Variable-to-fixed length codes provide better large deviations performance than fixed-to-variable length codes

It is proved that for finite-alphabet, finite-state unifilar sources, variable-to-fixed length codes provide better large deviations performance of the empirical compression ratio, than fixed-to-variable length codes. It is shown how to construct a universal variable-to-fixed length code that achieves the optimal performance. >

[1]  Frederick Jelinek,et al.  Buffer overflow in variable length coding of fixed rate sources , 1968, IEEE Trans. Inf. Theory.

[2]  Zeev Waksman,et al.  Fused trees and some new approaches to source coding , 1988, IEEE Trans. Inf. Theory.

[3]  Giuseppe Longo,et al.  The error exponent for the noiseless encoding of finite ergodic Markov sources , 1981, IEEE Trans. Inf. Theory.

[4]  Jacob Ziv,et al.  Variable-to-fixed length codes are better than fixed-to-variable length codes for Markov sources , 1990, IEEE Trans. Inf. Theory.

[5]  Frederick Jelinek,et al.  On variable-length-to-block coding , 1972, IEEE Trans. Inf. Theory.

[6]  Katalin Marton,et al.  Error exponent for source coding with a fidelity criterion , 1974, IEEE Trans. Inf. Theory.

[7]  Abraham Lempel,et al.  Compression of individual sequences via variable-rate coding , 1978, IEEE Trans. Inf. Theory.

[8]  Neri Merhav,et al.  Universal coding with minimum probability of codeword length overflow , 1991, IEEE Trans. Inf. Theory.

[9]  Raphail E. Krichevsky,et al.  The performance of universal encoding , 1981, IEEE Trans. Inf. Theory.

[10]  Frederick Jelinek,et al.  Variable-length encoding of fixed-rate Markov sources for fixed-rate channels , 1974, IEEE Trans. Inf. Theory.

[11]  D. A. Bell,et al.  Information Theory and Reliable Communication , 1969 .

[12]  Frans M. J. Willems,et al.  Variable to fixed-length codes for Markov sources , 1987, IEEE Trans. Inf. Theory.

[13]  John B. Shoven,et al.  I , Edinburgh Medical and Surgical Journal.