The paper is concerned with the possibility of using a decomposition techni
que to solve fixed channel assignment (FCA) problems. The main idea is firs
t to transform and then to decompose the original large FCA problem into a
number of smaller-size connected integer linear programming (ILP) sub-probl
ems. A sequential branch-and-bound algorithm is then used to solve optimall
y and efficiently the sub-problems. The computational results obtained when
the technique is applied to a number of different benchmark problems confi
rm that the proposed technique is computationally efficient in solving larg
e-scale FCA problems and that it is capable of significantly reducing the c
omputation time involved. The technique also makes it possible in some case
s to solve large-scale FCA problems.