Optical Switches and Communication
Gurjit Kaur and Pradeep Tomar
INTRODUCTION The optical network such as Wavelength Division Multiplexing which is also referred as wavelength cross connect which have capability to combine different optical signals into one optical fiber link at various wavelengths [1]. In recent years, it is conclude that by using various materials having high electro-optic effect in optical device will be appropriate for WDM network. These materials have special characteristics which can optimize the switching voltage and other important parameters in optical devices hence reduce the price of any network and will provide a reliable system. This kind of optical switching network can be used for many commercial purposes [2]. Many Cross Connect (OXC) switching devices have been proposed using 2x2 switches which are organized in a different multi-stage network which can be a good candidate in higher order switches [3]. As these switching device work as splitter and coupler can transfer signal from one input port to any other output port [4].Most of the switching operations are required to have high switching ability and more speed to make system more reliable [5,6].
It is known that on the basis of Mach-Zehnder interferometer, different 2x2 switches have been proposed [8]. In last past years, it seems that many designers have given different studies on the basis of different switch design. One of the major benefit of optical switching network is that it does not require any conversion of electrical signal data to optical signal data due to which routing of optical signal is become easier as it does not depend on data rate and data protocols. There are many other features of optical switching functions instead of electrical function as it will reduce the system components and hence increase the speed of system operations. Another parameters which will make network more reliable which is throughput also get increase and the operating power will reduce accordingly [5]. There is one more thing which we have to look after is effective cost of overall system which will reduce as we go with the optical networks. Now-a-days, backbone of any network is usually optical. Hence, it is required to improve and establish more optical networks. For improvement of any optical network, new hardware are being designed and various protocols have been studied to make the transmission more reliable and quicker. In most of the optical network, they are using electronic components which does not able to compensate with the speed of optical network due to which bandwidth get waste as they are able to use only small amount of bandwidth due to these components. Optical amplifier are the devices which amplify different wavelength simultaneously because of that the requirement of OEO regenerators get reduced. The main advantage of optical cross connect is that they does not require the conversion of electrical data signal to optical data signal as it switch signal optically. One more device is available and useful in large network which are Optical Add-Drop Multiplexers (OADM). OADM are simply use for adding or dropping any particular data signal from optical fiber channel [5,24]. Many solution have been proposed to establish optical light path. One of them is Optical Burst Switching(OBS) [24] which is capable of establishing the light path for a small interval of time sat for milliseconds. At the sender, it does not require to establish the data path utterly to begin the transmission of optical signal. In the optical switching network, the nodes have ability to buffer data during transmission if the required data is not switched already. Another proposed solution available is Optical Packet Switching (OPS). It is simply used to switch the user data optically. Optical packet switching is more faster and cheaper switching process to use as compared to traditional switching operation as it require less space to establish, low power and dissipates less amount of heat during transmission. Hence, by using OPS we can make our transmission more cheaper to maintain and much more reliable. Fig 1.1 shows the basic layout design of OPS [24]. And the third solution which has proposed is Generalized Multiprotocol Label switching(GMPL) is basically a switching operation such as network and space switching for time and wavelength as well as packet switching.
1.2 APPLICATIONS OF SWITCHES 1) Optical Cross-Connects: Optical switches are mainly used in optical cross connects to provide equipments to set up the light path or data path which may or may not be static it depends on the network. To reconfigure any optical network, OXCs play an important role in it. The optical switches are used in the OXCs to reconfigure and hence transmit new signal path in any network. Cross connects are usually handle large amount of wavelengths at a particular location such as hub or server. To route any optical signal or carrier, OXCs are the main element in the network topology as it has capability to optimize the transmission of signal path [17]. Optical cross connects can be very effective in large networks as it provide certain key features such as:
a) Service Provisioning: In large network, it is important to make the transmission automated so that when a time to deal with large amount of wavelengths at particular location should be error free and easy to implement without need of much expense. In this scenario, OXCs have capability to transmit data path in an automated way to avoid any further failure in the network.
b) Protection: To avoid any component failure and cut in the fiber link, cross connects are used to detect the failure and to find new route the data path quickly through it. This is one major requirement in any network so that it can avoid further loss of data transmit through path because of that failure which occurred in the fiber path of the optical network protection is one major key feature of cross connects require in any network.
c) Monitoring and loopback: Optical cross connect provide another function which is monitoring of optical data signal during transmission over data path. And it also allow a loopback function in which a optical signal can get back to the same node from where it had start in the beginning at the intermediate node.
d) Scalability : Cross connects provide large amount of scalability require in any optical networks. e) Wavelength conversions: With so many features like transmit signal from one input port to any other output port, it also have a capability of wavelength conversion.
The major benefits of using optical cross connects in any optical network is that they have ability to switch the optical data signal with high reliability, having low losses in output power and exhibit great uniformity of data signal as it does not depend on path length. Another important feature is that it does not interrupt any other optical signal path as it has capability to switch the data signal at desired optical data path. Recently in many switching network, cross connects are employing as an electrical core for switching operation in which the optical data signal first converted into electrical signal and then switched by using electrical devices and will again converted back into optical signal. This type of switching feature lead to many drawbacks. First, it require conversion at different stage due to which the cost of overall system will increases and also require extra equipments for conversion which also make system more complex and require more maintenance as well. Second, the speed of electronic switching devices can not able to compensate with the optical switching devices which lead to a major issue in transmission of any data signal [18].
2) Network Provisioning: To establish a new optical data route or need to modify the existing route in any network it require network provisioning. It is comparatively have capability to switch quickly which will require in reconfiguration for the desired request in few seconds. On the other hand, the manual process is very sluggish in reconfiguration of any path as it takes few weeks or maybe more. To increase the network flexibility, it require a high capability configuration switches so that they can respond promptly and in an automated manner to any particular services required in the network [18]. 3) Protection Switching: The main function of protection switching is that it conclude the nature and the origin of failure in any network so that it can rectify the issue occurred in the network as well as will inform the nearby nodes to circulate the problem occurred in the network to other nodes. A basic protection switching. The operation performed by protection switching effective but it slow down the speed of switching operation in the network as compared with the optical switching. A Protection switching require smaller port optical switches of 1x2 or 2x2 [19].
4) Optical Add/Drop Multiplexing: The basic requirement of optical add-drop multiplexers in any network is that we can easily add or drop any wavelength(optical signal) to or from optical transmission path through the use of OADM in a different wavelength channel. It allow us adding or dropping optical signal without any requirement of electronic processing. This feature of OADM provide a cost effective transmission in large optical networks as it handles the coming traffic accordingly [19].
1.3 OPTICAL SWITCH FABRICS The use of all-optical switching fiber is very beneficial in any transmission to switching as it expand the attempt of switching operation in optical network. One major advantage of all-optical switching devices is that they provide direct switching in optical domain, hence does not require any conversion of O/E/O for the switching operation which make network more reliable and also increase the speed of switching operation. Many all optic switching fabrics are being studied and some are still under research. Most of the applications which provide for these devices that are feasible enough, as they would have one or more results to be considered. So before going through the main specification of optical fabric technology available today, it first need to go through with some parameters on which we analyzing these optical switch. There is one parameter that is switching time which is very significant parameter for any application and is different for different applications. Hence, listed some other essential parameters of a switch are as follows [8]. 1) Insertion loss: It is the ratio of signal power which is lost due to the switch. This loss is generally measured in decibels and it should be as small as possible. In addition, the insertion loss of a switch must be same for the connections of all input–output port (loss uniformity). 2) Crosstalk: This is the fraction of the power at a particular output from the required input to the all other inputs power. 3) Extinction ratio (ON–OFF switches): It is the proportion of the output power in the on condition to the output power in the off condition. This fraction must be as large as possible. 4) Polarization-dependent loss (PDL): For both states of polarization of the optical signal if the loss of the switch is not equal, the switch is said to have polarization-dependent loss. It is required that optical switches should have low PDL. Some of the other parameters that are taken into consideration include reliability, temperature resistance, energy usage and scalability. The term scalability can be define as the ability to design switches with large number of port counts that perform sufficiently. It is a specifically significant concern [7].
1.4 OPTICAL SWITCHING TECHNOLOGIES 1) Opto-mechanical Switches: Optomechanical technology was the first commercially available mechanics for optical switching. A basic design is shown in Fir 1.4. The switching function is implemented by some mechanical means in optomechanical switches. The mechanical means used in optomechanical switches include directional couplers, prisms and mirrors. These mechanical switches consist high extinction ratio, low crosstalk, less fabrication cost low insertion losses and low polarization-dependent loss. These switches have switching speeds in the order of a few milliseconds, which may not be suited for many types of applications and can be adjust according to the specific requirement. They have a disadvantage which is the lack of scalability [6].
And most of the mechanical components require long-term reliability of some concern. Opto-mechanical switch constructions are restricted to 1X2 and 2X2 port sizes. Larger port counts can only be acquired by combining several 1X2 and 2X2 switches, but this may increases cost and downgrades performance of the switch. The opto-mechanical switches are generally used in fiber protection and very less number of port count wavelength add/drop applications [18].
2) Microelectromechanical System Devices: Although Micro Electro Mechanical system (MEMS) devices have been contemplate as a one of the category of opto-mechanical switches, they are proposed distinctly, especially because of telecommunications industry has shown extensive interest in them, but also because of they have perform much better than as compared with other optomechanical switches [18]. MEMS switches uses small reflective surfaces so as to redirect the light beams to a particular port at of neighboring reflective surfaces by either deflecting the light off to a port or by guiding the light beam straight to a port [19].
There are two MEMS approaches for optical switching: two-dimensional (2-D), or digital, and three–dimensional (3–D) or analog, MEMS. In 2-D MEMS, the switches are in digital form, as the position of mirror is bi-stable (ON or OFF), through which switch driving become very straightforward. A top view of a 2-D MEMS device having microscopic mirrors, which arranged in a crossbar configuration to acquire cross-connect functionality. Collimated light beams are propagate parallel to the crystal(substrate) plane.[20] When a mirror is operated, it will proceed towards the path of the beam and passes the light to one of the outputs, as it makes a 45 degree angle along the beam. This alignment also permits light to be passed along the matrix without striking a mirror. This additional operations can be used for inserting or deleting (dropping) optical channels (wavelengths). The settlement for the clearness of the mirror control in a 2-D MEMS switch can be optical loss.[5]
3) Electro-optic Switches: A 2x2 optoelectronic switch utilize a directional coupler whose coupling ratio can be altered in the coupling region by varying the refractive index of the material. Lithium niobate LiNbO3 is one of the common material used in designing switches . switch is constructed on a lithium niobate waveguide. The change of substrate’s index of refraction occurs when an electrical voltage applied to the electrodes. The change in the index of refraction distorts the light along the suitable waveguide path to the particular port [17]. The electro-optic switches have capability of changing its state eminently quickly, usually in less than a nanosecond. The limit of switching time is determined with the capacitance of the electrode configuration. Electro-optic switches are also reliable for any system, but they have to pay the cost of high insertion loss and possibly they are polarization dependence. At the cost of a higher driving voltage, switch can be polarization independent, which may limits the switching speed of the operation. Larger switches can be designed by integrating many 2X2 switches on a single substrate. However, they could have a comparatively high insertion losses and PDL and are much more expensive than mechanical switches [20].
4) Thermo-optic Switches: The functionality of these switches are based on the thermo-optic effect. It exhibits by varying the refractive index of a dielectric material, as there is temperature variation in the material itself. The thermo-optic switches are of two types categories as: interferometric switches and digital optical switches.
a) Interferometric switches: are mainly based on Mach–Zehnder interferometers. the devices comprise of a 3-dB coupler in which the signal will split into two beams, which then passes along two individual arms of same length, and the second 3-dB coupler, will merges the two splitted beams and finally splits the signal again. By heating one arm of the interferometer causes changes in its refractive index. Accordingly, arm is experienced a variation of the optical path of that. By heating one arm of the interferometer, it is then possible to alter the phase difference between the two light beams. Hence, as interference can be constructive or destructive, hence the power on alternate port can be minimized or maximized. Then output port will be selected [8].
b) Digital optical switches: are generally made of silica on silicon and these are integrated optical devices. The switch is made up of two interacting waveguide arms along which light propagates. To determines the output port, its required to calculate the phase error between the beams at the two arms. The refractive index of one of the arms get changes by heating, and the light is transmitted through one path in place of the other. a 2x2 digital optical switch. Thermo-optical switches are usually small in size but have some fault as it require high-driving-power characteristics and have difficulty in optical performance. This technology have some drawback which include high-power dissipation and limited integration density (large die area). Many commercially available thermo-optic switches entail forced air cooling for having reliable operation [7]. Some of the optical performance parameters, such as crosstalk and insertion loss, may not be acceptable for some of these applications. On the other side, this technology have some benefits it allows the variable optical attenuators design integration and also allow wavelength selective elements (arrayed waveguide gratings) design on the same chip with the same technology [18,7].
5) Liquid-Crystal Switches: The liquid-crystal state is having a phase that is arranged by a large number of organic materials with certain temperature ranges. the liquid-crystal phase, because of their permanent electrical dipole moment, molecules can accept a certain mean relative orientation. It can be possible, by applying a desirable voltage across a cell filled with liquid-crystal material, to act on the orientation of the molecules. Therefore, optical properties of the material can be modify. Liquid-crystal optical switches are depend on the change of state of polarization of incident light with a liquid crystal having the application of an electric field over the liquid crystal [8,18].
