Figure 7.3 shows the transfer functions for both the optical intensity and the optical field (I/Q component) against the drive voltage. For instance, in coherent detection systems, the transformation between the electrical drive voltage and optical field is of concern, whereas in the conventional direct detection systems the transformation between the electrical drive voltage and the optical power is of concern. The null bias point for CO-OFDM up-conversion signifies a fundamental difference between the optical intensity modulation and the optical field modulation. The quadrature bias point has been widely adopted for both analog and digital direct detection systems. Thus, we see no urgent need to further investigate the reliability and robustness of proposed methods or to develop alternative concepts based on DSP in direct detection systems. With significant SPM, the channel memory length is likely to exceed the equalizer memory length, which also degrades the estimation (compare discussion in Section 10.3.3).Īlthough the MLSE principle might be an interesting candidate to solve problems with nonlinearity in coherent detection systems, the importance of direct detection systems with equalizers based on MLSE is likely to decline in the future. Similarly, the estimation of pure OSNR is altered by the influence of XPM. However, as soon as the system exceeds the reference parameter space or if different system conditions refer to similar state models, the estimation becomes unreliable. As the parameter space can be reduced to typical operation points of the system, a reference-based method comparing the actual system state to a set of known system states delivers respectable results. Although there is a systematic separation of deterministic signal components and noisy distortions, the identification of individual deterministic distortions like CD or PMD is not possible, nor is the identification of different noise sources. In conclusion, OPM from the state model of MLSE equalizers in direct detection systems is elaborate and only allows for limited access to individual channel parameters. Hauske, Maxim Kuschnerov, in Optical Performance Monitoring, 2010 10.2.5 Conclusion We conclude the chapter by presenting our view on the future of short-reach transmission systems.įabian N.
We then introduce advanced DD systems where novel transmitter and receiver designs are employed, so as to enable ultra-high interface rates for short-reach applications. In this chapter, we review the principles of conventional DD systems and discuss their limitations. Nowadays, typical transmission distances for DD systems range from few meters to about 100 km. The applications of DD systems were then limited to cost-sensitive short-reach applications such as metro transports, intra- and interdatacenter interconnects, and passive optical networks. Later, core networks transitioned to coherent systems, as they offer higher receiver sensitivity and better spectral efficiency, and therefore are more suitable for long-distance transmission at interface rates of 100 Gb/s and beyond. Schematic structure of a typical transceiver used in a direct-detection system.Īt the time when the required rates per transceiver interface were 10 and 40 Gb/s, DD systems were widely deployed for various kinds of fiber communication systems ranging from short-reach metro connections to long-reach links in core networks.