Please use this identifier to cite or link to this item:
|Title:||Channel characterization and signal transmission for digital signal processing based optical communication systems|
Signal processing -- Digital techniques.
Hong Kong Polytechnic University -- Dissertations
|Publisher:||The Hong Kong Polytechnic University|
|Abstract:||The presence of various degradations in optical transmission systems raises the demand for channel characterization. In the aid of optical performance monitoring (OPM) techniques or channel estimation algorithms, monitoring or estimating the impairments such as optical signal-to-noise ratio (OSNR), chromatic dispersion (CD), polarization mode dispersion (PMD) can be achieved. The OPM can facilitate link supervision, impairment-aware routing and channel compensation. With theoretical analysis, three problems in channel characterization are solved for different generations of optical transmission system. For non-coherent system, OSNR monitoring technique using polarization diversity is studied. The relation among input polarization states, electrical bandwidth and monitored OSNR results are theoretically investigated. It is proposed to use a polarization scrambler (PS) to search for the particular polarization states. Based on the analysis, a simplified OSNR monitoring technique is derived, which advantages in low cost and insensitivity of CD and first order PMD. The proposed technique can achieve modulation format transparent OSNR monitoring at the intermediate nodes for non-coherent systems and enable reliable network operation, supervision and routing. For coherent system, a fast and robust CD estimation technique is proposed by examining a peak in the auto-correlation of the received signal power waveform. Theoretical analysis shows that the location of the peak is indicative of the accumulated CD. Due to its search-free nature, the required number of symbols is less than 1/10 of other searching based CD estimation techniques. The reduction in required symbols can reduce the delay and extra payload in building up a connection. It can also help to reduce the complexity and power consuming of the receiver DSP.|
Finally the channel of quasi-single-mode (QSM) is analytically characterized, in which the single-mode fiber (SMF) is replaced by few-mode fiber (FMF) to suppress the nonlinear effect and allows higher signal power and OSNR, with the tradeoff of additional loss by higher mode power stripped off at the end of each link and the presence of modal coupling induced multi-path interference (MPI). In order to characterize the aggregate performance improvement, 6 liner-polarized (LP) mode model is built to analyze the effect of modal coupling strength, differential modal delay (DMD) and differential modal loss (DML). It is found that for weakly coupling and long-haul transmission, the in- and cross-polarization channels can be modeled as Rayleigh and Gaussian channels. And the MPI effect can be adaptively compensated by the decision-directed least-mean-square (DD-LMS) algorithm for practical implementations. In the experiments for 256 Gb/s PM-16-QAM system, the transmission distance for QSM reaches 2600 km with 501-tap DD-LMS, almost twice as that with SMF. After theoretically and experimentally investigating the performance improvement of QSM over traditional SMF, QSM is shown to be a good candidate in future optical communication systems. The full compatibility with SMF system makes it easy in upgrading the current links and deploying new links. The marginal incremental cost coming from MPI compensation DSP brings about almost double transmission distance as a reward. All of these attributes grant QSM a promising technique in the long-haul transmissions.
|Description:||xiv, 106 pages : illustrations ; 30 cm|
PolyU Library Call No.: [THS] LG51 .H577P EE 2015 Sui
|Rights:||All rights reserved.|
|Appears in Collections:||Thesis|
Show full item record
Files in This Item:
|b27981873_link.htm||For PolyU Users||203 B||HTML||View/Open|
|b27981873_ir.pdf||For All Users (Non-printable)||2.73 MB||Adobe PDF||View/Open|
Checked on Jan 15, 2017
Checked on Jan 15, 2017
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.