We present a formula to calculate the bit error rate (BER) of maximum-distance-separable (MDS) codes based on the weight distributions of the codes. Because of the exact number of a certain weight of the codes, the probabilities of decoding error and decoding failure can be achieved for an incomplete decoder. The proposed formula is built by summing up all the probabilities of decoding error and decoding failure properly. Two MDS codes, RS (7,5) and RS (63,45) codes, with a 8PSK modulation scheme in the AWGN channel are chosen to verify the formula. Compared with the conventional upper bound formula, our proposed formula is more consistent with the simulated results, no matter short codes or long codes. The formula can be used widely to calculate the accurate BER of MDS codes.
[1]
J. Macwilliams.
A theorem on the distribution of weights in a systematic code
,
1963
.
[2]
G. David Forney,et al.
Generalized minimum distance decoding
,
1966,
IEEE Trans. Inf. Theory.
[3]
Tadao Kasami,et al.
Some Remarks on BCH Bounds and Minimum Weights of Binary Primitive BCH Codes
,
1969,
IEEE Trans. Inf. Theory.
[4]
Z. M. Huntoon,et al.
On the computation of the probability of post-decoding error events for block codes (Corresp.)
,
1977,
IEEE Trans. Inf. Theory.
[5]
I. Reed,et al.
Polynomial Codes Over Certain Finite Fields
,
1960
.
[6]
R. Blahut.
Theory and practice of error control codes
,
1983
.