Design of Circular Microstrip Antenna for Circular Polarization Using ANN

Circular microstrip antenna is one of the popular and widely used types of antenna due to low cost, low profile nature. But it suffers from disadvantages like narrow bandwidth, low polarization purity and low power handling capacity. Circular polarization (CP) is useful for number of applications such as radar, communication and navigational systems.
This paper presents use of artificial neural network, and its use to improve and optimize circular microstrip antenna the design of CMSA in which CP is achieved using single feed and cutting two notches at x-axis. This design eliminates use of external power divider, reducing size of the antenna assembly. This paper also demonstrates approximation capability of artificial neural network. Feed forward network architecture with backpropagation algorithm is used. Depending upon size of the database number of hidden layers and number of neurons in hidden layer are decided. Number of neurons in hidden layer is varied and effect on percentage deviation in the output parameter is observed. Depending upon nature of input and output parameters synthesis and analysis artificial neural network (ANN) models are trained. Results of optimized antenna shows considerable enhancement in the bandwidth. This design provides good VSWR bandwidth. It also considers need of polarization purity required in applications like radar. Axial ratio is achieved near ideal value which ensures circular polarization. Drastic decrease in value of axial ratio is one of the great successes of this work.