A set, , in a real inner product space is convex if for all and , where , it holds that .
The set of solutions to an LMI is convex.
That is, the set is a convex set, where is an LMI.
An LMI, , in the variable is an expression of the form
where and , .
Consider and , and suppose that and satisfy Lemma 1.2.
The LMI is convex, since
From Lemma 1.1, it is known that an optimization problem with a convex objective function and LMI constraints is convex.
The following is a non-exhaustive list of scalar convex objective functions involving matrix variables that can be minimized in conjunction with LMI constraints to yield a semi-definite programming (SDP) problem.
- , where , , , and .
- Special case when and , where, , and .
- Special case when 2, , and , where .
- , where , , , , , and .
- Special case when and , where , , and .
- Special case when , and , where .
- Special case when , and , where , .
- Special case when , , , and , where .
- , where and .
Relative Definition of a Matrix[edit | edit source]
The definiteness of a matrix can be found relative to another matrix.
For example,
Consider the matrices and . The matrix inequality is equivalent to or .
Knowing the relative definiteness of matrices can be useful.
For example,
If in the previous example we have and also know that , when we know that .
This follows from .
Strict and Non-strict Matrix Inequalities[edit | edit source]
A strict matrix inequality can be converted to a non-strict matrix inequality.
For example,
is implied by , where . Similarly, is implied by , where
Converting a strict matrix inequality into a non-strict matrix inequality is useful when working with LMI solvers that cannot handle strict constraints.
A useful property of LMIs is that multiple LMIs can be concatenated together to form a single LMI.
For example,
satisfying the LMIs and is equivalent to satisfying the concatenated LMI
More generally, satisfying the LMIs , is equivalent to satisfying the concatenated LMI .