UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Numerical simulation studies of the effects of noise and optical-fiber distortion on wideband frequency-modulated cable television systems Coenen, Robert B.

Abstract

In the late 1980s, the idea of using optical fiber to carry cable television (CATV) signals in analog form was first proposed and studied. The most widely employed approach today is an amplitude-modulated vestigial-side-band (AM-VSB) implementation in which the optical output of a laser is amplitude-modulated with the multi-channel CATV signal. A serious implementation issue in such systems is receiver noise: as the number of channels increases, the power per channel drops while the receiver noise remains constant. To offset the poor sensitivity, which is limited by receiver shot noise in a well-designed system, the received power must be high. To improve the sensitivity, a radically new approach is required. Wideband frequency modulation (WFM) can improve the sensitivity of the CATV fiber system by exploiting the large bandwidth available in an optical fiber. To date, a careful study of the limiting design factors in a WFM fiber optic system has not been carried out. This work examines the fundamental limiting factors in a WFM fiber optic system and calculates their quantitative effects for a general WFM CATV system. The specific design example of a highperformance 80-channel WFM CATV supertrunking system is also examined. Numerical simulation is used to study the effects of fiber dispersion and other system effects on the WFM system. Although previous studies of WFM CATV systems assume dispersion-shifted fiber, most practical systems today employ existing non-dispersion-shifted fiber. Simulations of group-velocity dispersion and self-phase modulation (SPM) in the WFM system show that fiber dispersion affects the WFM system in an unexpected and very different way than that for the AM-VSB system. In particular, fiber dispersion is shown to degrade the carrier-to-noise ratio (CNR) of the CATV channels by an effect we call dispersion-induced power loss. Thus, CNR becomes the fiber dispersion-limiting factor, rather than composite second-order (CSO) and composite-triple-beat (CTB) distortions. This behavior is the opposite of what occurs in an AM-VSB system, where fiber dispersion limits the system performance through distortions, and has not been reported for WFM before. Plots of the CNR penalty as a function of fiber distance and carrier frequency offer a guide for designing such systems. Analytical calculations support and elaborate the numerical simulation results. A new equation for calculating dispersion-induced power loss in optical fiber in the large-signal case is a product of the analytical calculations, and can be used to measure chromatic dispersion and modulation chirp. The simulation software allows an examination of the effects of optical transmitter frequency response. To aid in the design of practical systems, tables of the maximum allowable non-linearity in the transmitter phase response, validated by comparison to published data, are provided. The effects of finite photoreceiver bandwidth are shown not to be a limiting design factor. Analytical calculations of FM modulator phase noise and double Rayleigh backscattering are performed. Examination of the minimum noise performance required in a WFM system shows that an opto-electronic modulator is not suitable for a high-performance 80- channel WFM system. This result was not expected since previous work indicates that an optoelectronic FM modulator is favored. However, that work is based on more relaxed CNR requirements and fewer channels. Target values for an opto-electronic modulator and electronic modulator phase noise are developed in a format useful in designing a WFM CATV system for any number of channels. It is shown that a WFM CATV system may suffer from excessive noise at the low-frequency channels due to low-frequency FM modulator phase noise. Finally, the contributions of double Rayleigh back-scattering noise and discrete reflection noise are calculated and shown not to be a limiting factor. This result provides a mathematical explanation of other experimental results.

Item Media

Item Citations and Data

Rights

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

Usage Statistics