Dynamic Provisioning of Random Access Capacity in mMTC Slice Based on Beam Splitting/Merging

In massive Machine Type Communication (mMTC), an enormous number of user equipments (UEs) try to access the network in a short period. It is usually challenging to support such massive access with limited resources. Fortunately, directional beams can offer spatial degrees of freedom for random access (RA). In this article, a novel technique called beam splitting/merging is proposed to improve the performance of RA in mMTC network slice with limited resources such as preambles. We describe the framework of beam splitting/merging with the preambles reused, derive the RA throughput of the split beams which is approximately linearly increased with the number of beams, and give a method to obtain beam configuration ensuring the global RA success probability. With the derived analytic results, we further give an analysis of beam configuration in a typical scenario of beam splitting. Moreover, we design an RA control algorithm based on beam splitting/merging technique, by which the mMTC slice can dynamically provision RA capacity according to current access intensity. The updated RA configuration of each beam is broadcast by system information [e.g., system information block 1 (SIB1)]. Detailed analysis and simulation results show that the proposed beam splitting/merging algorithm can effectively improve system efficiency and guarantee RA performance of mMTC network slice with little increase of signaling overhead and delay. The proposed beam splitting/merging framework is applicable to both grant-based and grant-free RA procedures.