On the counting function of the lattice profile of periodic sequences (Q2465280): Difference between revisions
From MaRDI portal
Added link to MaRDI item. |
Removed claims |
||
Property / author | |||
Property / author: Fang-Wei Fu / rank | |||
Property / author | |||
Property / author: Harald Niederreiter / rank | |||
Revision as of 01:13, 10 February 2024
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | On the counting function of the lattice profile of periodic sequences |
scientific article |
Statements
On the counting function of the lattice profile of periodic sequences (English)
0 references
9 January 2008
0 references
A sequence \(\mathbf{S} = s_0,s_1,\ldots\) over a finite field \(\mathbb F_q\) is said to pass the \(t\)-dimensional \(n\)-lattice test if the vectors \(\underline{\mathbf{s}}_i - \underline{\mathbf{s}}_0\), \(i = 1,\ldots,n-t\), span \(\mathbb F_q^t\), where \[ \underline{\mathbf{s}}_i = (s_i,s_{i+1},\ldots,s_{i+t-1}),\quad 0 \leq i \leq n-t. \] The lattice profile \(T(\mathbf{S},n)\) of \(\mathbf{S}\) at \(n\) is then defined to be the greatest integer \(t\) such that \(\mathbf{S}\) passes the \(t\)-dimensional \(n\)-lattice test, and the lattice profile of \(\mathbf{S}\) is defined as \[ T(\mathbf{S}) = \sup_{n \geq 2}T(\mathbf{S},n), \] which is finite if \(\mathbf{S}\) is a periodic sequence. The authors use relations between linear complexity and the lattice profile for periodic sequences presented in \textit{H. Niederreiter} and \textit{A. Winterhof} [AAECC 13, 319--326 (2002; Zbl 1033.11038)], and \textit{G. Dorfer} and \textit{A. Winterhof} [Proceedings of MCQMC 2002, 199--211 (2004; Zbl 1043.65007)] to obtain explicit formulas for the expected value and the variance for the lattice profile of a random \(N\)-periodic sequence. Furthermore for certain period length closed formulas for the number of sequences with given lattice profile are provided.
0 references
linear complexity
0 references
discrete Fourier transform
0 references
expected value
0 references
variance
0 references