Why Machine Learning in Antenna Design?

Ms. Annie Threse Edwis is a teaching professional in SQA Approved Center, University of Stirling RAK Campus. She has completed her graduation from Madurai Kamaraj University, India in Electronics & Communication Engineering  &  master’s degree from Anna University, India in Communication Systems.

She has around 16 years of teaching experience at various ranks in UAE & India. She had presented 2 papers in National conferences & published a paper in  IUP Journal of Electrical & Electronics Engineering . Her area of interest includes Digital Electronics & Logic Design.

She has proven experience in the development of strong rapport with students, Colleagues & Administrators. In addition to her teaching responsibilities, she is the Data Manager of SQA, RAK Centre. She is also the head of Quality assurance & Exam cell. She is part of various committees namely Research & Innovation Cell and Academic Enhancement Cell.

Ms. Annie Threse Edwis
Assistant Professor

An Antenna is a transducer, which converts electrical signal into electromagnetic waves and vice versa. In the field of communication systems, whenever there is a need for wireless communication, there arises the necessity of an antenna. Antenna is a device which is capable of sending or receiving the electromagnetic waves for the purpose of communication, where laying down the cable is hard.

Source: https://zhuanlan.zhihu.com/p/114524922

In order to meet the increased demand of higher data rate communication, 5G plays a significant role. It is potentially 100 times faster than 4G. 5G offers high speeds, low latency and reliable connectivity to wireless devices. To enjoy these benefits, antennas play a vital role. The right high-performance antenna located in the correct position improves transmission distance and signal quality and allows it to better exploit the benefits of 5G. Over the last few decades, antennas and their associated systems have evolved rapidly due to unprecedented changes happened in their geometric and material profiles to meet modern communication requirement like 5G. 


Due to increasingly stringent specifications and the realization of additional performance requirements, present-day antenna structures are usually topologically and electromagnetically complex with large number of sensitive design parameters. Consequently, finding the best designs, which fulfil the desired performance of the antennas, could be very challenging.   Since, the antenna design is involved with a computationally expensive EM simulator, the simulation time ranges from a few seconds to a few days, depending on the size of the antenna, operating frequency, and computational power. Since, the antenna needs to be optimized to maximize the performance while reducing the size, rapid optimization techniques are highly desirable.

Machine learning is one of the rapidly emerging disciplines that can be widely applied in the fields of engineering, science, medicine, economics etc. Machine learning is based on algorithms that can learn from data without relying on rules-based programming. Recently, the application of machine learning has also been extended to electromagnetics. Rather than using an EM simulator, machine learning algorithms can work as an alternative to the general optimization techniques which can provide optimized values while maintaining high accuracy levels, with a minimization of error and time saving.