Nonlinear State–Space Model of Semiconductor Optical Amplifiers With Gain Compression for System Design and Analysis

Kuntze, S.B.; Zilkie, Aaron; Pavel, Lacra; Aitchison, J. S.

doi:10.1109/jlt.2008.922212

Cited by 24 publications

(13 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the past, models without spatial resolution have been employed for the description of semiconductor amplifiers [BER03b,KUN08,ERN09a,KIM09a]. As soon as strong spatial inhomogeneities arise, such models are, however, bound to fail.…”

Section: Quantum-dot Semiconductor Optical Amplifier Modelmentioning

confidence: 99%

Nonlinear and Nonequilibrium Dynamics of Quantum-Dot Optoelectronic Devices

Lingnau

2015

Springer Theses

View full text Add to dashboard Cite

In this thesis the dynamics and performance of optoelectronic devices based on semiconductor quantum-dots are investigated.In the first part, the dynamics of quantum-dot lasers under external perturbations is discussed. Using a microscopically based balance equation model that incorporates detailed charge-carrier scattering dynamics and the possibility to describe nonequilibrium between intra-band electronic states, the relaxation oscillations of the quantum-dot laser are investigated. Three qualitatively different dynamic regimes are identified in dependence of the scattering rates -the "constantreservoir" regime for slow scattering, the "overdamped" regime, and the "synchronized" regime for high scattering -characterized by a varying degree of nonequilibrium between the quantum-dot and reservoir states.Important differences to conventional lasers are found in the modulation response and the dynamics in optical injection and feedback setups. Common theoretical models and approaches used to describe these applications are shown to yield inaccurate predictions, especially in the "constant-reservoir" and "overdamped" dynamic regimes. An important consequence is that the amplitude-phase coupling in quantum-dot lasers, commonly described by the α-factor, differs from conventional descriptions due to the desynchronization of gain and refractive index. While the α-factor describes bifurcations of fixed points accurately, it fails in describing dynamic solutions and overestimates the extent of complex dynamics. The observed low sensitivity to optical perturbations in quantum-dot lasers can therefore be attributed partly to the charge-carrier nonequilibrium. Three quantum-dot laser models on different levels of sophistication are presented that can accurately describe the quantum-dot nonequilibrium dynamics.In the second part of the thesis, the performance of quantum-dot semiconductor optical amplifiers is investigated, and two types of applications unique to quantum-dots as active medium are discussed. The ground and excited states of the quantum-dots allow an ultra-broad-band amplification of optical data streams. Amplified signals on the ground-state frequencies are shown to generally exhibit higher quality than on the excited state, due to a lower sensitivity of the groundstate to carrier-density variations. Nevertheless the quantum-dot amplifier is found to allow effective amplification on both frequency ranges. Furthermore, a parameter range is identified that allows for a simultaneous amplification of data signals on the ground and excited state in a counter-propagating setup.The long microscopically polarization dephasing times in quantum-dots are found to enable quantum-coherent interactions on a macroscopic scale at room-temperature. By comparison with experiments, the occurrence of Rabi-oscillations by amplification of ultra-short pulses is demonstrated. Quantum-dot based devices could therefore be used for future applications based on quantum-coherent effects.

show abstract

Section: Quantum-dot Semiconductor Optical Amplifier Modelmentioning

confidence: 99%

Nonlinear and Nonequilibrium Dynamics of Quantum-Dot Optoelectronic Devices

Lingnau

2015

Springer Theses

View full text Add to dashboard Cite

show abstract

“…The SOA is modeled as a traveling wave amplifier [12,13]. In the amplifier the saturation effect and phase change is due to time dependent gain and refractive index coupling respectively.…”

Section: Theoretical Analysismentioning

confidence: 99%

Design and analysis of all-optical AND, XOR and OR gates based on SOA–MZI configuration

Singh

Tripathi

Jaiswal³

et al. 2015

Optics & Laser Technology

View full text Add to dashboard Cite

“…It has been previously demonstrated that the SOA can be treated through lumped, input/output element modelling, by averaging the longitudinal carrier density distributions to a single, time dependent output value [6][7][8]. This assumption holds under conditions that the SOA photon transient time is much shorter than the time scale of modulated carrier density variations, and much longer than the SOA propagation delay [7,8].…”

Section: Introductionmentioning

confidence: 99%

“…This assumption holds under conditions that the SOA photon transient time is much shorter than the time scale of modulated carrier density variations, and much longer than the SOA propagation delay [7,8]. Considering the typical values of optical transient times (100 ps À 1 ns) and active region lengths (400 μm-800 μm) in RSOA, and today's typical applications for up to 10 Gb/s transmission by using direct RSOA signal modulation, we assume the carrier density ℵ to be constant over the RSOA active region length, ℵðz; tÞ ¼ ℵðtÞ [7]. Thus, the partial differential equations governing the photon and carrier densities can be approximated by first order nonlinear ordinary differential equations [7].…”

Section: Introductionmentioning

confidence: 99%

“…Considering the typical values of optical transient times (100 ps À 1 ns) and active region lengths (400 μm-800 μm) in RSOA, and today's typical applications for up to 10 Gb/s transmission by using direct RSOA signal modulation, we assume the carrier density ℵ to be constant over the RSOA active region length, ℵðz; tÞ ¼ ℵðtÞ [7]. Thus, the partial differential equations governing the photon and carrier densities can be approximated by first order nonlinear ordinary differential equations [7]. Therefore, under the aforementioned assumptions, real or complex-valued (CV) neural networks are suitable solutions to model RSOA as a parametric identification of large-signal equivalent model, to speed-up the design process and reduce the time to market-production.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation