Poster Session Abstracts
Incoherently Pumped Tunable Raman Fibre Laser Operating Across 1625-1653nm
Optimization of Fibre Optical Parametric Amplifiers for QAM Signal Amplification Mariia Bastamova, Aston Institute of Photonic Technologies
Nura Adamu, Hao Liu, Kyle R. H. Bottrill, Periklis Petropoulos, Optoelectronics Research Centre, University of Southampton Fibre laser systems operating in the U-band (1625- 1675 nm), are attractive for a range of applications, such as methane sensing, deep tissue imaging, eye surgery and LIDAR. However, achieving efficient
Fiber optical parametric amplifiers (FOPA) offer theoretically unlimited bandwidth of operation in almost arbitrary wavelength ranges, the capability of phase-sensitive noiseless amplification, high gain, and applicability for ultra-fast transient-
laser operation in this region based on doped optical fibres is very challenging. This makes Raman fibre lasers (RFLs) attractive, due to their design simplicity and versatility in their operating wavelength. We demonstrate a RFL based on an incoherently pumped ring cavity configuration realized using a non-zero dispersion shifted fibre (NZDSF). Amplified spontaneous emission (ASE) from an erbium- doped fibre amplifier (EDFA), shaped with a gain flattening filter, was used as the incoherent pump source, giving rise to a broad gain bandwidth. An optical signal-to-noise ratio (OSNR) greater than 55 dB was obtained at 1650 nm. This work shows feasibility of constructing RFLs using off-the-shelf components to deliver efficient gain in the challenging 1625-1675nm region.
free applications. These features are crucial for applications involving the amplification of extremely low-power or few-photon signals, such as space or quantum communications. We demonstrated that the challenge of gain fluctuations in FOPA arising from pump phase modulation, necessary for mitigating Stimulated Brillouin Scattering, does not cause degradation for coherently detected QAM signals compared to OOK signals. The reason is that QAM signals rely on the electric field amplitude being the square root of power. The design of a polarization-diverse FOPA architecture allows for further compensation of the impact of pump phase modulation by adjusting the pumps’ optical path difference and the pump phase modulation frequencies. Theoretically this approach can decrease the required optical signal-to-noise ratio penalty by a factor of 10.
Scan to watch poster preview.
Scan to watch poster preview.
Automating Network Growth: Graph Machine Learning Approaches to Fibre Expansion Akanksha Ahuja, Sam Nallaperuma, Albert
Scaling Quantum Networks: Fidelity and Coherence in Distributed Systems
Rafel, Paul Wright, Andrew Lord and Seb J. Savory, Department of Electrical Engineering, Darwin College, University of Cambridge Optical network providers must expand the physical infrastructure of national-level core
Irene Cáceres Muñoz, Georgios Zervas, UCL We explore the scalability of quantum distributed systems by analyzing how the intranode and communication noise affects their performance and reliability in different topologies. The current bottleneck in quantum processor development is often attributed to decoherence resulting
networks efficiently and regularly by adding new fibre onto the network to increase capacity, maintain optimal utilisation and ensure resilience. Physical topologies can be represented as graphs, where each node represents a geolocation hosting network equipment, and each edge represents an optical fibre connecting these regions. Fibre prediction or edge prediction is the forecasting of new connections in optical networks and is critical for automating the scalability of the infrastructure. Our research aims to predict edges among existing nodes to inform optical network expansion planning and infer patterns of network design. We represent the physical topology as a graph and develop machine learning and graph embedding techniques to predict fibre connections, quantitatively demonstrating high edge prediction accuracy on real-world core networks. The predictive graph learning approach outperforms heuristic methods and captures the intelligence of network design.
from attempts to scale the number of qubits per node. Meanwhile, networked systems face challenges related to the connection with quantum communication channels and increased operations per node to perform equivalent circuits. We focus on the creation of Greenberger-Horne-Zeilinger (GHZ) states among the network, which exhibit maximal entanglement between all the qubits. We investigate the impact that network size, node subdivision and qubit allocation have on their creation in an environment that accounts for thermal noise and depolarizing quantum channels. Our analysis aims to uncover insights into the scalability limits and coherence preservation within distributed quantum systems. This research contributes to optimizing large-scale quantum networks in the era of transformative quantum technologies.
Scan to watch poster preview.
14
Made with FlippingBook - professional solution for displaying marketing and sales documents online