VLSI architectures for computing higher-order cumulants have gained co
nsiderable attention recently, as a consequence of the increasingly ev
ident importance of using cumulant functions as an effective signal pr
ocessing tool. This paper presents a new approach for designing linear
systolic arrays for computing third-order cumulants. First, the estim
ation of third-order cumulants is formulated elegantly as a sequence o
f matrix multiplication operations. Then a special structure systolic
array for this matrix multiplication is developed. The resulting struc
ture is an efficient unidirectional array which not only is adequate f
or VLSI implementations but also is suitable for supporting fault-tole
rance techniques.