We study a relatively excellent BHD(Balanced Homodyne Detector)with large bandwidth and low noise pre-amplification, which is different from the specific BHD implementation scheme and can better adapt to the subsequent cascade scheme, thereby achieving the implementation of BHD with specific needs. Specifically, we optimize our BHD performance by compromising various parameters of BHD, focusing on optimizing and increasing BHD bandwidth, and choosing a more reasonable BHD implementation solution. Judging from the test results, the produced BHD has good performance in extracting effective signals. The bandwidth reaches about 2GHz, which meets the requirements of detection applications. It has a high signal-to-noise ratio of 10 dB and a common-mode rejection ratio (CMRR) of 20 dB near the 2 GHz frequency point.
We proposed a Raman and erbium-doped fiber amplifiers (EDFA) hybrid bidirectional optical amplifier (HBOA). The hybrid amplifier consists of a fiber Raman amplifier (FRA) and two EDFA, which are ingeniously combined by four WDMs. Gain and noise figure (NF) are the metrics used to evaluate the performance. The effects of Raman and EDFA pump on the gain and NF of the proposed hybrid amplifier are explored, resulting in a maximum gain exceeding 40 dB and an NF below 0 dB. Furthermore, when the input signal power is small (−35 dBm), the proposed hybrid amplifier, exhibits a total gain increased of ∼14 dB, and NF reduction of ∼4 dB across the wavelength span of 1545 to 1565 nm, as compared with EDFA-only. Finally, The Allan deviation can reach 1.41×10−14 when HBOA was used, significantly better than the stability of 2.62×10−14 obtained by Bi-EDFA, which confirmed the feasibility of the proposed amplifier for radio frequency synchronization systems.
A single-longitudinal-mode (SLM) narrow-linewidth Brillouin erbium-doped fiber laser (BEFL) is proposed and demonstrated experimentally. The erbium-doped fiber (EDF) is employed to act as both the linear gain and Brillouin gain medium, which makes it easy to excite the stimulated Brillouin scattering (SBS) in the EDF by cooperating with the ring cavity structure. In order to realize stable SLM and narrow-linewidth laser output, the fiber saturable absorber (SA) and the self-injection feedback structure are added to BEFL for the first time, which can suppress the multimode phenomenon effectively and obtain a stable SLM status. The wavelength stability is less than 1.32 pm over 45 minutes and the linewidth is as narrow as 283 Hz.
Temperature variation is a key factor affecting the stability of fiber optic frequency transfer systems. In a long-distance optical fiber frequency transfer system, a dispersion compensation fiber (DCF) placed in the machine room and a non-buried single-mode fiber (SMF) are collectively referred to as a bare optical fiber. Due to the impact of the external environment, the temperature of the bare optical fiber varies drastically, which seriously deteriorates the long-term stability (10,000 s stability) of the frequency transfer system. To investigate the relationship between the temperature variation of the bare optical fiber and the frequency transfer stability, we use VPItransmissionMaker optical simulation software to simulate and study the fiber frequency transfer system of the bare optical fiber under the impact of different temperatures. The simulation found that the system stability is 1.4 × 10 − 16 / 104 s in a 3000 km link when the temperature variation of DCF is 5°C, whereas the impact on the system stability is on the order of 10 − 16 / 104 s when the length of the non-buried SMF is more than 300 km and the temperature variation is more than 10°C. To verify the correctness of the simulation, the stability of the actual 80 km fiber optic frequency transfer system was compared with that of the simulation, and the stability of the simulation was found to be 4.2 × 10 − 17 / 104 s, whereas the stability of the experimentally measured system was 7.9 × 10 − 17 / 104 s, and the simulation results were basically consistent with the experimental results.
Shannon's information theory teaches us that the amount of information gained in a measurement is inversely proportional to its predictability. Difficult to capture, flash-like signals contain far more information than repetitive waveforms. The Photonic Time Stretch data acquisition invented two decades ago, has emerged as the most successful solution to single-shot measurements of transient events. This talk will review the fundamentals of photonic time stretch and its numerous applications in science, biomedicine and as mathematical inspiration for a new class of numerical algorithms.
KEYWORDS: Wavelength division multiplexing, Artificial intelligence, Nonlinear filtering, Machine learning, Digital signal processing, Data modeling, Binary data, Signal to noise ratio, Receivers, Performance modeling
Wavelength Division Multiplexing (WDM) is the key technology in ultra-high capacity links that form the backbone of the internet. Hundreds or more data channels each at a different wavelength travel through a single fiber resulting in aggregate data rates exceeding many Terabits per second. The fundamental limit to the data transmission rate is the optical crosstalk between channels induced by the inevitable nonlinearity of the fiber. Traditional methods for compensating for the reduction in the bit error rate caused by the crosstalk include numerical backpropagation as well as nonlinear Volterra filter, both implemented in the digital domain at the receiver. Backpropagation through the canonical nonlinear Schrodinger equation is computationally expensive and beyond the capability of today’s DSP at the data rates that optical networks operate. Volterra filters scale superlinearly with an increasing number of taps and which in turn scale with the amount dispersion in the fiber. Therefore, they are not the ideal solution for high data rates. In this talk, we report on the application of machine learning, and neural networks in particular, on the compensation of optical crosstalk in WDM communication. We compare the performance of different machine learning models such as support vector machine (SVM), decision tree, convolutional neural network (CNN) in terms of the achievable bit error rate on both binary and multilevel modulated data. We further evaluate the sensitivity of the error rate to the resolution of the analog to digital converter (ADC) and to the signal to noise ratio as well as the latency of our algorithms
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.