TY - JOUR
T1 - Path Identification in a Power-Line Network Based on Channel Transfer Function Measurements
AU - Pagani, Pascal
AU - Ismail, Amr
AU - Zeddam, Ahmed
N1 - KAUST Repository Item: Exported on 2020-10-01
PY - 2012/7
Y1 - 2012/7
N2 - The development of very high data-rate power-line communication (PLC) systems requires an accurate knowledge of the transmission phenomena over the electrical network. In particular, the detection of the multiple propagation paths enables a compact description of the channel models, and gives an indication of the network topology, which may, in turn, be exploited to improve the communication techniques over PLC. In this paper, two high-resolution algorithms for the identification of the propagation paths are studied and adapted to the PLC channel characteristics, namely, the frequency-domain maximum-likelihood (FDML) algorithm and the Matrix Pencil (MP) algorithm. A parametric study is then detailed in order to analyze the performance of both algorithms in terms of resolution, computation time, and residual error. The study demonstrates that the MP algorithm provides a quicker convergence and a lower residual error when compared to the FDML algorithm. Finally, the MP algorithm is validated through its application on experimental network measurements. Results show a good agreement between the measurement and the synthetic channel recomposed from the detected paths.
AB - The development of very high data-rate power-line communication (PLC) systems requires an accurate knowledge of the transmission phenomena over the electrical network. In particular, the detection of the multiple propagation paths enables a compact description of the channel models, and gives an indication of the network topology, which may, in turn, be exploited to improve the communication techniques over PLC. In this paper, two high-resolution algorithms for the identification of the propagation paths are studied and adapted to the PLC channel characteristics, namely, the frequency-domain maximum-likelihood (FDML) algorithm and the Matrix Pencil (MP) algorithm. A parametric study is then detailed in order to analyze the performance of both algorithms in terms of resolution, computation time, and residual error. The study demonstrates that the MP algorithm provides a quicker convergence and a lower residual error when compared to the FDML algorithm. Finally, the MP algorithm is validated through its application on experimental network measurements. Results show a good agreement between the measurement and the synthetic channel recomposed from the detected paths.
UR - http://hdl.handle.net/10754/655941
UR - https://ieeexplore.ieee.org/document/6205345/
UR - http://www.scopus.com/inward/record.url?scp=84862992245&partnerID=8YFLogxK
U2 - 10.1109/TPWRD.2012.2195336
DO - 10.1109/TPWRD.2012.2195336
M3 - Article
SN - 0885-8977
VL - 27
SP - 1081
EP - 1089
JO - IEEE Transactions on Power Delivery
JF - IEEE Transactions on Power Delivery
IS - 3
ER -