stillingsbetegnelse:
PhD scholarship in Machine learning for Emulation of Nonlinear Optical Fibre Impairments

jobbeskrivelse:
Applications are invited for a 3-year PhD position within the field of machine learning for the design of optical fibre emulation algorithms for next generation arbitrary waveform generators. The project is partly sponsored by Keysight Technologies (former Agilent Technologies) who is one of the world leaders of test and measurement equipment and software. The start date is flexible, but no later than November 1, 2015. The project is interdisciplinary and will cover topics within the field of machine learning, control theory and optical communication. The DTU Fotonik and Keysight group have already a strong track record in this novel area of applying machine learning techniques to test and measurement equipment. Responsibilities and tasks The main focus of the project is nonlinear inference, especially within the context of optical communication systems that operate in the nonlinear regime. The project will deal with topics such as Bayesian filtering, Markov Chain Monte Carlo and nonlinear signal transmission. The project will be carried out at the Department of Photonics, Technical University of Denmark (DTU). However, some parts of the project may be carried out at Keysight Technologies. In order to satisfy future demands on the data capacity, optical communication systems are experiencing a paradigm shift by moving to sophisticated modulation formats such as quadrature amplitude modulation (QAM). For higher order QAM signaling, optical fiber nonlinearities can severely limit the achievable transmission distance. It is therefore crucial for system-designer to test the robustness of the receiver to optical fibre channel impairments. This can be achieved by performing extensive recirculating loop experiments or by using a novel approach, persued in this project, of having an instrument (arbitrary waveform generator) that can generate nonlinear impairments in a controlled way. However, the second approach requires intelligent signal processing algorithms that can identify, learn and synthesize nonlinear optical fibre impairments. This is a very challenging task and novel approaches are required especially for multi-channel systems. The objective of the project is to explore machine learning techniques to perform optical fibre nonlinearity inference and enable emulation of nonlinear impairments in a controlled way. Most promising techniques will be incorporated and evaluated in an experimental test-bed environment. Qualifications Candidates should have a master's degree in engineering or a similar degree with an academic level equivalent to the master's degree in engineering. Candidates with experience in digital communication, machine learning, optical communication, machine learning, pattern recognition or some other related fields of signal processing are encouraged to apply. Moreover the candidate shall have additional skills as: •Ability to work independently, to plan and carry out complicated tasks, and to be a part of a large, dynamical group •Good communication skills in English, both written and spoken •Experience with Bayesian filtering •Experience with Monte Carlo Markov Chain Methods •Experience with nonlinear optical impairments •Experience with experimental optical setups •Experience using MATLAB or similar •Innovative skills and the ability to generate new ideas Approval and Enrolment The scholarships for the PhD degree are subject to academic approval, and the candidates will be enrolled in one of the general degree programmes of DTU. For information about the general requirements for enrolment and the general planning of the scholarship studies, please see the DTU PhD Guide. Assessment The assessment of the applicants will be made by Associate Professor Darko Zibar. We offer We offer an interesting and challenging job in an international environment focusing on education, research, public-sector consultancy and innovation, which contribute to enhancing the economy and improving social welfare. We

Arbejdsgiver Navn: Danmarks Tekniske Universitet

Arbejdsgiver telefonnummer: 45252525

Arbejdsgiver faxnummer: 45881799

Job kort beskrivelse:
Undervisning og forskning på universiteter og højere læreranstalter

land: Danmark

region: Hovedstaden

krævede uddannelse: Ikke angivet

Kilde til jobtilbud: AMS, Offentlige arbejdsformidlinger, Danmark