Arbitrary Distribution Random Variable Generator for Channel Emulators

Channel emulation is a fundamental part in designing, testing and validation of wireless communication systems. It brings the possibility of modeling certain channel characteristics without the necessity of performing field tests. In standards, channel models are specified in terms of several power spectral distributions that require the generation of random variables with predefined statistics. However, current emulator approaches consider different generators for each specific distribution. In this work it is presented an efficient hardware architecture for arbitrary random variable generation that can be used for all the distributions of a wireless channel emulator. The proposed architecture is based on the inversion method implemented via piecewise polynomial approximation. The architecture was validated under the WiMAX standardas the parameters generator of the channel´s temporal variations. Implementation results show that this single module can produce all random parameters for current communication standards.