Consider system
\left\{ {\begin{array}{*{20}{c}}
{{{\dot x}_1} = {\varphi _1}(.) + {\rho _1}{x_{l + 1}},} \\
{{{\dot x}_m} = {\varphi _m}(.) + {\rho _m}{x_n},} \\
{{{\dot x}_{m + 1}} = {\varphi _{m + 1}}(.) + {\mu _1},} \\
{{{\dot x}_n} = {\varphi _n}(.) + {\mu _1},}
\end{array}} \right. where x1, …, and xn is the state of the system, u1, …, and ul are controls, n/l is not an integer, and l ≥
... [Show full abstract] 2. It is supposed that only outputs x1, …, and xl are measurable, (l > n) and ϕi(·) are non-anticipating arbitrary functionals, and 0 < ρ–≤ ρi (t, x1, …, and xl) ≤ ρ+. Using the backstepping method, we construct the square Lyapunov function and stabilize the control for the global exponential stability of the closed loop system. The stabilization by means of synchronous modulators with a sufficiently high impulsion frequency is considered as well.