Massive Multiple Access using Superposition Raptor Codes for M2M Communications


Machine-to-machine (M2M) wireless systems aim to provide ubiquitous connectivity between machine type communication (MTC) devices without any human intervention. Given the exponential growth of MTC traffic, it is of utmost importance to ensure that future wireless standards are capable of handling this traffic. In this work, we focus on the design of a very efficient massive access strategy for highly dense cellular networks with M2M communications. Several MTC devices are allowed to simultaneously transmit at the same resource block by incorporating Raptor codes and superposition modulation. This significantly reduces the access delay and improves the achievable system throughput. A simple yet efficient random access strategy is proposed to only detect the selected preambles and the number of devices which have chosen them. No device identification is needed in the random access phase which significantly reduces the signalling overhead. The proposed scheme is analyzed and the maximum number of MTC devices that can be supported in a resource block is characterized as a function of the message length, number of available resources, and the number of preambles. Simulation results show that the proposed scheme can effectively support a massive number of M2M devices for a limited number of available resources, when the message size is small.

In this work, we propose a novel RA strategy for M2M communications which provides major improvements in terms of access delay and QoS, by shifting from conventional identification/authentication based RA strategies. The proposed strategy aims to 1) minimize the access delay by enabling the collided devices to transmit at the same data channel, consisting of several RBs, 2) minimize the signalling overhead by signaling once for each group of devices which have selected the same preamble, and 3) minimize the resource wastage due to efficient usage of available resources.

The proposed scheme contains two phases, the RA phase and the data transmission phase. The devices do not need to be identified by the BS in the RA phase; instead, the device ID is sent along with its message in the data transmission phase and later is decoded by the BS. In the proposed scheme, collided devices are transmitting at the same data channel by using the same Raptor code. More specifically, a single degree distribution is used for Raptor codes in all the devices, which significantly simplifies the system design as the code is not dependent on the number of devices or network condition. The BS will need to know only the number of active devices in a data channel to perform the decoding and device identification. In fact, the received signal at the BS can be realized as a superposition of coded symbols sent from the devices, which is then shown to be capacity approaching, when an appropriate successive interference cancellation (SIC) is used for the decoding. This is particularly suitable for M2M communications with strict power limitations, especially when the data size is very small and the number of devices is very large; thus, low rate Raptor codes in the low SNR regime can be effectively used for their data transmission. The maximum number of M2M devices which can transmit at the same resource block is then characterized as a function of the data size and the available bandwidth. The proposed scheme shows an excellent performance in highly dense M2M networks, which makes it an excellent choice for future wireless technologies.

For more details please refer to the Arxiv version of the paper here.

About the author

Provide your thoughts

Your email address will not be published. All fields are required.

Please note: All comments are moderated before being published.

The material or views expressed on this Blog are those of the author and do not represent those of the University.  Please report any offensive or improper use of this Blog to
Skip to toolbar