On uncertain suboptimum estimation with hypothesis switching

The problem model considered here is y(k+1) = A(k)y(k) + ??G(k)w(k) (1) x(k) = ??y(k) (2) z(k) = ??C(k)x(k) + v(k) (3) where x(k) is a Gauss-Markov n vector, the state of the system, w(k), the plant noise vector, is a white gaussian m vector with known mean and variance, v(k) is the observation noise vector and is zero mean gaussian r vector, z(k) is the r vector of observation, A(k), G(k), and C(k) are known coefficient matrices, ??, ??, and ?? are the sources of uncertainty for this model.