1: \begin{abstract}
2: In this paper, a concurrent learning based adaptive observer is developed
3: for a class of second-order linear time-invariant systems with uncertain
4: system matrices. The developed technique yields an exponentially convergent
5: state estimator and an exponentially convergent parameter estimator.
6: As opposed to \textit{persistent} excitation required for parameter
7: convergence in traditional adaptive methods, excitation over a finite
8: time-interval is sufficient for the developed technique to achieve
9: exponential convergence. Simulation results in both noise-free and
10: noisy environments are presented to validate the design.
11: \end{abstract}
12: