Ellis et al., 2011) For any fixed k and sufficiently large n , if I ⊂ S n is k -intersecting then | I | ≤ ( n − k )! , with equality if and only if I is a k -coset. D. Ellis, E. Friedgut and H. Pilpel, Intersecting Families of Permutations, J. Amer. Math. Soc. 24 (3) (2011) 649-682.

  18. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement Relation between intersecting families and permutation codes Observation: A t -intersecting family of S n is a permutation code with maximum Hamming distance at most n − t. Theorem Let C be a permutation code with maximum Hamming distance n − t . Then | C | ≤ ( n − t )! .

  19. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement Graph theory model Definition A subgraph of a graph is called a clique if any two of its vertices are adjacent. An independent set is a subgraph in which no two vertices are adjacent. We define a Cayley graph Γ( n, d ) := Γ( S n , S ( n, d − 1)) , where S ( n, d − 1) is the set of all the permutations with more than n − d fixed points.

  20. An improvement to the Gilbert-Varshamov bound for permutation codes Under Hamming distance Our improvement Graph theory model By the definition, Γ( n, d ) is a regular graph of degree which equals the size of the generating set, i.e., d − 1 � n � � ∆( n, d ) = | S ( n, d − 1) | = D k . k k =1 The codewords of an ( n, d ) permutation code are vertices of an independent set in Γ( n, d ) . Conversely, any independent set in Γ( n, d ) is an ( n, d ) -permutation code.

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend