In this paper, we analyze the performance of a cooperative cognitive radio network when secondary users (SUs) are operating in different environments. The variable operating conditions result in variable SNRs at each SU. Besides, closed forms of detection and false alarm probabilities are computed assuming that cognitive terminals are performing spectrum sensing using energy detection technique. Simulation results show that the system performance are significantly improved by increasing the number of SUs. An optimum local false alarm probability is also computed numerically. This optimum value maximizes the fusion center reliability for a given interference level. An analysis of the effect of the false alarm rate at the SUs on the overall system functioning is given and is illustrated by simulation.