Hybrid OFDM-Digital Filter Multiple Access PONs
Research output: Contribution to journal › Article › peer-review
Electronic versions
Documents
- 2018-Hybrid OFDM-Digital Filter Multiple Access PONs
Accepted author manuscript, 796 KB, PDF document
Licence: Other
Hybrid OFDM-digital filter multiple access (DFMA) passive optical networks (PONs) are, for the first time, proposed and extensively investigated, where digital signal processing (DSP)-based software-reconfigurable digital filtering is employed for each individual OFDM signal from various optical network units (ONUs), and a single FFT operation and its following data recovery processes are implemented in a pipelined approach in the optical line terminal (OLT). As a direct result of combining both OFDMA and DFMA PON characteristics, the proposed hybrid OFDM-DFMA PON still maintains highly desirable features associated with DFMA PONs, more importantly, it also offers a number of unique advantages including great relaxation of digital filter DSP complexity and considerable reduction in transceiver cost, significantly enhanced upstream PON performance and its robustness and flexibility, as well as excellent backward compatibility with existing OFDM-based 4G networks. In this paper, an analytical hybrid OFDM-DFMA PON model is developed, and numerical simulations are performed to verify the validity of the proposed technique and also to examine the upstream hybrid OFDM-DFMA PON performance characteristics over representative 25km SSMF IMDD PON systems. It is shown that in comparison with the DFMA PONs, the hybrid OFDM-DFMA PON can improve the upstream ONU EVM performance by >10dB for relatively large received optical powers, and also increases the differential ONU launch power dynamic range by as large as 16dB.
Keywords
- Passive optical networks (PONs), digital signal processing (DSP), digital filter multiple access, OFDM, 5G
Original language | English |
---|---|
Pages (from-to) | 5640-5649 |
Journal | Journal of Lightwave Technology |
Volume | 36 |
Issue number | 23 |
Early online date | 24 Oct 2018 |
Publication status | Published - 1 Dec 2018 |
Total downloads
No data available