Side-channel attacks (SCA) have become more and more popular among hackers all over the world because they can provide access to sensitive information stored in a circuit, only by exploring physical data leakages. The most well-known SCA's are based on power consumption (e.g. Differential/Simple Power Analysis), electromagnetic radiations (e.g. Differential/Simple Electromagnetic Analysis), analyze of consumed time while performing cryptographic operations and inducing of faults (e.g. Optical Faults). Both power and electromagnetic analysis require acquiring a large number of traces by using specialized equipment. The problem appears when the environmental conditions are rough and the captured traces are noisy, making it hard to extract the signal from the noise. In this paper we focus on finding an optimal mechanism to denoise the signals captured from the electromagnetic radiations emitted by a microcontroller. In order to do so, Wavelet functions were used and results were compared depending on the threshold's levels and methods that were applied. Wavelet functions are a comprehensive tool with applications like signal denoising and data compression. All the obtained results are presented in this paper and conclusions regarding this process are drawn.