Optimal Estimation of Schottky Diode Parameters Using Advanced Swarm Intelligence Algorithms
- PDF / 1,991,913 Bytes
- 8 Pages / 612 x 792 pts (letter) Page_size
- 0 Downloads / 189 Views
PECTROSCOPY, INTERACTION WITH RADIATION
Optimal Estimation of Schottky Diode Parameters Using Advanced Swarm Intelligence Algorithms A. Rabehia,b,*, B. Nailb, H. Helala, A. Douarab, A. Zianea, M. Amrania, B. Akkala, and Z. Benamaraa a Laboratoire
de Micro-électronique Appliquée, Université Djillali Liabés de Sidi Bel Abbés, BP 89, Sidi Bel Abbés, 22000 Algeria b Institute of Science and Technology, Tissemsilt University Center, Tissemsilt, 38000 Algeria *e-mail: [email protected] Received May 11, 2020; revised May 11, 2020; accepted July 6, 2020
Abstract—This work deals with estimation of the Schottky diode (Au|GaN|GaAs) optimal parameters. For this purpose, advanced swarm intelligence (SI) algorithms have been applied, i.e., Harris hawks optimization, ant lion optimizer (ALO), grey wolf optimizer, and whale optimization algorithm. The performance of the SI algorithms has been investigated by a comparative study following the analytical methods developed by Kaminski I, Cheung and Cheung, Norde, and Mikhelashvili. The comparative results show that the ALO algorithm gives minimum RMSE criteria, with best parameters estimation against all the SI optimizers and the analytical techniques. Keywords: Schottky diodes, Au, GaN, GaAs, swarm intelligence algorithms, parameters estimation, barrier height DOI: 10.1134/S1063782620110214
1. INTRODUCTION Fabrication and characterization of metal-semiconductor Schottky contacts have attracted much attention in recent years [1–5]. This is due to the potential application in various electronic and optoelectronic devices, such as high-frequency field-effect transistors (FETs), microwave FETs, RF detectors, photodiodes, laser diodes, and solar cells [3]. It is well known that for ideal Schottky contacts, the dominant current is the thermionic emission (TE) [5–9]. But in practice, this mechanism deviates due to the effect of tunnel currents such as thermionic field emission (TFE) and field emission (FE) mechanisms [10]. A non-dimensional parameter called the ideality factor n, as well as the series resistance are included in the model of the current–voltage I(V) characteristics to take into account non-ideal Schottky contacts behavior [7, 11]. Accurate knowledge of the electrical parameters is of importance for the quality control and the evaluation of the Schottky device’s performance. Several methods have been proposed to extract the electrical parameters such as ideality factor n, barrier height φ bn, and series resistance RS from the I(V) characteristics for non-ideal Schottky contacts [7, 12–15]. However, most of these methods lack credibility and reliability. They are essentially based on identifying each parameter from restricted regions of the I(V) characteristics, while the effect of other parameters is assumed to be negligible [9].
Evolutionary algorithm (EA) techniques based on artificial intelligence are effective methods in semiconductor device modeling. EA such as genetic algorithm, the penalty-based differential evolution, the particle swarm optimization, and the teac
Data Loading...
