Reliability Control Framework for Random Access of Massive IoT Devices

The Internet of Things (IoT) devices with enhanced machine type communication (eMTC) technology require random access (RA) to transmit data. The success rate of data delivery in the eMTC depends on the probability of failure in the RA. Access class barring (ACB) can decrease the probability of failure in the RA procedure. However, it is hard to precisely predict the success rate of the RA with the ACB. In this paper, we aim to control the failure rate of the RA to the desired probability by designing a reliability control framework for the RA in the eMTC. The framework includes an algorithm that estimates the number of active devices in a cell from the number of undecoded preambles and the probability of preamble loss. The framework also consists of an algorithm to obtain the probability of RA failure by changing the number of preamble transmissions. In addition, the adaptive ACB factor decision algorithm adjusts the ACB factor based on observed state information. The performance of the proposed framework is evaluated using an RA simulator in the environments recommended by the third generation partnership project (3GPP). The results of the evaluation indicate that the number of preamble transmissions selected by our algorithm successfully determines the probability of RA failure. In addition, the simulation results suggest that the number of supportable devices decreases as a tradeoff for increased reliability due to decreased RA failure rate.