A Case Study of Nelder Mead Simplex Optimization Algorithm: Trade-Offs of Sprienski Fractal Bowtie Antenna Parameters

Abstract

In this study, tri-band antenna design adapted for wireless communication, Internet of Things (IoT) and RFID systems is examined. The simulation results indicate that the proposed antenna has three distinct frequency bands. Band 1 (lower band) covers the frequency range of 1.64-1.78 GHz with a resonance frequency of 1.7 GHz. Band 2 covers the range of 3.06-3.9 GHz with a resonance frequency of 3.4 GHz with a high gain of 10 dBi and a radiation efficiency of 92% for long-range communication. Band 3 radiates from 6.25 to 7.6 GHz with a resonance frequency of 6.62 GHz, which is suitable for higher-frequency applications. The antenna design is simulated and analyzed regarding S11, VSWR, gain, radiation efficiency, and bandwidth. Especially, Band 2 (mid-band) provides notable performance, with its 10 dBi gain and 92% efficiency, which makes the proposed antenna an ideal structure for high-data-rate, long-distance communication systems, and 5G (midband) applications. This study also employs the Nelder-Mead Simplex algorithm to observe the optimization of the physical parameters of the proposed antenna via multiple objective functions. The optimization results outlines that longer the arm length of the proposed antenna causes to decrease the resonance frequency of Band 3. Addition to this, the gain is higher with the lower arm length except for the arm length of 90.467 mm and flare angle of 64.77o. That’s, the trade-off condition occurs between minimum return loss and gain. At this point, it can be concluded from this optimization algorithm results that each objective function should be evaluated separately due to this trade-off condition.