1: \begin{definition}\label{def-ext} In the following, we list several definitions (see for instance \cite{Wirth}).

2: \begin{description}

3: \label{bnorm}

4: \item[-] A norm on $\mathbb{R}^n$ satisfying Conditions $1.$ 

5: and~$2.$ of Theorem~\ref{th1} is called 

6: a \textit{Barabanov norm}. Given such a norm $v(\cdot)$ we denote by $S:=\{x\in\mathbb{R}^n:v(x)=1\}$ the corresponding Barabanov unit sphere.

7: \item[-]  Given a Barabanov norm $v(\cdot)$ a solution $x(\cdot)$ of~\eqref{sys0} is said to be $v$\textit{-extremal} (or simply \textit{extremal} whenever the choice of the Barabanov norm is clear) if $v(x(t))=v(x(0))$ for every $t\geq0$.

8: \item[-] For a norm $w(\cdot)$ on $\mathbb{R}^n$ and  $x\in\mathbb{R}^n$, we use $\partial w(x)$ to denote the sub-differential of $w(\cdot)$ at $x$, that is the set of $l\in\mathbb{R}^n$ such that $l^T x=w(x)$ and $l^T y\leq w(y)$ for every $y\in\mathbb{R}^n$. {This is equivalent to saying that $l^T (y-x)\leq w(y)-w(x)$ for every $y\in\mathbb{R}^n$.}

9: \item[-] We define the $\omega$-limit set of a trajectory $x(\cdot)$ as follows:

10: $$

11: \omega(x(\cdot)):=\{\bar x\in\mathbb{R}^n:\exists (t_n)_{n\geq1}~\text{such that}~ t_n\to+\infty~\text{and}~x(t_n)\to\bar x~\text{as}~n\to+\infty\}.

12: $$

13: \end{description}

14: \end{definition}

15: