Network Infrastructure and Architecture: Designing High-Availability Networks

8

CIRCUITS FOR OPTICAL-TO-ELECTRICAL CONVERSION

8.1 Introduction

8.2 Optical to Electrical-to-Optical Conversion

8.2.1 Principle of Operation

8.2.2 Optical Transceiver Architectures

8.2.3 Integrated Circuits for Optical Transceivers

8.3 Signal Amplification

8.3.1 Trans-Impedance Amplifier

8.3.2 Limited Amplifier

8.3.3 Laser Driver

8.4 Phase-Locked Loop

8.4.1 Phase-Locked-Loop Architecture

8.4.2 Voltage-Controlled Oscillator

8.4.3 Phase and Frequency Detectors

8.5 Clock Synthesis and Recovery

8.5.1 Clock Synthesis

8.5.2 Clock and Data Recovery

8.5.3 Jitter Requirements

8.6 Preemphasis and Equalization

8.6.1 High-Speed Signal Impairments

8.6.2 Preemphasis

8.6.3 Equalization

Key Points

References

8.1 INTRODUCTION

Although light seems to be ideal for sending signals in the core of a network, electrical signals are obviously used on the network's edges, as all terminal equipment, be it PCs, cell phones, or large servers, utilizes electronics. The fact that the edge of the network is electrical and the core is optical implies that somewhere in the network a conversion process from optical to electrical (O-E) needs to take place. So, how is O-E conversion done? The goal of this chapter is to show you how electrical signals are converted to optical siguals, and conversely, at the end of the optical transmission, how they converted back to the electrical domain. Since we have already discussed light sources and detectors in Chapter 2, it is probably quite intuitively clear how this task ...