Performance Analysis of Polar Codes for Wireless Sensor Networks

Abstract

Towards 2020 and beyond, the Internet of Things (IoT) will become the main driving forces for the development of mobile communications. As an important part of IoT, wireless sensor networks (WSNs) are energy-constrained networks. Energy consumption is the most important factor in the design of WSNs. In addition, link reliability is also a crucial factor to consider when the networks are designed. Using error-correcting codes (ECC) in WSNs can not only increase link reliability, but also reduce the transmitting energy, but commonly at the expense of an increased decoding complexity and power. The aim of this paper is to detailedly discuss the performance metric of polar codes in WSNs. From the simulation results in our work, it can be seen that the bit-error-rate (BER) performance of polar codes is better than LDPC codes but worse than Turbo codes. Besides, a concatenated polar code whose outer code is a “short” BCH code and inner code is a “short” polar code is designed to improve the BER performance and reduce the decoding complexity of “long” polar code. This paper also presents two different interleaving schemes, random interleaving (RI) and blind interleaving (BI) in the design of the concatenated polar code. From the final simulation results, we can learn that the BER performance of the concatenated polar code with the BI scheme is the best among all the coding schemes considered in the simulation. At the same time, the decoding complexity of “long” polar codes can be greatly reduced by constructing the concatenated polar codes.