New combinatorial designs and their applications to authentication codes and secret sharing schemes

This paper introduces three new types of combinatorial designs, which we call external difference families (EDF), external BIBDs (EBIBD) and splitting BIBDs. An EDF is a special type of EBIBD, so existence of an EDF implies existence of an EBIBD. We construct optimal splitting A-codes by using EDF. Then we give a new bound on the number of shares required in robust secret sharing schemes (i.e., schemes secure against cheaters). EDF can be used to construct robust secret sharing schemes that are optimal with respect to the new bound. We also prove a weak converse, showing that if there exists an optimal secret sharing scheme, then there exists an EBIBD. Finally, we derive a Fisher-type inequality for splitting BIBDs. We also prove a weak equivalence between splitting BIBDs and splitting A-codes. Further, it is shown that an EDF implies a splitting BIBD.