FIR (Finite Impulse Response) filters with linear phase response are highly required in applications where the linearity of the phase is crucial. In this paper, we present a novel evolutionary based technique for designing FIR filters. Typically, the required filter has a given set of specifications to be met. The specifications include the cut off frequency, band-stop region, band-pass region, ripple factors, and most importantly the linearity of phase transfer function. Medical applications require high linearity in the filter phase function to prevent undesired distortions in the detected signals. The compromise between ending with a tight magnitude transfer function and a linear phase function is crucial here, as a cost function is used to measure performance. The evolutionary method we are using is a modified version of the Genetic Algorithm (GA) we called it messy GA (MGA). It is an optimization algorithm with high capabilities to span the space of filter parameters. It is used to optimize a complex objective function that reflects constraints and design requirements of the FIR filter. Comparison between MGA designed filter and a standard filter design methods is implemented. Testing for both filters is done using different noisy artificial ECG signals. The ECG signals are made using combinations of Hermite functions .