Distributed massive wireless access for cellular machine-to-machine communication

Traditional cellular networks play an essential role in the evolution of machine-to-machine (M2M) networks, due to their ubiquitous network coverage. However, M2M communications have some unique characteristics, which are different from the conventional human to human communications, such as the massive number of connected machine type devices (MTD) and stringent QoS requirements. This requires an innovative design of cellular access networks in order to accommodate these requirements. In this paper, we consider the design of M2M communications by using the existing cellular network infrastructure. Since the MTDs do not have their own radio resources, they send their data through the cellular users´ (CU) sub-channels. We propose a distributed algorithm that matches a group of MTDs with a particular CU that aims earning more profit by sharing its resources. Then the MTDs in each group access the sub-channels of their matched CU in a TDMA manner. We prove that the proposed algorithm results in a stable matching after a small number of iterations. It is shown that the weighted sum data rate of the MTDs in the proposed algorithm approaches that of the centralized algorithm as the price step-number is sufficiently small, with a significantly lower overhead than the centralized approach.