Master Thesis: Robust digital predistortion (f/m)

Job description

Job Description

Master Thesis: Robust digital predistortion

Development of a digital predistortion scheme that can cope with a highly non-linear PA characteristic and a corresponding method to train the DPD coefficients.

The "explosion" of transmit bandwidth being introduced by 5G NR will create completely new challenges for mobile terminals. The increase of transmit bandwidth is faster than the deployment of new process nodes that traditionally improve the baseline current consumption year by year. There will be a strong need to introduce new methods and evolved digital circuitries so that current consumption does less scale with the increasing transmit bandwidth. In parallel the importance of digital predistortion will significantly increase due to introduction of 5G NR. The ultra-high bandwidth will boost the complexity of the predistorter and may cause excessive battery current consumption and heat dissipation if there is no further innovation. In addition there is an increasing demand for adaptive predistortion schemes. During UE operation the linearity characteristic of the transmitter might be degraded due temperature changes or due to antenna mismatch effects causing PA load pulling.

Adaptive predistortion can recover the linearity characteristic by measuring the TX signal and by updating the coefficients based on a comparison with a reference signal. Objective of the work is to develop and to comprehensively analyse a predistorter scheme that can cope with a highly non-linear PA characteristic such as gain expansion effects and hard clipping. The work includes system simulations and lab testing to verify the simulation results.

Inside this Business Group

Communication & Devices Group: The wireless revolution at Intel! We are one team - passionate engineers and technologists from diverse industry backgrounds working together to realize a world of connected computing. We are bringing the best ideas from the brightest minds to deliver future mobile experiences into the market. We are on the journey towards making Intel a wireless leader with exciting products for the Internet of Things, 5G and an opportunity to change the world with your work.

Desired profile

Qualifications

Programming languages: Matlab
Knowledge: Digital signal processing, Digital Predistortion,Theory of non-linear system Volterra theory, Wiener system, Hammerstein system, etc.