Monash University
Browse
4725511_monash_173210.pdf (2.57 MB)

Practical physical layer network coding

Download (2.57 MB)
thesis
posted on 2017-03-05, 22:57 authored by Kramarev, Dmitry
Physical-layer network coding (PNC) is a new technology which has the potential to increase network throughput beyond existing standards based on routing. Despite the fact that PNC has been well investigated from information-theoretic point of view, only a few partial prototypes have been reported in the literature. The implementation of a PNC system is burdened with many challenges such as carrier-phase, symbol and frame asynchrony. In this research, we mainly focus on software-defined radio prototyping of a two-way relay network utilizing PNC relaying. We present the first real-time implementation of a generalized PNC algorithm, namely compute-and-forward relaying. In addition, we propose an improved compute-and-forward relaying scheme which simplifies the use of power-of-two size constellations typically used in practical communication systems. The presented testbed provides a valuable platform for verification of the theoretical research on PNC and evaluation of synchronization requirements. Our experimental results show that when the signal-to-noise ratio is high, PNC relaying outperforms other relaying strategies in terms of the network throughput. In addition, we propose a new method of multiplierless pulse-shaping filter design which allows essential reduction of the hardware utilization as well as out-of-band power. Therefore, a multiplierless filter designed with the proposed method is especially suitable for FPGA/VLSI implementation.

History

Campus location

Australia

Principal supervisor

Emanuele Viterbo

Year of Award

2016

Department, School or Centre

Electrical and Computer Systems Engineering

Degree Type

DOCTORATE

Faculty

Faculty of Engineering

Usage metrics

    Faculty of Engineering Theses

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC