Pricing-Based Load Control of M2M Traffic for the LTE-A Random Access Channel

Existing cellular infrastructures have to be revisited for emerging machine-to-machine (M2M) traffic as semi-synchronized M2M arrivals may create a significant congestion resulting in a high access delay. In such a case, there is a strong need for service differentiation, since the delay requirements of IoT applications may vary greatly from delay-tolerant metering applications to security applications with stringent requirements. This problem has been receiving significant interest from the research community in the context of the LTE-A random access channel. Most studies, however, consider load control schemes with few fixed service classes, which can provide limited service differentiation. We propose an alternative scheme where the load is controlled by the price announced by the base station. The proposed method controls the load effectively and provides negligible delay for messages with the highest priority. It also enables low-cost wireless access to delay-tolerant messages by generating most of the revenue from high-priority messages. We derive pricing-based load control schemes for throughput and revenue maximization, and present the relationship between delay, revenue, and cost in both schemes. Our results suggest that dynamic pricing is a promising solution for major problems associated with cellular M2M traffic.