In technologies of communication most of devices emit radiated and conducted emission unsafely. Thus every device is tested according the standards and determined whether they are safe. In order to analyze and reduce a device noise, the noise must be separated to components. In this paper, some methods were researched to learn how to separate the noise and two theories were dwelled upon. These theories' patterns and formulas were shown and their advantages and disadvantages were explained. Then the theory that was truer was selected for studies. A noise separator sample was simulated on Microwave Office. Its scattering parameters were examined. It is determined that S31 and S41 are stable at -6 dB and S11 and S21 is approximately -50 and -80 dB. After simulation, the noise separator pattern was laid out on board. The separator's difference from the other designs that used the theory is elliptic and shorter ways. As a result of this, we aimed to provide reduced transmission line characteristic impedance and ideal voltage standing wave ratio. Measurement of the noise separator's scattering parameter was performed with Agilent N9926A vector network analyzer and it was determined that S11 parameter decreased as far as 10 dB at some points.