Page 244 - Optical Communications Essentials
P. 244

Performance Measures



          234  Chapter Fourteen


                        In a digital system a photodetector in the receiver produces an output voltage
                      proportional to the incident optical power level. This voltage is compared once
                      per bit time to a threshold voltage level to determine whether a pulse is present
                      at the photodetector in that time slot. Ideally the measured voltage would
                      always exceed the threshold voltage when a 1 is present and would be less than
                      the threshold voltage when no pulse (a 0) was sent. In an actual system, vari-
                      ous noises and interference effects cause deviations from the expected output
                      voltage, which leads to errors in the interpretation of the received signal.

          14.1.1. Bit error rate
                      The most common figure of merit for digital links is the bit error rate, which
                      commonly is abbreviated as BER. This is defined as the number of bit errors N E
                      occurring over a specific time interval, divided by the total number of bits N T
                      sent during that interval; that is, BER   N E /N T . The error rate is expressed by
                                          9
                      a number, such as 10 , which states that on the average one error occurs for
                      every billion pulses sent. Typical error rates specified for optical fiber telecom-
                      munication systems range from 10  9  to 10  15 . The BER also is known as the
                      error probability, which commonly is abbreviated as P e .
                        The SNR is related to the BER through the expression
                                                                            2

                                                           2
                                                          2
                                                                          Q /2
                                              1           x         1   e
                                     BER       exp     dx                               (14.1)
                                             2π   0                 2π    Q




                      where the symbol Q traditionally represents the SNR for simplicity of notation.
                                                                               3
                      The approximation on the right-hand side holds for BER  10 , which means
                      it is accurate for all cases of interest in optical fiber communications. Figure 14.1
                      shows how the BER or P e varies with Q. Some commonly quoted values are
                                         9
                      Q   6 for BER   10 , Q   7 for BER   10  12 , and Q   8 for BER   10  15 .
                        The Q value is defined as
                                                         I 1   I 0
                                                    Q                                   (14.2)
                                                        σ 1   σ 0
                      where I 1 and I 0 are the average detected signal currents for 1 and 0 bits, respec-
                      tively, and  σ 1 and σ 0 are the corresponding detected  root-mean-square (rms)
                      noise values, assuming a non-return-to-zero (NRZ) code and an equal number
                      of 1 and 0 pulses.
                        Certain signal degradation effects such as fiber attenuation are linear processes
                      that can be overcome by increasing the received optical power. Other types of
                      noise sources, such as laser relative intensity noise (RIN), are independent of
                      signal strength and can create a noise floor that limits the system performance.
                      Section 14.3 describes the effects of RIN.
          14.1.2. Eye diagrams
                      The eye diagram technique is a simple but powerful measurement method for
                      assessing the data-handling ability of a digital transmission system. This


                 Downloaded from Digital Engineering Library @ McGraw-Hill (www.digitalengineeringlibrary.com)
                            Copyright © 2004 The McGraw-Hill Companies. All rights reserved.
                              Any use is subject to the Terms of Use as given at the website.
   239   240   241   242   243   244   245   246   247   248   249