The invention relates to a technique for optical communications and, more particularly, to a technique for transmission of polarization-multiplexed optical signals containing data.
This section introduces aspects that may help facilitate a better understanding of the invention. Accordingly, the statements of this section are to be read in this light and are not to be understood as admissions about what is prior art or what is not prior art.
Wavelength division multiplexing (WDM) and signal coding, e.g., differential quadrature phase shifting keying (DQPSK), are commonly used in optical communications to efficiently utilize limited transmission bandwidth. The optical communication industry currently is in hot pursuit of 100 G optical technology whereby a transmitted wavelength can be used to carry information at a rate on the order of 100 Gbps. To realize the 100 G optical technology, use of a polarization-multiplexed scheme in addition to current optical transmission schemes, e.g., DQPSK and WDM, has been contemplated to increase the transmission spectral efficiency.
The invention is premised upon a recognition that use of multiple optical multiplexers (e.g., arrayed waveguide gratings (AWGs)) in a typical transmission system to multiplex modulated optical signals of different polarizations is not desirable. Among other things, the multiple AWGs need to be spectrally aligned, e.g., to the same frequency (wavelength) grid, defined by a standard promulgated by the International Telecommunication Union (ITU) in order for the system to work properly. As a result, for example, two thermo-electric coolers (TECs) may be needed in the system to individually adjust the temperatures of the AWGs to tune and align their operating frequency ranges.
In accordance with an embodiment of the invention, only one passive optical device (e.g., an AWG) is used in an optical transmission system for multiplexing first and second sets of optical signals, which contain data and are associated with first and second polarization states, to provide first and second multiplexed signals, respectively. The first and second multiplexed signals are then transformed to be in the first and second polarization states (e.g., orthogonal to each other), respectively. The resulting signals are combined to form a combined signal for transmission. In that embodiment, among other advantages, only one optical multiplexer needs to be thermally controlled to achieve the proper spectral alignment.
In another embodiment, a passive optical device (e.g., an AWG) is configured to multiplex the first and second sets of optical signals in a bidirectional manner. The optical device has first and second sides thereof for receiving the optical signals in the first and second sets, respectively. The optical signals in the first set converge onto a first output on the second side of the device, thereby providing a first multiplexed signal at the first output. At the same time, the optical signals in the second set converge onto a second output on the first side of the device, thereby forming a second multiplexed signal at the second output. The first and second multiplexed signals are then polarization-multiplexed to form a combined signal for transmission.
Each pair of polarized carrier signals, C-na and C-nb (1≦n≦N), from an optical splitter 107-n is routed to a pair of modulators, H- and V-modulators 109-na and 109-nb, corresponding thereto. For example, H- and V-modulators 109-na and 109-nb may be nested Mach-Zehnder modulators. 2N streams of data (e.g., at a combined bit rate on the order of 100 GHz) are provided to the N pairs of modulators where the N pairs of carrier signals are modulated with the respective data streams, e.g., in accordance with a conventional DQPSK scheme. The modulated optical signals of λnH and λnV are fed to multiplex section 110 which includes two N×1 arrayed waveguide gratings (AWGs) denoted 111a and 111b, respectively, in accordance with a WDM scheme. Specifically, AWG 111a collects the modulated optical signals of λnH which are co-polarized, and AWG 111b collects the modulated optical signals of λnV which are co-polarized. As is well known, AWGs are passive optical signal multiplexers. The channel spacing of AWGs 111a and 111b corresponds to the spacing between the lasers' wavelengths, i.e. Δλ. Each set of modulated polarized optical signals is introduced into the corresponding AWG through its injection ports on its input side, where the signals in the set enter the sequentially-numbered injection ports in order of wavelength, e.g., with their wavelength increasing with the injection port number. That is, the modulated signal of λnH (λnV) enters input port n of AWG 111a (AWG 111b). The injection ports of AWG 111a are optically coupled to the collector output port thereof such that the incident modulated signals converge onto the collector output port, resulting in a multiplexed signal—polarized WDM signal 114H—emerging therefrom. Similarly, the modulated signals of λnV are multiplexed by AWG 111b to form polarized WDM signal 114V at its output port.
WDM signals 114H and 114V are polarization-multiplexed by polarization combiner 120, resulting in PM-DQPSK signal 125 to be transmitted through a transmission medium, e.g., an optical fiber. In a well known manner, polarization combiner 120 at the first stage may make the polarization states of WDM streams 114H and 114V nominally orthogonal to each other, for example, using polarization rotators such as retardation plates. Combiner 120 at the second stage may merge the two orthogonally polarized streams, for example, using a polarization beam combiner. Alternatively, a power combiner may be used at the second stage, however at the cost of additional power losses.
The invention is premised upon a recognition that use of the multiple AWGs for multiplexing the respective sets of modulated polarized signals as in system 100 is not desirable. For example, the multiple AWGs need to be spectrally aligned, e.g., to the same frequency (wavelength) grid, defined by a standard promulgated by the International Telecommunication Union (ITU) in order for the system to work properly. For example, the most common frequency (wavelength) grid that is used for dense WDM (DWDM) is defined relative to 193.1 THz and extends from about 191.7 THz to about 196.1 THz, with 100-GHz spacing (see ITU-T G.694.1). While defined in frequency, the grid is also often expressed in terms of wavelength, in which case its wavelength range is from about 1528 nm to about 1564 nm, with about 0.8-nm channel spacing. For practical purposes the grid is often extended to cover the range from about 186 THz to about 201 THz and subdivided to provide 50-GHz and 25-GHz spaced grids.
In system 100 two thermal control circuits, e.g., thermo-electric coolers (TECs), typically are used to individually adjust the temperatures of AWGs 111a and 111b to tune their operating frequency ranges. By contrast, only one such TEC needs to be used for AWG 211 in system 200, thereby obviating the otherwise duplication of the tuning effort and incurrence of the cost of an extra TEC. Thus, an advantage of using a single AWG 211 in system 200 is that the transmission passband of this multiplexer can be aligned with the lasers' wavelengths by controlling a single grating. In addition, AWG 211 is much more compact than AWG 111a and 111b combined, and thus may be more easily co-integrated with the signal sources and/or modulators in modulation section 203 to form a single chip.
In
It is noteworthy that, unlike the signal paths leading to AWGs 111a and 111b in multiplex section 110 of
Referring back to
It should be noted at this point that in general, the size of an AWG is proportional to the number of ports on either side of the AWG, whichever number is larger. Thus, AWG 311, with each side having N+1 ports, is advantageously smaller than AWG 211, with its input side having 2N ports. As a result, AWG 311 is more conducive to co-integration with the signal sources and/or modulators, e.g., in modulation section 203 to form a single chip. In one embodiment, AWG 311 is selected to be used in multiplex section 310 because of its low level of optical crosstalk, less than 40-45 dB, possibly leaking from the injection ports on one side of the AWG to the injection ports on its other side, and feeding back to the modulators and the laser sources, thereby impairing their performance. In another embodiment, this optical crosstalk or leakage is further reduced by using a slightly larger AWG, denoted 311′, with an increased number (N+k, where k>1) of ports on each side thereof to better separate spatially the collector port from the injection ports.
The foregoing merely illustrates the principles of the invention. It will thus be appreciated that those skilled in the art will be able to devise numerous arrangements which embody the principles of the invention and are thus within its spirit and scope.
For example, in the disclosed embodiment, the collector ports of AWG 311 are arranged at an end of respective sides A and B so that WDM signals 114H and 114V may be conveniently fiber-tapped from a free end of the AWG, without interfering with any other connections (e.g. fibers or optical waveguides) leading to the injection ports. However, it will be appreciated that a person skilled in the art may design to have any port on each side of the AWG to be a collector port to suit his/her particular need.
In addition, the above embodiments disclose inventive techniques for polarization-multiplexing of signals modulated according to a DPQSK modulation scheme. It will be appreciated that a person skilled in the art may apply these inventive techniques to signals modulated according to a different modulation scheme, as well.
Finally, although system 200 and its components, as disclosed, are embodied in the form of various discrete functional blocks, such a system and components could equally well be embodied in an arrangement in which the functions of any one or more of those blocks or indeed, all of the functions thereof, are realized, for example, by one or more optical processors or devices.