Our society is undergoing a wireless revolution in which there is a rapidly increasing demand for personal communication services (PCS). Several alternate technologies have been proposed to meet this demand, including a 1.25 MHz CDMA system. It is important to provide system designers with tools to accurately predict system performance. Analytical techniques are useful for providing general information on system performance, but may not fully capture the complexities of the mobile radio channel needed to accurately predict the performance of complex system designs. This thesis presents a novel simulation technique in which a CDMA system is modeled by software, but the channel impulse response model is derived from actual measurements. This thesis describes a bit error simulation tool for a 1.25 MHz bandwidth CDMA system. The simulator incorporates measurement-based channel impulse response data with the simulated CDMA system. Bit error rate for COMA is simulated under a variety of actual channel conditions, including heavy urban, urban, suburban and rural environments. Simulation results for both 915 MHz and 1920 MHz are presented. Vertical and circular antenna polarization are compared. In the simulation, a RAKE receiver is implemented, the advantage of this RAKE receiver implementation is demonstrated by comparison to a single correlation receiver. Results for outage probability are also presented. / Master of Science