A number of numerical schemes exists for Monte Carlo Simulation of stochas-tic differential equations with refiection driven by Wiener processes. In contrast, reflectedSDEs driven by purely discontinuous processes, or Poisson random measures, respectivelyare weakly investigated. In this paper, we suggest an algorithm for simulating SDEs driv-en by certain Poisson random measures. The idea of this approach is to approximate theunderlying point process arising by cutting the diffusion at the level set of the boundaryand parametrizing these excursions by the local time. This method is easily to implementon Computers.