TY - JOUR
T1 - Grobner bases for lattices and an algebraic decoding algorithm
AU - Aliasgari, Malihe
AU - Sadeghi, Mohammad Reza
AU - Panario, Daniel
PY - 2013
Y1 - 2013
N2 - In this paper we present Grobner bases for lattices given in a general form, including integer and non-integer lattices. Grdot{o}bner bases for binary linear codes were introduced by Borges-Quintana et al.. We extend their work to non-binary group block codes. Then, given a lattice Λ and its associated label code L, which is a group code, we define an ideal for L. A Grobner basis is assigned to Λ as the Grobner basis of its label code L. Since the associated label code for integer and non-integer lattices are group codes, the assigned Grobner bases can be obtained for both cases. Using this Grobner basis an algebraic decoding algorithm is introduced. We provide an example of the decoding method for a lower dimension lattice. We explain that the complexity of this decoding method depends on the division algorithm and show this decoding method has polynomial time complexity. Experiments for some versions of root lattices (E7 and E8) show that for low SNR the performance of these lattices is near to the lower bounds given in.
AB - In this paper we present Grobner bases for lattices given in a general form, including integer and non-integer lattices. Grdot{o}bner bases for binary linear codes were introduced by Borges-Quintana et al.. We extend their work to non-binary group block codes. Then, given a lattice Λ and its associated label code L, which is a group code, we define an ideal for L. A Grobner basis is assigned to Λ as the Grobner basis of its label code L. Since the associated label code for integer and non-integer lattices are group codes, the assigned Grobner bases can be obtained for both cases. Using this Grobner basis an algebraic decoding algorithm is introduced. We provide an example of the decoding method for a lower dimension lattice. We explain that the complexity of this decoding method depends on the division algorithm and show this decoding method has polynomial time complexity. Experiments for some versions of root lattices (E7 and E8) show that for low SNR the performance of these lattices is near to the lower bounds given in.
KW - division algorithm
KW - Grobner bases
KW - label code
KW - lattices
UR - http://www.scopus.com/inward/record.url?scp=84877875546&partnerID=8YFLogxK
U2 - 10.1109/TCOMM.2013.13.120317
DO - 10.1109/TCOMM.2013.13.120317
M3 - Article
AN - SCOPUS:84877875546
SN - 0090-6778
VL - 61
SP - 1222
EP - 1230
JO - IEEE Transactions on Communications
JF - IEEE Transactions on Communications
IS - 4
M1 - 6466341
ER -