04-04-2012, 03:51 PM
Research and Development in Optical CDMA Systems
Sohan.pdf (Size: 1.08 MB / Downloads: 89)
Introduction
Over the last one to two decades, there has been a lot of interest and research in
optical CDMA systems. More than 250 papers have been written in this area since 1985
[12]. A vast number of different schemes using time domain or frequency domain
encoding approaches have been proposed.
Time Domain Encoding Using Optical Delay Line Loops
The first optical CDMA proposals were found in Hui [13] and Prucnal [14-15]. It
was intended as a multiple access protocol in a local area network (LAN).
Intensity modulation and direct detection (IM-DD) has been established as the
most suitable signal modulation and detection scheme in optical communication systems.
Although coherent detection schemes are also possible, it is very costly to have a local
oscillator at the optical frequency. In contrast, most radio communication systems make
use of coherent schemes, which detect the electrical field.
Non-coherent Spectral Intensity Encoding
Zacarrin and Kavehrad first described this approach [28-33]. It is similar to the
coherent phase encoded system in the sense that the frequency components from a
broadband optical source are resolved first. Each code channel then uses a spectral
amplitude encoder to selectively block or transmit certain frequency
Non-coherent spectral intensity encoded optical CDMA system.
Since there is a subtraction between two photodetectors, it is possible to design
codes so that full orthogonality can be achieved with the non-coherent spectral intensity
encoding approach. In principle, orthogonality eliminates the crosstalk from other users.
Bipolar signaling can also be obtained by sending complementary spectrally encoded
signals
Principle
The block diagram of a non-coherent spectral intensity encoded bipolar CDMA
system is shown in Fig. 3.1. As mentioned before, one of the important goals here is to
obtain full orthogonality. The optical signal intensity that we are dealing with here is a
positive quantity. However, to achieve full orthogonality, both positive and negative
quantities are required. The way to obtain negative quantities with the positive optical
intensity signal is use an analog subtraction operation.
Conclusion
In this chapter, we have discussed the general scaling complexity for
demultiplexing signals in TDM, WDM and CDM systems. It is interesting that in all
three multiplex access systems, logarithmic demultiplexing complexity can be achieved
theoretically. For the WDM system, the simplest demultiplexer is a tunable wavelength
filter, which has a complexity of 1 and independent of the number of wavelengths.
However, there are still technical challenges that limit its use.