A New Class of Balanced Near-Perfect Nonlinear Mappings and Its Application to Sequence Design

作者：Chung Jin Ho^*; Yang Kyeongcheol

来源：IEEE Transactions on Information Theory, 2013, 59(2): 1090-1097.

DOI：10.1109/TIT.2012.2224146

• Balancedness
• frequency-hopping sequences (FHS)
• Hamming correlation
• nonlinear mappings
• periodic correlation

A mapping from Z(N) to Z(M) can be directly applied for the design of a sequence of period with alphabet size, where Z(N) denotes the ring of integers modulo. The nonlinearity of such a mapping is closely related to the autocorrelation of the corresponding sequence. When is a divisor of, the sequence corresponding to a perfect nonlinear mapping has perfect autocorrelation, but it is not balanced. In this paper, we study balanced near-perfect nonlinear (NPN) mappings applicable for the design of sequence sets with low correlation. We first construct a new class of balanced NPN mappings from Z(p2-p) to Z(p) for an odd prime p. We then present a general method to construct a frequency-hopping sequence (FHS) set from a nonlinear mapping. By applying it to the new class, we obtain a new optimal FHS set of period p(2) - p with respect to the Peng-Fan bound, whose FHSs are balanced and optimal with respect to the Lempel-Greenberger bound. Moreover, we construct a low-correlation sequence set with size p, period p(2) - p, and maximum correlation magnitude p from the new class of balanced NPN mappings, which is asymptotically optimal with respect to the Welch bound.

• 出版日期2013-2