Exponentially Distributed Random Access in LTE-A networks

To sustain dense user equipment (UE) deployments, one of the difficult issues is to provide a proficient way to various channel access in the cellular communication networks, such as long-term evolution-Advanced (LTE-A) networks. In LTE-A, random access (RA) is the necessary process to set up the wireless connection between a UE and an evolved node B (eNB). The performance of the RA straightforwardly influences the performance of the whole system. Currently, a discrete uniform distribution (DUD) is used to avoid the potential collision in the network, while multiple UEs try to access the channel resources. However, in a DUD, every UE has an equal chance to choose similar contention preamble near to the expected value of the DUD, which may cause an increase in a collision among the UEs. In this work, we propose a method based on the existing alternatives, such as continuous exponential distribution (CED). CED distributes the random values between two bounds in a poison point process, in which random variables occur continuously and independently with a constant average rate. In this way, our proposed method can distribute the UEs in a parametric set of the probability distribution. Our proposed mechanism is named as CED RA (CED-RA), which replaces the DUD mechanism in RA with CED to show enhanced reliability and lower latency.