Bandwidth Analysis of Dual-Feed Slotted Antenna Using Artificial Neural Networks

2. Designing And Data Creation

Figure 1 describes the geometry of the conventional antenna. A radio wire has 33.4 mm×40.6 mm altered ground plane and 23.8 mm×31 mm of slotted cross patch measurements. The microstrip receiver has the length L the width W, which is shown in Figure 1.

Figure 1.

Geometry of proposed microstrip antenna

The substrate chosen for this plan is FR-4 glass epoxy with dielectric consistent (ε_r) = 4.4 and tallness of the substrate (h) = 1.6 mm. The fringing field is increased by using low dielectric constant thus the radiated power is increased. The feed is given at the point (x, y). The anticipated antenna is simulated on IE3D software to get the sufficient date for ANN. Figure 1 shows the format of a coaxial test feed slotted antenna (Abhishek et al., 2018; Grilo & Correra, 2015; Kunal et al., 2018; Manju & Singh, 2018; Niharika et al., 2018; Princi et al., 2018; Rahul et al., 2018; Rawat & Sharma, 2014). By differing the feed position of Port-1 and Port-II of the anticipated antenna the training date have been generated.

2.1 ANN Model

The ANN model has been produced for opened cross patch receiver as appeared in Figure 2. The feed forward system has been used to ascertain the data transfer capacity of the antenna by placing, substrate dielectric constant (ε_r) and substrate width (h). This is characterized as examination ANN model. So by giving the different probes at the input of ANN model, the large bandwidth is received without complex calculation.

Figure 2.

Artificial Neural Network Model