The number of Machine-to-Machine (M2M) applications is rapidly increasing in cellular communication systems. In order to ensure a maximum system capacity, the impact of this special kind of traffic on common Human-to-Human (H2H) communication needs to be analyzed. In this paper, a system model for performance evaluation of cellular networks like Long Term Evolution (LTE) in the presence of M2M communication and under different Quality of Service (QoS) constraints is presented. By means of a Markovian model, which is parameterized by laboratory measurements and ray tracing simulations, an estimation of the behavior of LTE for different traffic characteristics is shown. We present blocking probabilities for an LTE network with heterogeneous M2M and H2H traffic and compare different transmission strategies for M2M communication to minimize the impact on human users. The results show that particularly a large number of devices with a low data rate influences the utilization of an LTE cell very negatively.