f{sub B} and f{sub B{sub s}} using NRQCD

The authors present results from a calculation of f{sub B} and f{sub B{sub s}} in the quenched approximation. The heavy quark is simulated using an O(1/M{sup 2}) improved NRQCD action, and the tadpole improved clover action is used for the light quarks. The axial current includes the O(1/M) corrections in one-loop perturbation theory and the O(1/M{sup 2}) ones at tree-level. The authors find f{sub B}=152(11)({sup +8}{sub {minus}12})(9)(6) MeV, f{sub B{sub s}}=181(8)({sup +7}{sub {minus}0})(11)(7)({sup +7}{sub {minus}0}) and f{sub B{sub s}}/f{sub B}=1.20(4)({sup +4}{sub {minus}0}), where, in each case, the first error is statistical and the rest are various systematics