Open Collections

In-vehicle powerline communication using software-defined radio

University of British Columbia

Powerline communication is an attractive solution for in-vehicle communication. However, the research of communication over powerlines requires field-testing and full access to communication protocol layers, particularly to the physical and the Media Access Control (MAC) layers. This ability can be accomplished through the use of software-defined radio along with real-time signal processing executed on a personal computer. In this work, we present the design and implementation of an IEEE 1901-based transceiver aimed for vehicular powerlines, written for GNU Radio and operated on Ettus Universal Software Radio Peripheral (USRP) N210 hardware. The software components include a C++ physical layer signal processing library and several complementary GNU Radio blocks including a MAC layer block written in Python. The implemented capabilities include several channel estimation methods, a noise power spectral density estimator and an adaptive bit loading algorithm. We make all the software components available as an open source project to facilitate further development by the broader research community. We then show experimental results obtained with the system applied to a vehicle harness and a real vehicle powerline network. In the first part of the experiments, we demonstrate the correctness of the implementation, compare between several channel estimation methods, and test the system performance. In the second part, we examine the feasibility of reliable communication with IEEE 1901 over powerlines in a car. Our experiments show that IEEE 1901 along with the implemented receiver algorithm are capable of operating in the scenarios tested. The vehicular impulsive noise is identified as the primary cause for errors. In particular, our experiments show that it affects mainly the frame synchronization. Hence we believe that further investigations of in-vehicle powerline communication should focus on alleviating the effect of impulse noise on synchronization.

Text

2016-08-12

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/58762

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/