Access control of MTC devices using reinforcement learning approach

MTC (Machine Type Communication) applications are one of the promising applications in 3GPP system because it connects a huge number of devices into one network. In MTC applications, a huge number of devices attempt to access a system using contention-based random access scheme in a short period. It makes a system overloaded. To solve the overload problem, we propose an access control scheme of devices using Q-learning algorithm. Experimental results show that the scheme adaptively adjusts access control parameter.