Machine-to-Machine communications have emerged as a promising technology to enable trillions of multirole devices, namely machine-type communications (MTC) devices, to communicate with each other with little or no human intervention. It has many potential applications, such as intelligent transportation systems (ITS), healthcare monitoring, retail, banking, smart grids, home automation and so on. It is expected that in the next a few years over 2 billion MTC devices will become directly attached to cellular networks to provide M2M communications . Thus, there will be a massive number of MTC devices with no/low mobility in each cell, which is significantly more than the number of users in current cellular networks. Moreover, M2M traffic involves a large number of short-lived sessions, attempting to deliver a small amount of data (few hundred bits) to the base station, which is quite different from thos e in human-to-human (H2H) communications.
In most existing wireless access networks, the first step in establishing an air interface connection, is to perform random access (RA) in a contention manner. In fact, short-lived sessions with small amount of data in M2M communications, makes it inefficient to establish dedicated resources for data transmission. Thus, one of critical challenges in M2M communications is to design an effective medium access scheme to support a large number of devices. In current RA approaches, when two or more devices select the same RA preamble in the first phase of the contention phase, a collision will occur and the respective devices will not be scheduled for data transmission. Existing random access schemes still suffer from very high access delays in highly dense networks. Additionally, different MTC devices have diverse service requirements and traffic patterns in M2M communications. Conventional random access approaches are mostly inefficient for M2M communications as they are generally designed for a fixed payload size and thus, cannot support 2 M2M applications with different service requirements.
We consider a realistic model for M2M communications, which supports both regular and random traffics with different delay and service requirements. We develop a practical transmission scheme for M2M communications based on recently proposed analog fountain codes (AFCs) to enable massive number of devices communicating with a common base station (BS) while satisfying QoS requirements of all devices. We further show that the proposed scheme can closely approach the fundamental limits of M2M communications in terms of throughput and can satisfy the delay requirements of all devices at the same time. The main advantages of our proposed scheme are summarized below:
1. Efficient Random Access Proposal
2. Efficient Maximum Throughput Transmission
3. QoS aware Random Transmission
We have recently published a paper titled as “Probabilistic Rateless Multiple Access for Machine-to-Machine Communication” in IEEE Transactions on Wireless Communications, which explains the idea in more details. Interested readers are refer to this paper for more details, which can be found in IEEE eXplore and ArXive.