ABSTRACT
Let a and b be positive integers and let M be an (a + 1)b × (a + 1)b matrix. The columns and rows of M are indexed by pairs (i, j) and (k, ) such that (i, j) is the (bi+ j)-th column and (j, ) is the (bk+ )-th row of M . Thus, the entry in the (bi+ j)-th column and (bk+ )-th row (in the usual sense) is also the element in the (i, j)-th column and (k, )-th row, and will be denoted by M(i, j, k, ). Diagonal entries in the matrix have the form M(k, , k, ).
