Traffic Models for Machine Type Communications

Machine-to-machine (M2M) or Machine-type Communication (MTC) is expected to significantly increase in future wireless networks. It exhibits considerably different traffic patterns than human-type communication, thus, claims for new traffic models and simulation scenarios. The challenge in designing such models is not only to accurately capture the behavior of single MTC devices but also to handle their enormous amount (e.g., up to 30 000 devices per cell) and their coordinated behavior. Source traffic models (i.e., each device is modeled as autonomous entity) are generally desirable for their precision and flexibility. However, their complexity is in general growing quadratically with the number of devices. Aggregated traffic models (i.e., all device are summarized to one stream) are far less precise but their complexity is mainly independent of the number of devices. In this work we propose an approach which is combining the advantages of both modeling paradigms, namely, the Coupled Markov Modulated Poisson Processes (CMMPP) framework. It demonstrates the feasibility of source traffic modeling for MTC, being enabled by only linearly growing complexity. Compared to aggregated MTC traffic models, such as proposed by 3GPP TR37.868, CMMPP allows for enhanced accuracy and flexibility at the cost of moderate computational complexity.