Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.759711
Title: The design and analysis of quartic double well potential with stochastic resonance for communication systems
Author: Gunes, Nurhan
ISNI:       0000 0004 7431 7406
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
Non-linearity and noise are two phenomena that are expected to be essential to future advanced technologies. Although largely abstained, in general, from introduction into current communication systems, the counter-intuitive phenomenon called Stochastic Resonance (SR) can be introduced into communication systems in an innovative form. Therefore, in this thesis, the most prominent dynamical system in the SR field, the double well potential, namely the over-damped Duffing equation with symmetric bistable potential, has been studied in order to reveal its signal processing capabilities for communication systems. Within this thesis, the double well potential was designed in order to detect a binary pulse amplitude modulated (BPAM) signal subject to a background noise. The bit-error-rate (BER) performance was enhanced by adding various resonant signals to the input. In addition, the eye patterns of system output indicated that, while decreasing BER, a resonant causes a strong fluctuation. It was eliminated by a use of two systems coupled in parallel, which provided further performance improvement. The results inferred that the double well potential performs filtering and modulation. Following that, the double well potential was designed as a lowpass filter by determining the DC gain and cut-off frequency. Through simulations, as a filter, its noise suppression performance was shown to be better than that of various orders of Butterworth filters. The analog and digital modulation capabilities of the double well potential have also been investigated. In order to clarify the relation between input signal and modulation parameters, the differential equation driving the output was solved, and thus the output was expressed as a function of modulation parameters. It was shown that the output is a multivariate analog modulated signal. In terms of digital modulation, the output of system processing a PAM signal has been interpreted by means of a Markov chain. The results indicated that this process consists of a convolutional coding and multidimensional modulation. In addition, the presence of noise induced coding was found. Finally, the system was designed to obtain a pulse width position modulated (PWPM) output. Throughout the project, detection, filtering, modulation and coding capabilities have been demonstrated, it has been concluded that the double well potential is an sophisticated signal processing tool.
Supervisor: Not available Sponsor: Millî Eğitim Bakanlığı (Turkey)
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.759711  DOI: Not available
Keywords: TK Electrical engineering. Electronics Nuclear engineering
Share: