Proposed in this paper is a novel packet-level self-synchronization scheme for slotted optical packet-switched networks (OPNs), where a clock denoting the beginning of each packet is extracted to perform address recognition, ultra-fast optical sampling, or payload delineation. The effect of self-phase modulation (SPM) in semiconductor optical amplifier (SOA) and the interfering characteristics of Mach-Zehnder Interferometer (MZI) are combined to make the proposal simple, fast and integrable, with no special treatment at wavelengths, polarization, intensity, or bit-period of the marker pulses.

In this paper, the all-optical logic exclusive OR using integrated SOA-based Mach-Zehnder interferometer with differential modulation scheme is proved to exceed the operation speed limit of the logic XOR with original modulation effectively through numerical analysis for the first time. The SOA Parameters, the signal parameters and the time delay are discussed and designed to optimize the switching performance of the proposed logic XOR. The operation of the proposed logic XOR with 80Gb/s RZ pseudorandom bit sequences is simulated with fairly high performance.

In this paper, we present a technique, using packet compressor, to perform both compression and decompression of optical time division multiplexed (OTDM) packets. The compressor is based on optical fiber delay line, and we design a mathematical model for the compressor. With this model, we discuss several questions associated with the compressor, such as Link Budget, Differentia Budget, and Effects of Time-out. At last, compression of 2.5Gbit/s 16bit packet to 100Gbit/s is experimentally demonstrated.

Optical gate switch is a vital part of all-optical signal processing and being investigated. A kind of optical gate switch, called ultrafast nonlinear interferometer (UNI) consisting of a SOA, which is used as a nonlinear component in the path of one-arm MZI, has the advantages of a simple structure and high speed. In this letter, we report a 10GHz all-optical switch based on an UNI. Through our experiment, we found that the switching window depends on many parameters, including the injection current of SOA, the power of control pulse, and the power of continual light.

We experimentally demonstrate ultra-high-speed (40 Gb/s, 80 Gb/s, and 160 Gb/s) optical signal transmission in point-to-point terrestrial systems and transparent networks built on cross connects and add-drop multiplexers. Without using forward error correction, we demonstrate 40-Gb/s transmission over 2000 km based on a simple system design, 80-Gb/s transmission over 1200 km, and 160-Gb/s transmission over 300 km, all with 100-km terrestrial type transmission span. In addition, a new approach to generating phase-correlated ultra-high-speed signals is introduced. We also demonstrate 40-Gb/s transmission through wavelength-selective optical cross connects spaced by 400 km, as well as 40- and 160-Gb/s signal transmission through a new type optical add-drop multiplexer that can adapt to different data rates.

An evident model for high-speed wide-bandwidth Raman-amplified WDM system simulation is introduced, and the routines to realize numerically this model are also detailed. Using this model, the major physical effects in Raman-amplified WDM systems are simulated and discussed. Comparison with VPItransmissionMaker software shows good agreements.

The wavelength-division multiplexing (WDM) passive optical network (PON) is one of the promising access network systems that can satisfy cost-effectiveness, inter-operability, and providing simultaneously of various services. Semiconductor optical amplifier (SOA) transceiver is an attractive devices achieving inter-operability which corresponds to wavelength independency in each optical network unit (ONU). This paper reviews the configuration of the WDM access systems and the performance of the SOA transceivers applied to the systems.

Ultra-high speed optical transmissions require the knowledge of pulse broadening due to higher order dispersion. In this paper, we report the pulse broadening created by Lorentzian input pulse and it is compared with the pulse broadening created by Gaussian input pulse. The pulse broadening due to higher order dispersion terms are obtained for both Lorentzian and Gaussian input pulses. The output optical pulses created by Lorentzian input pulse are compared with the results obtained from Gaussian input pulse. This is followed by the estimation of intersymbol interference (ISI) and the maximum transmission length for Lorentzian input pulse is obtained. It is shown that intersymbol interference strongly depends on pulse profile. Further, the maximum transmission lengths are obtained for both Lorentzian and Guassian input pulses.

In this paper a novel online chromatic dispersion monitoring method employing the spectral shift in the semiconductor optical amplifier (SOA) is proposed and experimentally investigated. The power of the long wavelength side of the output signal through the SOA is extracted by a narrow bandwidth filter, and is utilized to measure the dispersion of the signal. The experiment shows that the ratio of the filtered power to the total power decreases 10 dB while the absolute value of the dispersion increases 700 ps/nm, with the input average power kept constant.

In this paper, based on the proposed simulation model, the impact of dispersion on the systems performance in 160Gbit/s optical communication systems is detailedly discussed. As chromatic dispersion (CD) and polarization mode dispersion (PMD) are caused by different physical effect, the effects of CD and PMD on systems performance are also analyzed separately in the paper. At first, the CD and CD slope tolerance in 160Gbit/s systems are carefully simulated and analyzed, at the same time, self-phase modulation (SPM) effect is also discussed in this part, and the result shows that for 1dB eye-opening penalty, CD tolerance would be about ±3.1ps/nm and CD slope and SPM also all seriously limit the transmission distance of systems. Furthermore, the impact of PMD on systems property are discussed by virtue of statistical method, and result shows that PMD tolerance of 160Gbit/s is 0.7ps with the outage probability of 10^-5 of 1dB eye-opening penalty and it will increase to 1.8ps for the first-order PMD compensated systems. These results show that CD, CD slope, PMD and SPM all are key challenges in 160Gbit/s systems.

Based on the proposed polarization mode dispersion (PMD) compensation simulation model and statistical analysis method (Monte-Carlo), the critical parameters initialization of two typical optical domain PMD compensators, which include optical PMD method with fixed compensation differential group delay (DGD) and that with variable compensation DGD, are detailedly investigated by numerical method. In the simulation, the line PMD values are chosen as 3ps, 4ps and 5ps and run samples are set to 1000 in order to achieve statistical evaluation for PMD compensated systems, respectively. The simulation results show that for the PMD value pre-known systems, the value of the fixed DGD compensator should be set to 1.5~1.6 times of line PMD value in order to reach the optimum performance, but for the second kind of PMD compensator, the DGD range of lower limit should be 1.5~1.6 times of line PMD provided that of upper limit is set to 3 times of line PMD, if no effective ways are chosen to resolve the problem of local minimum in optimum process. Another conclusion can be drawn from the simulation is that, although the second PMD compensator holds higher PMD compensation performance, it will spend more feedback loops to look up the optimum DGD value in the real PMD compensation realization, and this will bring more requirements on adjustable DGD device, not only wider adjustable range, but rapid adjusting speed for real time PMD equalization.

In this paper, with the superimposed technique a 4-channel chirped grating in a fiber for multi-channel dispersion compensation is been written. We have performed a dispersion compensation experiment with a system using a 10Gbit/s externally modulated laser. The experimental results show that measured BER of the all channels has the power penalty less than ±0.3dB when the grating is inserted through a circulator in front of the 100km transmission fiber.

As a promising solution for the next generation optical Internet, optical burst switching still has much to be improved, especially on the design of core routers. This paper mainly focuses on the channel scheduling algorithms of core routers and proposes a new scheduling algorithm. The new scheduling algorithm integrates burst segmentation, one of contention resolution schemes that are another major concern in the core router design. The simulation results show that the new algorithm, compared to existing scheduling algorithms, can effectively lower the packet loss probability and is of great benefit to the network performance.

Optical burst switching (OBS) is a switching technology proposed to efficiently support the growing broadband traffic over all optical WDM networks. In this paper, we analyze the blocking performance of an OBS network with limited number wavelength conversion, where only a fraction of input wavelengths can be converted to a set of (but not all) output wavelength. We have derived exact analytical results for the blocking probability for the case of exponentially distributed burst size. Analytical and simulation results indicate that by using wavelength conversion, we can improve the burst blocking performance evidently. These results also indicate that, by using limited number wavelength conversion (for example, the number of wavelengths that can be converted is seven-eighths of the total number of wavelengths), we can achieve almost the same performance as that of using full wavelength conversion.

Micromechanical optical switch has a good quality for free-space optical cross connects, particularly in terms of the low insertion loss, low crosstalk, low polarization dependent loss (PDL), wavelength-independence, and bit-rate transparency. In research of micromechanical optical switches, the electrostatic torsion beam actuator is widely used. But this actuator needs high-applied voltage. In this paper, the actuator with the slant counter electrode was proposed and fabricated with tilted 3° (111) silicon wafer.

A novel self-aligning SIOB (SIlicon Optical Bench) for fully passive alignment of optical components is proposed for integrated transceiver modules. The coupling efficiency from laser diode directly to fiber is about 8% in theory because of severe mode mismatch and large NA difference. To pursue high coupling efficiency, a ball lens is added between laser diode and fiber. The calculations and simulations are done to optimize the optical system according to optical component specifications. The coupling efficiency is up to 45%. We designed the layout and fabricated the SIOB in (100) silicon wafer to obtain the demanding cavities by MEMS technology.

Hybrid hierarchical optical cross-connects enhance the performance/cost ratio of optical networks by providing transparent (optical) switching of sets of wavelengths (wavebands) in addition to opaque (electrical) switching of individual wavelengths. As the network bandwidth is getting cheaper, and performance bottleneck is moving to switching nodes, these systems provide an attractive scalable solution for next generation optical networks. We describe key technological components (including flexible non-uniform wavebands) of hybrid hierarchical optical cross-connects and discuss their performance/cost implications.

Optical MEMS (Micro-Electro-Mechanical Systems) is the integration of mechanical elements, sensors, actuators, and electronics on a common silicon substrate through microfabrication technology. And Optical MEMS has created a new fabrication paradigm for optical devices and systems. Using tools developed for the integrated-circuits industry, Optical MEMS brings unprecedented levels of miniaturization and integration to optical communication systems. Today, the capacity of optical communication networks is limited by the switches that are electronic in their core. Optical MEMS is transforming the telecommunications infrastructure by providing all-optical switches that enable high-capacity networks. Optical MEMS devices are driving the trend toward all-optical telecommunication networks. With the ability to directly manipulate an optical signal, MEMS have several applications that eliminate unnecessary optical-electrical-optical (O-E-O) conversions. In optical communication, electrical components such as inductors and tunable capacitors can be improved significantly compared to their integrated counterparts if they are made using optical MEMS technology. With the integration of such components, the performance of communication circuits will improve, while the total circuit area, power consumption and cost will be reduced. In addition, the mechanical switch, as developed by several research groups, is a key component with huge potential in various microwave circuits. Optical MEMS technology is currently used in low- or medium-volume applications. There are also some of the obstacles preventing its wider adoption. Such as a mechanism giving smaller organizations responsive and affordable access to optical MEMS fabrication. And the packaging of optical MEMS devices and systems needs to improve considerably from its current primitive state. Currently the designer of an optical MEMS device requires a high level of fabrication knowledge in order to create a successful design. This paper presents the technology of optical MEMS and the application of this technology in the optical communication.

Recently, the k-ary n-cube direct interconnect network is considered to construct high performance switching fabric for Internet routers. Routing algorithm, which specify how packets can be sent along switching elements in network, are crucial for the high throughput and low latency of switching fabric. For maximum performance, the routing algorithm should exhibit the following important features: avoidance of deadlocks, ability to work well under varying traffic patterns and supporting QoS (Quality of Service). In this paper we present a recovery-based minimal routing algorithm to reach such targets. As with recover-based approach, the detection scheme has significant impact on the performance. A simple and efficient recover scheme is proposed in this paper, queue length and timer-based deadlock detection scheme (QTDS). Then, QTDS-based routing algorithm is proposed (QTDS-R). QTDS-R is applied on a three dimensional (3D) torus architecture. Changing the size of the fabric, the length of packets and the number of virtual channel, the performance of QTDS-R is measured in terms of throughput, average packets latency. The performance comparisons have been done on OPENT simulation tool. Numerical results show that this algorithm is better than other existed adaptive algorithms.

We demonstrate a novel modulation scheme based on a combination of PolSK and duobinary encoding called duobinary PolSK, which is used for the long-haul optic transmission. We present a comparison between NRZ intensity modulation and duobinary PolSK modulation. Results show that duobinary PolSK can become an interesting candidate to go beyond the limits of the transmission distance and bit rate in standard NRZ systems.

A novel all-optical NRZ-to-CSRZ converter using an SOA-based fiber loop mirror is demonstrated in this paper for the first time. The converter is operated at 10Gb/s and has the potential to work at 40Gb/s. This converter is optimized by adjusting the optical power ratio of control light and signal light respectively and its performance is evaluated. The minimum power penalty of the converter is measured to be -0.8dB at -13dBm input signal power for NRZ-to-CSRZ conversion and the dynamic range of input NRZ signal is 8dB (from -18dBm to -10dBm) for 0dB power penalty at 10^-10 BER. Format conversion between RZ and CSRZ is also realized using such a converter.

This paper presents a new optical Minimum-Shift Keying (MSK) modulation scheme with high spectral efficiency. The transmitter is realized in an Offset-Quaternary Phase-Shift Keying (QPSK) arrangement. The receiver uses a delay & add filter (DAF) and photodiodes for direct detection. An MSK precoder for the transmitter is derived.

In this paper, an overview of promising solutions for new modulation formats in optical WDM-transmission is presented and some recent results are summarized. We compare these formats with the standard conventional intensity modulation scheme. For real world applications it is necessary that these formats not only offer bandwidth reduction but also prove sufficient performance with respect to interference and intermodulation from adjacent channels, non-linear fibre properties, chromatic and polarization mode dispersion and noise resistance. Last but not least the implementation-effort at high speed must comply with economic constraints.

In this paper a four-channel OADM based on fiber gratings with smart temperature stable sets was implemented. It solved the conflict between the tunable capability and the temperature stability of the center wavelength. Using a cantilever beam method, the tunable range of 1nm is obtained. Within the temperature range -20°C ~+ 60°C, the center wavelength shift is less than 0.004 nm/°C and the thermal stability is 6.75 times as that without thermal stable sets. The wavelength spacing is 0.8 nm, which accords with ITU-T G.692 recommendation. 35 dB of the adjacent isolation in the OADM is obtained, which is the best one reported as we known.

The diffraction characteristics are analyzed for a polymer arrayed-waveguide grating (AWG) multiplexer around the central wavelength of 1.55 μm with the wavelength spacing of 1.6 nm. The diffraction loss and diffraction efficiency in the input and the output slab waveguide are investigated and discussed for different values of parameters, such as the core width, pitch of adjacent waveguides, the number of arrayed waveguides, taper end width of waveguides, and number of output wavelength. Finally, we give a set of parameters which have been optimized in this device.

Optical add/drop multiplexers(OADMs) are finding their applications in future all-optical communication networks to link the network with local transmitters and receivers. A novel fully dynamic and reconfigurable OADM based on AOTF with low crosstalk is presented. The PMD compensation issue of AOTF is addressed as well.

Optical Burst Switching (OBS) provides statistical multiplexing capabilities at the optical layer with relaxed hardware requirements when compared to optical packet switching. Burst blocking probability of OBS networks is a very serious problem without efficient wavelength converter or optical buffer so that OBS is not available in the near future. We proposed a new switching architecture and technical implementation issues. The technical features of the proposed network are use of ring topology, one-way reservation, out-of-band transmission of the burst packet control signal, optical burst transmission unit, loss-free transmission of burst, small delay jitter, and dynamical intelligent wavelength add-drop multiplexer (DI-WADM) module. Our proposed switching architecture and technical implementation issues can make most use of both optic and electric technology and provide a good resolution of optical burst packet contention with our proposed access protocols. The implementation frameworks of the proposed network and node are simple.

Reconfigurable Add/drop Multiplexer (ROADM) is a broad definition of a functionally reconfigurable filtering device for dynamic networking. We focus on a class of ROADM architecture that allows scalability of wavelength channels, add/drop port counts as well as functional capability of integrating variable attenuation and monitoring elements. We demonstrate the proposed concepts of integrating a ROADM with a variable optical attenuator and a performance monitor array.

In this paper, we study the impact of noise constraints due to the accumulated Amplified Spontaneous Emission (ASE) noise for partially shared buffering (PSB) optical packet switches under self-similar traffic. Noise constraints have a severe impact on the performance of PSB optical packet switches. We then propose a new priority switching algorithm that can effectively alleviate the performance degradation caused by the noise constraints on re-circulation limits in WDM PSB type OPS.

In this paper, we investigate the somewhat untraditional approach of contention resolution in WDM optical packet switches. The most striking characteristics of the developed switch architecture are that (1) contention resolution is achieved by a combined sharing of fiber delay-lines (FDLs) and tunable optical wavelength converters (TOWCs); (2) FDLs used for contention resolution is in non-degenerate form, i.e., buffers are achieved by non-uniform distribution of the delay lines; (3) TOWCs just can achieve wavelength conversion in partial continuous wavelength channels, i.e., sparse wavelength conversion. We describe and analyze the concrete configuration of FDLs and TOWCs under non-bursty and bursty traffic scenarios. Simulation results demonstrate that for a prefixed packet loss probability constraint, e.g., 10^-6, the developed architecture provides a different point of view in the optical packet switching (OPS) design. That is, combined sharing of FDLs and TOWCs can, effectively, obtain a good tradeoff between the switch size and the cost, and TOWCs which are achieved in sparse form can also decrease the implementing complexity.

It has been proved that through burst segmenting, and thus dropping the segment of one burst that is overlapped with another burst, the packet loss probability will be significantly improved for optical burst switched networks. Based upon this, two main segment dropping strategies, head- and tail-dropping, can be used to resolve contention. In this paper, we point out that tail-dropping policy, which is adopted in most literatures, may not be a feasible solution, while head-dropping policy, even though it may make the packets to arrive out of order, is in deed a solution which can resolve the contention effectively. Moreover, with respect to tail-dropping policy, the increase of the disorder of packet arrivals for head-dropping policy is trivial. To support service differentiation and decrease the disorder of packet arrivals generated by contention resolution, we further propose a new threshold-based hybrid-assembly scheme. The most striking characteristic of the hybrid-assembly scheme is that the low and high packet classes are aggregated into one burst simultaneously. Once contention occurs, head-dropping policy is adopted to drop the overlapping packets. We describe the concrete implementation of QoS supporting, and the corresponding dropping strategy -- improved head-dropping policy, which aims at guaranteeing a better QoS support and a feasible implementation, is also detailed. Simulation results demonstrate that the proposed burst assembly scheme, together with the head-dropping policy, perform well in terms of performance metrics such as the average packet loss probability and service differentiation.

This paper introduces the scattering matrix for Total Internal Reflection (TIR) electro-optic switch and Spanke switch fabric. The transfer functions of the switch and switch fabric are derived through Mason's gain formula. The scattering matrix of the switch fabric is formed using star product of switch element's scattering matrices. These transfer functions can be made to analyze the responses of the switch and fabric. Finally, generic model for transfer function of 'n' port switch fabric is proposed.

This paper introduces a novel wavelength switch based on the grating light valve (GLV) technology, which has the advantage of high speed, high reliability, low cost and compactness. A theoretical model is presented on this wavelength switch and the configuration parameters are calculated. The insertion loss is found to be reasonably low, leading to the practical significance of this novel wavelength switch.

Optical transmission systems made impressive progress in the field of system capacity, reach and flexibility. Beside all improvements achieved so far, signal distortion on the one hand side and noise accumulation on the other hand side are the limitations of optical transmission system performance. In terms of signal distortion besides advanced link design strategies optical regenerators are of high interest for signal conditioning along a fiber link. Our paper reviews signal regeneration techniques focusing on fiber nonlinearity based setups both in interferometric and non interferometric architecture. We analyze the impact of regenerator design parameters as fiber nonlinear parameters, regenerator input power and characteristics of filters on signal improvement and demonstrate e.g. the increase of system reach. In order to address the upcoming techniques of advanced modulation formats we introduce a fiber based regenerator scheme that shows attractive potential especially for phase modulated signals. Combining these new optical formats with regenerative methods is attractive for high performance systems. For the differential phase shift keying (DPSK) modulation format we investigate the potential of a cross phase modulation based type of regenerator in more detail. The availability of suitable regenerator components, especially highly nonlinear fibers is one of the challenges. Especially the new fiber types known as photonic crystal fibers can act as bases for new structures and thus will open the door for ultra high performance optical transmission systems.

This paper reported an improved optical switching network configuration based on optical interconnection technology with vertical cavity surface emitting laser (VCSEL) array. The optical switching network consists of two-level optical interconnection backplane. It can connect 64 nodes with parallel optical links. The first level of optical interconnection backplane includes eight 8×8 crossbar interconnect sub-networks. Instead of one 8×8 crossbar interconnect sub-network in the second level of the optical interconnection backplane adopted in our original configuration, the second level of optical interconnection backplane has two 8×8 Crossbar interconnect sub-networks in this improved configuration. So the blocking rate is decreased. VCSEL-based parallel optoelectronic I/O interface is used as O/E conversion. Every I/O parallel interface between optical interconnection network and every node includes 18 VCSEL emitter pixels, 18 PIN receiver pixels. In order to couple 18 signal light beam array into optical fiber array ribbon, a fabrication technique based on the high precise position slot is used for assembling optical fiber array interface. A configuration of coupling packaging for the VCSEL pixel array to the fiber array with 45° end surface is also presented in this paper. An optical data transmission rate between interconnection nodes is 5Gb/s which is transmitted by the optical fiber ribbon-based parallel optical data links with 2 channels at data rate of 2.5Gb/s per channel. The aggregate bandwidth of 360Gbps for an 8×8 Crossbar optical fiber interconnect network backplane is achieved. The reliability of the fiber array with 45° end surface is tested in our experiment.

This paper investigates the implementing techniques concerning 2-to-1 node receiver for all-optical packet switching. We point out that the state-of-the-art technologies introduced in the literatures are all associated with one main limitation, i.e., even if two incoming packets are directed to the different ports and thus no contention will occur, they can not be directed to the output ports simultaneously. To overcome this problem, a novel node receiver model is proposed in this paper. This work presents a solution that makes use of fiber delay-lines (FDLs) which allow multiple packets to be concurrently stored or transmitted. With a novel switch control, it is shown that this solution is very efficient to resolve contention, and overcomes the limitation existing in the current models (e.g., Quadro, M-Quadro and COD architectures). We describe and analyze the concrete scheduling of the switches. Simulation results based on bursty and non-bursty traffic scenarios demonstrate that the proposed node architecture (1) performs well in terms of performance metrics such as packet loss probability; (2) has a simple control requirement. In addition, we find that the required number of the recirculating time of the packets is minimal.

The advantages and the background of the Optical Burst Switching networks are reviewed based on a comparison study with wavelength-routed networks and Optical Packet Switching networks. Then the key attention-drawing issues concerning further develop of the OBS networks are analyzed. Currently, lots of effort has been taken in developing proprietary signaling protocols for resource reservation in OBS networks, the framework of OBS core switching node and edge routers. On the other hand, some related topics, such as contention resolutions strategy, QoS support strategy, control and scheduling algorithms, and hardware implementation issues about Tunable Wavelength Converter, Fiber Delay Line based optical buffers, high speed optical switches, are all need further research. All these issues will be reviewed in this work. At last, some problems not yet been studied in current works are presented, and further research directions are proposed.

As data rate in optical fiber communication system increases to 40Gbit/s in single channel, the impact of polarization mode dispersion (PMD) on system performance becomes more important in high speed and long distance transmission system. This paper studies on the system penalties induced by PMD with different electrical bandwidths of receiver for four different modulation formats such as NRZ (not return to zero), RZ (return to zero), CSRZ (carrier suppressed return to zero) and duobinary. We find that the system penalties induced by PMD show a strong dependence on both the electrical bandwidth of the receiver and modulation format.

Theoretical calculations are made on the basis of relationship between electrical power and differential group delay (DGD). Curves are given representing the change of electrical power with DGD, central frequency of electrical signal and power ratio γ between principal states of polarization (PSPs). Experimental results confirming the theoretical calculations are also presented for voltage V changing with DGD. Factors affecting the feedback signals are proposed and means are given to make the fluctuations smaller. Eye patterns show the effectiveness of a PMD compensation prototype for a 10Gbit/s pseudo random sequence. Receiver sensitivity under 3 different bit error rates are also measured using BER measuring equipment. Comparisons of results under different conditions are also made, confirming the results and deterioration caused by the simulator is well compensated for by using the compensation system.

Polarization effects such as polarization mode dispersion (PMD) and polarization dependent loss (PDL) have become the critical limiting factors for long haul high bitrate optical fiber transmission. This paper use Monte-Carlo numerical simulations to investigate the statistical performance of the PMD compensator in the 10Gbit/s NRZ optical transmission link with PMD and PDL, and the PMD compensator use the tunable SBG-FBG as differential delay difference (DGD) compensation element and output optical signal degree of polarization (DOP) as the feedback control signal. Statistical simulation results with the outage probability of the compensated signal eye-diagrams opening (EO) indicate the PMD compensation performance significantly decreases in the presence of PDL.

The impact of XPM on the 1^st-order PMD compensator is numerically simulated for a two-channel WDM system with the bit rate in each channel of 40-Gbit/s. Simulation results show that a 1^st-order PMDC does not change the distribution of the remaining EOP in a single channel system; but in a WDM system, the EOP distribution will disperses over the whole EOP range and the occurrence probability of larger EOPs increases heavily due to the XPM effect when a 1^st-order PMD compensator is used to compensate for the DGD of the preceding fiber link. Furthermore, in WDM systems, the larger the input average power in the pump channel, the severer the impact of XPM on 1^st-order PMD compensation is. When the XPM-inducing input average power reaches 0dBm, the impact of XPM on the 1^st-order PMD compensator is visible.

In wavelength-routed WDM networks, all-optical connection (or lightpath) is required to be established between communication entities before data transmission. In order to establish and take down such a connection, we need control mechanisms to select a route, assign a wavelength, and reserve resource to it. These mechanisms are expected to be scalable and be able to support highly dynamic and bursty traffic, meanwhile with a low degree of blocking for connection requests. In this paper, we propose to introduce a distributed control protocol for lightpath establishment in wavelength-routed WDM networks, which utilize a hybrid resource reservation scheme from Source Initiation Reservation (SIR) and Destination Initiation Reservation (DIR), and an efficient data transmission scheme, namely Conditional One-way. A simulation is developed to evaluate performances in terms of blocking probability and lightpath setup time, and results show that our protocol outperforms other approaches.

We consider the problem of routing and wavelength assignment problem (RWA) in wavelength-routed networks, which consists of the Wavelength Division Multiplexers (WDM) and optical cross-connect (OXCs) devices interconnected via point-to-point WDM links. Assume that none of the OXCs has wavelength conversion capability. The paper shows that the problem of RWA can be solved simultaneously by employing an efficient graph theoretic formulation, named as multi-layered graph model. On the basis of this notion, a heuristic RWA algorithm for static case is proposed. The numerical examples confirm that the proposed approach for static lightpath establishment (SLE) can obtain better solution for the number of requested lightpaths. The paper also considers the dynamic lightpath establishment (DLE) problem. We propose another heuristic DLE algorithm based on alternate-lightpath routing, according to a global selection criterion. We make a comparison of blocking probability according to different selection criterion. The performance of the proposed DLE algorithms is also evaluated and compared with that of the conventional algorithm via simulation of an arbitrary mesh network, where lightpath requests are assumed to arrive according to a Poisson process with exponentially distributed holding times.

A detailed analytical model for hot-potato ShuffleNet with limited wavelength conversion is presented in this paper for the first time. Network simulations are performed to validate the results of theoretical analysis. The analytical and simulation results show that the performance of ShuffleNet under limited wavelength conversion improves with wavelength conversion degree d_w, and the efficiency of contention resolution per wavelength conversion degree decreases largely with increasing d_w. ShuffleNet performs well with limited wavelength conversion and can achieve performance close to that of full wavelength conversion adopting a wavelength conversion degree as small as 8 for nodes equipped with 16 wavelength channels. The results presented in this paper are interesting and valuable in the designation of the other kind of multi-path optical packet networks.

In this paper, we propose four different RWA schemes which are introduced into the wavelength-routed intelligent optical networks. FSP+FF and FPLC+FF are traditional RWA schemes used in the centralized control environment. The other two schemes, FWR-4 and BWR-3, are distributed wavelength reservation protocols which invoke the GMPLS REVP-TE. These schemes are emulated in GMPLS-based NSFNET. The results are compared in terms of four performance metrics which are blocking probability (BP), route length (RL), setup time (ST) and control overhead (CO). Except for FWR-4, the BPs of the other three RWA schemes are similar, and in the case of lighter traffic load the BP of FPLC+FF is the smallest, but under heavier traffic load the BP of BWR-3 is the best. Because FPLC+FF does not always choose the shortest path compared with other RWA schemes using the shortest path routing algorithm, so its RL is the longest. The route and wavelength are determined by the source node, and the RL is shorter, therefore the lightpath setup in FSP+FF is the fastest among all schemes. In addition, the CO of FWR-4 is the fewest because it need not flood the routing information and retry. Though the four schemes respectively have their advantages, the performance of BWR-3 is outstanding among them if its ST can be accepted.

The technology of WDM is widely used in Optical Transport Networks. The rapid growth of data service demands a higher quality of the capacity of transmission equipments, network bandwidth and the ability of multi-service transmission. This paper introduces the application of 1.6Tbit/s WDM Multi-service Transmission Platform of Fiberhome Telecommunication Technologies Co., Ltd. (FT) in the backbone transmission network of China Unicom in Jiangsu Province, and analyzes the key technology of RAMAN/EDFA Fiber Amplification, GEthernet interface, Sub-data-rate Multiplexing, Data Wrapping and Forward Error Correction (FEC) and administration of networks. The present writers believe that IP over WDM (optical) will be more widely used in the future.

An implementation method of a core SONET/SDH switch with high capacity is introduced in the paper. High-speed serial I/O, switching architectures and design considerations for switching unit are involved. It supports strictly non-blocking for unicast traffic and re-arrangeably non-blocking for dual-cast. Dualcast traffic allows for efficient scheduling of working and protection paths in UPSR(Unidirectional Path Switched Ring)/BLSR (Bidirectional Line Switched Ring) applications.

In this paper we introduce three kinds of experiment that have been carried out to check how different signal modulation formats and fiber types behave in Ultra-Long-Haul system. From the test results, two helpful conclusions have been deduced. Under the experiment conditions presented in this paper, SMF-fiber's system can extend 600km longer than E-LEAF-fiber's and RZ system can extend 900km longer than NRZ system.

This paper considers the problem of Routing and Wavelength Assignment(RWA) in a reconfigurable optical network using wavelength division multiplexing (WDM). The RWA problem consists of two basic sub-problems, namely, routing problem and wavelength assignment problem. We couple the two sub-problems: the paths are decided by the state of the wavelength assignment. This is usually called dynamic (or adaptive) RWA. As Traditional static RWA algorithms only consider to have lower blocking probability, for high priority class light-path, they ignore to optimize the performance of low priority class light-path and result in higher blocking probability for low priority class light-path request. So the average of blocking probability is increased in all-optic networks. In fact, the most efficient service of optic network is for low priority class light-path request. In order to improve the performance of optic network, we must optimize low priority class light-path request. Our method is based on quality of service (QoS), using a layered-graph model and setting a threshold to assign the wavelength dynamically, this threshold decides whether to assign intensively the preferred wavelength in higher priority class wavelength. The value is tuned adaptively by nodes in all optic-networks with the usage of wavelength. Then the nodes select the routing dynamically according to the wavelength assigned. By using our method, the blocking rate of the high priority request is guaranteed to the lower degree, and at the same time performance of the low priority request is improved. So the network’s average blocking rate is minimized. Simulation results of the network performance showed that the proposed method performs well in different dynamic traffic situation.

A fast IPv6 route lookup scheme implemented by hardware is proposed in this paper. It supports a fast IP address lookup and can insert and delete the prefixes effectively. A novel compressed multibit trie algorithm that decreases the memory space occupied and the average searching time is applied. The scheme proposed in this paper is superior to other IPV6 route lookup ones, for example, by using SRAM pipeline, a lookup speed of 125 x 10⁶ per second can be realized to satisfy 40Gbps optical link rate with only 1.9Mbyte consumption of memory space. As there is no actual IPv6 route prefix, we generate various simulation databases in which prefix length distribution is different. Simulation results show that our scheme has reasonable lookup time, memory space for all the prefix length distribution.

The IEEE 802.17 Working Group has taken another step forward in providing standards to Resilient Packet Ring (RPR) architectures by approving draft 2.4 of the RPR standard. Some companies have already had their RPR equipments for good market perspectives before the final standardization of RPR specifications. These equipments can supply services such as Ethernet Private Line (EPL), Transparent LAN Service (TLS), Ethernet Line Aggregation (ELA) and so on. Quality of Service (QoS) is an important part implemented in these equipments. Only when there is promise of QoS for RPR equipment, can commercial clients put their key applications to its network. Both the edge and the ring of RPR equipment need take QoS into effect to get QoS guarantees. This paper mainly deals with edge QoS study of RPR equipment. Policing, classification, shaping, and scheduling that are used for edge QoS process are introduced in the paper. And a QoS process method for traffic across rings is discussed at the same time.

The paper discusses integrated QoS multicast routing and wavelength assignment algorithm in IP/DWDM Internet. Given a QoS multicast request with the bandwidth requirement, the multicast end-to-end delay and the delay variation bounds, we propose an algorithm, which can find a QoS-based multicast tree satisfying the above three constraints. The proposed algorithm constructs the multicast tree based on logical topology generated from physical topology. The logical topology is similar to the wavelength graph. Thus, the search process of the multicast tree means finding both route and available wavelength. By this way, the proposed algorithm integrates routing and wavelength assignment into a single process. We explain the correctness of the algorithm finally.

In this paper, a low-cost repeaterless transmission system is demonstrated at 10Gbit/s using two Sampled Chirped fiber Bragg gratings (SCBG) to realize the dispersion compensation of 210km transmission over G.652 single-mode fiber (SSMF). The two SCBGs are used as dispersion management components to mitigate the damage caused by the pulse broadening effort before the receiver terminal. Non-return-to-zero (NRZ) modulation format is adopted and EDFAs as boost amplifier and pre-amplifiers in this system. Low-gain Raman amplifier is used to improve the signal noise ratio (SNR) at the end of the transmission fiber.

The regenerator-free transmission distance of ultra-long-haul, dispersion-managed, optical WDM-transmission-links is ultimately determined by accumulating ASE-Noise of optical amplifiers and Polarisation-Mode-Dispersion (PMD) of the fiber-material. Therefore the applicability of the fiber-infrastructure of Deutsche Telekom to transport optical channels at 10 Gbit/s and 40 Gbit/s over sufficiently long distances is assessed both generally, applying PMD limit-values of ITU-T, and for a selected transmission link, evaluating numerical simulations using a worst-case simulation model.

The pressure on prices per bit and distance in operator networks requires an ongoing reduction of costs. In todays optical transport networks typical transparent transmission lengths are in the order of about 400 to 600 km. Then costly electrical 3R-regeneration is required. One way to deal with this cost-driver is to introduce ultra long haul (ULH) transparent WDM systems which are able to cover transmission length of e.g. up to 2000 km and thus decreasing the number of regenerative transponders. However, to maintain the flexibility of the network to add and drop traffic at required locations the use of optical add drop multiplexers (OADMs) or optical cross connects (OXCs) is mandatory in these systems, particularly for pan-European but even for European national networks. The design of such ULH-networks has to deal with a vast number of issues, in particular the physical layer design has to be very flexible and to allow for different optimization and adjustment schemes. In this paper based on computer simulations we will compare different approaches to reduce the design complexity of 10 Gbit/s ULH systems taking into account environmental constraints like availability of housing stations for different target distances.

Up to date Tsinghua University has almost taken part in all major/grand projects associated with optical networks which are supported by Nature Science Fund of China or by National High Technology Program during the 9^th and the 10^th National Five-Year-Plan. This paper presents a survey on these projects and their achievements.

Highspeed Optical wireless access system by using VCSEL single-beam and multi-optical-receivers system is proposed with a smart beam control mechanism. Sub-Gbit/s-data and sub-Mbit/s-control signals are simultaneously communicated to each other by using LD single beam.

Recently, optical differential phase shift keying (DPSK) has attracted much interest in high-speed optical transmission systems. It features constant intensity and thus is less vulnerable to the fiber nonlinearities. In addition to high-speed data transmission, it is desirable to carry relatively low-speed supervisory information simultaneously to facilitate the system monitoring and network management. In this paper, we propose and investigate an orthogonal modulation scheme for carrying relatively low bit-rate supervisory amplitude shift keying (ASK) data on a high-speed (10-Gb/s) optical DPSK data stream. The supervisory data is amplitude-modulated onto the constant intensity optical DPSK signal, thus no dedicated time slots are needed and this eliminates the bandwidth overhead. We investigate both experimentally and numerically the optimized operation conditions for such orthogonal modulation. The results suggest the design guidelines of the proposed ASK/DPSK orthogonal modulation for supervisory information dissemination.

A method for calculating the optical-signal-to-noise ratio (OSNR) of an optical performance monitor (OPM) is discussed. OSNR is one of the most import parameters determining DWDM system performance because of its dominance in determining the bit-error-rate (BER). However for the optical and electric components’ limitations, most OPMs are less accurate and dynamic range in measuring OSNR. In order to improve OSNR, heightening measuring accuracy of the signal power and the noise power are both required. The method for improving accuracy of measuring the signal power, namely Gauss Fitting, can heighten the signal power in the selected bandwidth by 1-2 dB. For improving accuracy of measuring the noise power, a kind of algorithm about deconvolution and signal recovery is discussed, which is used to sharpen the signal figure and eliminate effect of signal power to noise power (named cross talk). After deconvolution, the power of the pits of a 100 GHz channel signal can be depressed by 6-8 dB. In conclusion, a research on how to calculate OSNR in the OPM is discussed. By Gauss Fitting and deconvolution, the signal can be recovery efficiently and the result of OSNR can be heightened by 7-10 dB.

A new beam shaping method using cylindrical lensed fiber to change a beam with uniform angular divergence into an elliptical collimated beam with specific size is presented. One structure of the mechanical setting according to the beam shaping method is also provided.

A CCD signal acquisition system featuring a flexible and compact (Charged-Coupled Device) CCD driving mode based on a DSP device is designed for a new developed optical performance monitor (OPM) module. The design has prominent advantages, such as less power consumption, fewer and cheaper attached components and flexibility. The DSP-based system aims to implement the high accuracy conversion from successive video output signals of a CCD device to DSP Processor with least additional circuit components.

In the optical wireless communications systems, modulation schemes such as PPM, DPPM are often used because of its power efficiency. This paper presents a modified DPPM modulation technique for optical wireless communication. We study the properties of IDPPM and compare the power and bandwidth efficiency of OOK, PPM, DPPM and IDPPM. Also we analyse the packet-error probability on ideal distortionless channels with AWGN and examine the detection technology of the system. Finally we show the improving of performance using IDPPM modulation.

This paper compares the transmission properties of different modulation formats in 10Gb/s system through computer simulation. The modulation formats include the following four formats, they are return to zero (RZ), non-return to zero (NRZ), carrier-suppressed return-to-zero (CS-RZ) and optical duobinary. We compare these four formats in terms of the capability of combating nonlinear effects such as self-phase modulation (SPM), cross-phase modulation (XPM). And we also compare the chromatic dispersion tolerance of these four formats. It is shown that optical duobinary has wider chromatic dispersion tolerance relative to other formats. And CS-RZ has better effect on anti-nonlinearity.

In order to efficiently use the available optical bandwidth efforts are under way to achieve spectral efficiencies of at least 0.8 bits/s/Hz. For such spectrally efficient systems (considering co-polarized WDM-channels) the importance of filtering for different modulation formats, different channel bitrates from 10 - 160 Gbit/s and different dispersion map designs will be discussed. It turns out that the optimisation strategy for spectrally efficient WDM systems shows considerable differences to single channel systems. The specific modulation format and the chosen channel bitrate turn out to be of minor importance, thus enabling simple analytic expressions for the achievable system performance.

The method for the multi-phase computer-generated-holography (CGH) design of reconfigurable chip or board level optical interconnection is discussed in this paper. First, an improved direct search (DS) algorithm for the multi-phase CGH design is presented. Using this method, the high quality multi-phase CGH can be calculated very quickly, and the computation complexity is much lower than the other algorithms such as the most popularly used DS method and SA method. Then, how to use this algorithm for the CGH design of the reconfigurable chip or board level optical interconnection is given.

This paper focuses on the simulative optimization of modulation format, which is CS-RZ format with appropriate spectrum filtering, in a 42.7Gb/s spectrally efficient long-haul DWDM system. It is studied that the required bandwidth of the CS-RZ signal with spectrum pruning for channel spacing of 50GHz. The power loss and crosstalk are calculation under the Gaussian filter with various order and bandwidth. The influence on the CS-RZ waveform by the filter of various order and bandwidth was also investigated. Furthermore, the Q-value of the 42.7Gb/s CS-RZ signal was simulated before and after 640km fiber link transmission with 3 order Gaussian filter of various bandwidth, which simulative test system was designed as a 42.7Gb/s DWDM system with simultaneous C+L band discrete Raman amplifier by dispersion compensation fiber (DCF) as Raman medium. As a conclusion, the optimum filter bandwidth of 42GHz is obtained for the 42.7Gb/s 50GHz spaced DWDM system. Q-value of the signal is up to 8 and remains 6.7 after transmission of 640km fiber link, with the filter of the optimum bandwidth.

In this paper the performance of single channel 40 Gb/s non-return-to-zero (NRZ) signal is studied over standard configured dispersion-management transmission systems (1*80 km/6*80 km). Firstly, the optical and electrical filter bandwidth of receiver is optimized according to the power penalty. Then the nonlinear characteristic is considered. The pre- and post-compensating dispersion is discussed too. We get the conclusion that the optimal optical bandwidth will increase with the increase of SMF input power because of the fiber non-linear effects and the appropriate pre-compensating dispersion will improve the transmission performance and dispersion tolerance.

The transmission of high-bandwidth data over optical fiber is impaired by a variety of linear and nonlinear effects due to the inherent properties of the transmission medium. In this tutorial, we will explain how these physical effects affect the data transmission. We will also discuss the interdependence between these effects and give some rules of thumb to estimate their relative impact.

The main aspects of electromagnetic compatibility (EMC) are reviewed and a detailed EMC analysis of optical transceiver is given. This paper is focusing on the EMC problems and solutions of optical transceiver. Design rules and test results of the module are given to show good performances complying with the designing analysis.

We experimentally demonstrate multiple wavelength continuous wave (CW) sources based on longitudinal mode carving of the supercontinuum (SC) generated in optical fibers. The 10GHz SC with -20dB bandwidth of 63nm was generated with the actively modelocked fiber laser (AML-EDFL). By longitudinal mode carving of the SC we can generate more than 100 channels CW optical frequency chain.

In this letter, extending the normal OrCAD Pspice into electro-optical system of FLC, we construct an equivalent circuit model based on this principle; formulate the operation of the OASLM. Especially, we also show that the OASLM gray scale is realized not by FLC multidomain switching, but by uniform director rotation between bistable states. At the same time, the method could realize the dynamic switch simulation, memory characteristics and give the relation of memory threshold and critical pulse square, and show that this theory agrees with various experimental photographed trace results.

As is known, Ethernet has gained great popularity in LAN (local area network). Meanwhile, SDH/SONET is the backbone of today’s communication networks. Then, how to exchange information between Ethernets located in different regions over SDH/SONET is becoming a hotspot. This paper provides a single chip solution of Ethernet transmission over SDH/SONET (EOS), which supports up to thirty-two 10/100Mbit/s Ethernet ports or four 1000Mbit/s Ethernet ports over 2.5Gbit/s SDH/SONET. On designing of this EOS chip, we also present a packet loss policy called SDP (Selective Dropping Policy) to perform flow control, which can actively select packets to drop as the packet buffer overflows. Up to now, we have developed two policies of SDP: LPF (Longest Packet First) and LSPF (Least Significant Packet First). Simulation results show that with LPF, the average packet loss rate (PLR) of the overall system can be cut down significantly without increasing the buffer storage, and with LSPF, the packets of the least important group will be discarded discriminately in the case of overflow to lower average PLRs of other groups. Moreover, SDP can cooperate with Ethernet's Backpressure Mechanism fairly well and be implemented easily on chip.

A 5 to 10 Gbps bandwidth optical interconnecting and switching network system used on parallel computing is introduced in this paper. This system provides a high bandwidth to meet the request of high bandwidth of the parallel computing. Optics is used to be a media to carry data and optical crossbar interconnection board is used to switch data in this system. It comes over the inherent disadvantage of the R, L, C delay and clock skew of the electronics interconnection. This system has good stability and scalability.

In this paper, the transmission performance of soliton communication system using phase-sensitive amplifiers (PSAs) as in-line amplifiers was theoretically analyzed by computer simulation. For comparison, we also simulated the transmission performance of soliton system using erbium-doped fiber amplifiers (EDFAs) as in-line amplifiers. Our simulation considered both average-soliton regime and dynamic-soliton regime. The simulation results show that due to no ASE noise in PSAs, using them as in-line amplifiers in soliton system, the Gordon-Haus restriction can be overcome and soliton interaction can be greatly suppressed. For average-soliton, in the condition of no additional soliton control techniques, the soliton steady transmission distance can be lengthened greatly. For dynamic-soliton, PSA also has a good behavior of suppressing pulse broadening and soliton interaction. But after long distance transmission, the amplitude of soliton pulse descends significantly.

In the high-speed fiber communication system, the dispersion, non-linear effect, the performance of receiver and so on limit the further increase of transmission distance of fiber communication system. In order to reduce the effect of these adverse factors and improve the performance of fiber communication system, the FEC technology has been applied into the high-speed fiber communication system for further improving the quality of communication. The ITU-T stipulates that the basic FEC codes applied into the high-speed fiber communication system are BCH Code and RS Code. In this paper we give a detailed generalization and analysis of FEC technology, on the basis of analysis of some FEC technology and some present methods of implementing the function of FEC, we put forward the assumption of Advanced FEC and analyse its some performances in detail.

Nonlocal spatial solitons are the optical spatial solitons in nonlocal nonlinear media. In this invited paper, a review is given about the progress on this subject. The nonlocal spatial solitons are phenomenologically modeled by the nonlocal nonlinear Schrodinger equation, which can be simplified to a linear model for a strong nonlocality. An exact analytical Gaussian-shaped solution to the linear model is obtained, and a spatial soliton called an accessible soliton is found to exist. A phase shift of such an accessible soliton can be very large comparable to its local counterpart, and this finding might be applied in integrated all-optical devices. More intriguing is the collision between two nonlocal spatial solitons. The interaction of two coherent accessible solitons is always attractive, independent of their relative phase, unlike in a local nonlinearity. Such an interaction behavior is experimentally observed in nematic liquid crystals, and a power-dependent X junction, AND, NOR, and XNOR gates are also experimentally demonstrated. This collisional characteristics has implications to all-optical switching and all-optical logic.

Determination of blood glucose concentrations in diabetic patients is a frequently occurring procedure and an important tool for diabetes management. Use of noninvasive detection techniques can relieve patients from the pain of frequent finger pokes and avoid the infection of disease via blood. This thesis discusses current research and analyzes the advantages and shortages of different measurement methods, including: optical methods (Transmission, Polarimetry and scattering), then, we give emphasis to analyze the technology of near-infrared (NIR) spectra. NIR spectral range 700 nm ~2300 nm was used because of its good transparency for biological tissue and presence of glucose absorption band. In this work, we present an outline of noninvasive blood glucose measurement. A near-infrared light beam is passed through the finger, and the spectral components of the emergent beam are measured using spectroscopic techniques. The device includes light sources having the wavelengths of 600 nm - 1800 nm to illuminate the tissue. Receptors associated with the light sources for receiving light and generating a transmission signal representing the light transmitted are also provided. Once a transmission signal is received by receptors, and the high and low values from each of the signals are stored in the device. The averaged values are then analyzed to determine the glucose concentration, which is displayed on the device.

In recent years, more and more optical fiber cable has applied to the long haul trunk, and the optical network has become the backbone of communication network because of the rapid development of optical fiber communication. So higher requirement for management of optical cable network is brought forward. This paper analyzes the technique fo monitoring optical cable, gives an event analysis algorithm based on Gabor transform and minimum description length (MDL) criterion, and analyzes the performance of the algorithm.

According to the standard diffraction theory, apertures much smaller than the wavelength of electromagnetic waves transmit poorly. But it's observed experimentally that the transmission through arrays of subwavelength metallic structures exhibits peaks at some wavelengths. These properties implies novel applications including dynamic optical subwavelength filters, optic switches, and subwavelength optical fiber couplers. In this paper, we theoretically investigate the transmission of electromagnetic waves through subwavelength magnetic slit arrays. The calculated results show that the transmission can be enhanced at wavelengths without exciting surface polaritons. This enhancement is attributed to the slit-perpendicular surface current induced by the incident electromagnetic waves. Additionally, the position of the enhanced peaks shifts to longer wavelengths and the strength of the peaks exhibits more complicated properties with increasing the permeability of the magnetic material constructing the gratings.

In this paper, an off-axis holographic lens with the focal length of about 70mm is recorded in the azobenzene liquid crystal polymer film (azo-LCP) by 532nm YbVO₄ double-frequency lasers and is reconstructed by a 633nm He-Ne laser. The particularity and main advantage of the work consists in writing and reading out the hologram at the same place, without moving, post-exposure adjusting or chemical processing. The converging and imaging property of the holographic lens is measured and analyzed.

An experimental system used for the optical chaos communication with a semiconductor optical amplifier (SOA) based fiber ring laser as a dynamical light system is presented. The chaotic polarization dynamic of the light source is modulated through a cross phase modulation (XPM) effect in the SOA induced by the input light carrying the message. A chaos receiver based on a time-delayed configuration can establish the synchronization with the transmitter and the message is restored by a dynamical detection for the states of polarization (SOPs) of transmission lights. A fiber link of 1-km long was used as the communication channel. The round-trip time and SOPs of the ring in this system can be used as key parameters to establish the synchronization between the transmitter and receiver. The polarization dynamics of the system were investigated under different input optical powers, and an experiment for the optical chaos communication was carried out, in which the 16-bit RZ sequences at 140 Mb/s were successfully recovered from the chaotic carriers after 1-km long transmission.

This paper investigates turbo equalization (TE) with linear minimum mean squared error (LMMSE) equalizer, which has much lower complexity but much worse performance than TE with maximum a posteriori (MAP) equalizer because of inefficient information processing of LMMSE equalizer. Motivated by forward and backward recursions of MAP algorithm, three methods are presented to improve turbo equalization with LMMSE equalizer (TE-LMMSE). In the simulation, they all obtain more than 2.5-dB performance gain over TE-LMMSE with no or some complexity price.

This paper presents an approximate analysis of delay performance of a WD/SD optical packet/burst switch equipped with infinite optical buffers. By using M/G/1 model, we give the theoretical expression of delay performance of the switch fabric with/without assignment of priority class to service under fixed/variable packet length. Meanwhile, the results of numerical analysis are presented, too. According to these analyses, we give the following main conclusions: in the switch fabric with variable packet/burst length and no assignment of priority class to service, the mean packet/burst delay appears much smaller than that with fixed packet length; and in the switch fabric with assignment of three priority classes to service, the delay of two kinds of services with lower priority classes appears to bigger than that of service with no assignment of priority class.

This work is a part of the project entitled Optical Access Key featuring Low-Power Electronic Adaptive Function (OAKLEAF). A partnership has been formed between the Universities of Cambridge and Essex that aims to build on existing research and has two thrusts; verify that VDSL modem compensation techniques can include impairments present in a complete link comprising optoelectronic, radio and copper media components and to produce a demonstrator showing ho VCSEL-based bandpass modulators may offer an effective interface for hybrid xDSL/optoelectronic access networks. In order to analyze and assess the effect of VCSEL non-linearities in a bandpass modulation context a simulation study has been carried out on an 8-channel optical sub-carrier modulated QAM-VDSL system. Also, a new integrated circuit functional block, a combiner and laser driver with predistortion circuit has been designed and simulated to compensate for the non-linearities of the VCSEL.

We present novel linearization methods for both of single pulse system and single-channel chirped return-to-zero (CRZ) system with the linearization assumption. Both of them show that the Gaussian fit is a good approximation over about two orders of magnitude, but deviate strongly at low probability densities. Linearization allows us to efficiently and accurately compute eye diagrams and bit error rates (BERs) without the use of Monte-Carlo simulations and allows us to greatly increase the accuracy at small BERs at a fraction of the computational cost. We compare these results to the standard Monte-Carlo simulation technique and find that they are agreed very well.

With the increase of the Internet service, it is more and more important to provide better performance and quality in the WDM network. The paper introduces a novel dynamic routing (DR) algorithm based on OSPF with the balance parameter. Due to the limitation to the wavelength continuity, the problem of the routing and wavelength assignment (RWA) is also analyzed comprehensively. In designing the algorithm, RWA is considered, too. The protocol of OSPF may collect the network state information by flooding in WDM network. Furthermore, the algorithm can adapt to the traffic changes with rather little changes in transport performance and blocking probability. In order to prove the method efficient and robust, the simulation with two topologies is done and results are also analyzed and compared. At last, the conclusion and further work are given, too.

This paper presents a study of crosstalk evaluation of a non-blocking banyan switch fabric based on acousto-optic tunable filters (AOTFs). The non-blocking banyan switch fabric is based on the basic unit of 1x2 AOTFs. After describing the crosstalk that exists in the non-blocking banyan switch fabric based on 1x2 AOTFs, we place our emphasis on the two kinds of crosstalk of them: the accumulated AOTF's coherent crosstalk, and the first-order crosstalk caused by the lack of multiplexers. Analyzing the BER degradation induced by these two kinds of crosstalk, we give the theoretical expression of BER and power penalty.

In IP network, the coarse packet classification is a scalable QoS solution. Similarly, it can also be utilized to OBS network because of its scalability. However, the OBS network can support only few priorities of optical burst due to the limitation of current optical device and signal process technology. Hence, in an edge node of OBS networks, the bursts are usually assembled through the many-to-one composite class burst (CCB) assembly technique, e.g., N:1-CCB. Currently, the QoS issue in OBS is as active as that in IP networks. The burst assembly has a great impact on the QoS support of OBS network. According to the study of the N:1-CCB burst assembly technique, there are two aspects related with the scalable QoS support, i.e., the mapping relationship and the assembly resource allocation. This paper simultaneously focuses the two aspects, and based on the linear optimization theory, proposes a novel assembly mapping mechanism, i.e. Optimized TQ-MAP, which most important feature is adaptivity. Based on nonlinear programming and differential calculus, it allocates the burst assembling capacity between classes fairly, efficiently and differentially, and matches IP QoS requirement with OBS QoS capacity as possible as it can. The simulation results show that Optimized TQ-MAP is more adaptive, and can efficiently guarantee the coherence of QoS support from IP to OBS.

Currently optical burst switching (OBS) has been regarded as the most promising backbone networking technology for the next-generation Internet. In the OBS network, the data channel scheduling is one of key problems. Bandwidth efficiency and QoS support are its two concern focuses. However, the existing algorithms pay more attention to bandwidth efficiency. In this paper, we develop an efficient data channel scheduling algorithm, called BM-VF-SBD. It effectively integrates several mechanisms (i.e., void filling, burst migration and selective burst discard) to reduce the bandwidth fragment and support QoS. Its basic idea is in that a new burst is scheduled by migrating some bursts to other channels if none of voids in any channels can accommodate it; otherwise repeating the above processes after selectively dropping some bursts. Meanwhile under an effective data structure, such as the balanced binary search tree, its computational complexity will be o[(2w+1)log w] at most, and be close to LAUC-VF and ODBR. In the proposed algorithm, burst migration plays a key role in the improvement of bandwidth efficiency while selective burst discard has great effects on the two sides. The simulation results show that it performs much better than LAUC-VF and ODBR in burst loss probability (overall or individual) and bandwidth fragment ratio.

A new optical label scheme called Common optical SCM (Subcarrier Multiplexing) label method is proposed. The label paths are consisting of multiplex SCM paths with SCM/WDM technology, which separate from the data wavelength paths. These label paths are parallel and shared for any data wavelength path. We analyzed the modulation and transmission of SCM label channels and compared the bandwidth efficiency and switching performance of main several optical label schemes. The results show that because of no guard between label and data packets the bandwidth efficiency increases and the operation complexity decreases in common optical label schemes and multiple shared parallel SCM label channels can decrease the block probability of label packets.

We demonstrate a novel kind of dense dispersion-managed fiber line, which gradually decrease not only local dispersion in each fiber segment, but also averaged dispersion in every dispersion compensation cell during an amplified period. When an amplifier period contains several or more dispersion compensation cell (L_a/L_c >> 1), short dispersion compensation period can greatly reduce the degree of optical pulse breathing and interaction between neighboring pulse. Loss can make the Kerr effect becomes weaker when optical pulse propagating in an optical amplified length, so better counterbalance between dispersion and nonlinear effect can be obtained by appropriately reducing local and averaged dispersion parameter. Some results show that this kind of optical fiber links have some excellent advantages over constant local and averaged dispersion system, can greatly improve the performance on high speed optical fiber transmission system.

A new flexible broadcast optical switching structure (FBOSS) using tunable optical splitter is proposed. FBOSS supports both point-to-point and point-to-multipoint (broadcasting) connections, and any input optical signal can be broadcasted to any number of output optical ports without extra optical power loss. Its flexibilities of the optical port and optical wavelength help to enhance the capability of switching.

Packet Ring transport technology combines the simplicity of Ethernet and survivability of SONET and attracts more and more interests of researches. However, electrical processing cost will also be very high as data rate grows to gigabit level, which lowers its efficiency. This paper examines the value of optical bypass approach by minimizing the traffic to be processed in electrical domain. An Integer Linear Program (ILP) formulation is presented to solve this problem and results with different traffic patterns are given, which shows one more wavelength can bring notable electrical processing cost savings in the case of a small number of multiplexed wavelengths in one channel.

We have reported a novel 1x2 MEMS optical switch for telecommunication application. A vertical mirror was fabricated by wet anisotropic etching in (110)-oriented silicon wafer. Using DRIE (Deep Reactive Ion Etching) technique, we make a torsional actuator to turn a vertical mirror with a small angle. The actuator was composed of a cantilever beam and two electrodes with curved shape edges. The mirror size was 500 um (L) x 125 um (H) x 50 um (W), and the cantilever dimension was 3000 um (L) x 125 um (H) x 10 um (W). In the optical switch, ball lenses were added in the optics system to extend working distance of fibers. We also fabricated a micro optical platform integrated with the device to simplify the coupling. On the platform, a U-shape groove and fiber clamps were fabricated to accommodate and fix the SMF (single mode fiber). The surface roughness (Ra) of etched mirror was tested below 10 nm. The optical switch was tested on electric and optical characteristics: switching voltage 78.5V, resonance frequency 2.3 kHz, insertion loss 4 - 5 dB, crosstalk 45dB. The device can perform the switching function by the large.

In Optical burst switching (OBS) networks, one important issue is how the performance will be affected by burst assembled from packets. In this paper, we study the fundamental statistic properties of Possion and Pareto traffic, such burst length distribution, inter-arrival time distribution and the switching performance with different aggregation schemes. The results show that the traffic will approach the Gaussian distribution of both kinds of traffic, but the aggregation operation has little effect on the heavy tail of Long Range Dependent traffic. The aggregation algorithm based on aggregation time degrades the loss performance of burst switch because burst arrivals easily concentrate a moment at the period of aggregation time window.

The paper presents techniques for information coding and optical bandwidth reduction that leads to decrease in chromatic dispersion in the optical signal and thus results with an increase in the possible transmission ranges of transceivers.

This report proposes a “Plug and Play” quantum key distribution system using differential phase shift, in which any birefringence effects and polarization-dependent losses in the telecom fiber are automatically compensated by using a Faraday mirror. In this system, a light pulse is split into three pulses after traveling through four couplers and a light intensity feedback servosystem is plugged into this system in order to adjust the phase shifter to compensate the variation of these three pulses. The efficiency of key creation in this system is 8/3 higher than the conventional cryptosystem based on the BB84 protocol.

In this paper, manufacturing issues include Optical Sub-Assembly (OSA), Electrical Sub-Assembly (ESA) and reliability considerations of 10 Gb/s Ethernet transponder were studied by using experiments and implementation. In the growing optical communication industry, one of the star products is the Z-axis pluggable optical transceiver module. Under the broad usage of Ethernet means high port density, low cost, high utilities, compact size and still require excellent performance. After standardizing of 10 Gb/s ethernet (IEEE 802.3ae), many transceiver companies, silicon vendors and system vendors reached the agreement and signed up diversity of MSA (Multi-Source Agreement). These MSAs still keep modifying with system demands, customer requirements, cost and performance issue. This paper presents how to achieve these functions description in the MSA and own a highly manufacturability and reliability module design. According to composed block of transponder, we split it into OSA, ESA, mechanical design and related reliability experimental result. In the OSA, traditional TO-CAN package and optical components be introduced. Because the mature manufacture experience, vendor can easy to meet low cost and manufacturability requirements and only need to slightly modifications. A simply solution be implemented to solve this problem and discuss the critical point of the design. Thermal issue on OSA will also be mentioned because of the sensitive of light source and how to calculate the effect to find effective solutions. By the way, some manufacturability criteria will be discussed for OSA characteristics in 10 Gb/s applications. In the ESA, PMD (Physical media dependant) driving methods, Multi-Source Agreement related digital optical monitor function implement and performance comparison will be presented. On the other hand, we will examine the crosstalk effect between transmitter and receiver circuit and impact to the module Optical to Electrical convert interface design. We still need to take well-assembly between OSA and ESA into consideration. Perfect assembly will lead us to earn wonderful reliability and manufacturability. In the mechanical design, many reliability and maufacturability issue need to consider such as thermal, EMI, temperature, humidity effect etc. So, module mechanical design must be highly collocation with ESA, OSA, heat spreader and special EMI shielding design. Experimental data and thermal simulation result are also prepared to guide us forward to a successful product.

This paper describes the design of an OIF-approved 10Gbps very short reach parallel optical interconnect demonstrated system. It is a 12x1.25Gb/s channel parallel optics solution, leveraging the low cost transceiver (850nm VCSEL), and CMOS (SERDES) technologies originally developed for Gigabit Ethernet. The demonstrator comprises of SONET/SDH serial OC-192 interface, CPLD based convert IC, 1.25Gbps 12-channel parallel optical transmitter and receiver. The transmitter includes a 12-channel array of 850nm VCSEL, a 12-channel VCSEL driver LSI, and a precise coupling structure for 12 multi-mode-fibers ribbon. The receiver consists of a 12-channel array of pin-PDs, a 12-channel receiver LSI, and a precise coupling structure for 12 multi-mode-fibers ribbon. A CPLD chip, which maps the OC-192 framer onto the parallel optical links, and reassembles it after detection, has been developed. We also present the coupling package configuration for VCSEL arrays to fiber ribbon.

With the development of optical fiber communication technology, optical spectrum analyzer (OSA) suitable for field use is urgently required. The laboratory-based OSA can be available only in a laboratory environment and unsuited to field use. By eliminating unwanted functions and fine design, the portable OSA can meet the requirement for field test and provide low cost and high performance. In this paper, a kind of high speed and portable OSA suitable for plastic fiber field test is designed, whose wavelength measurement accuracy is better than 0.1nm and spectrum energy measurement sensitivity is -30dBm.

A reconfigurable tri-directional transmission module (TTM) is proposed and experimentally demonstrated in a dense wavelength division multiplexing (DWDM), 2.5 Gb/s x 100 Km lightwave system. Channel spacing is 100 GHz (0.8 nm) in our study. Bit error rate floor of 10^-4 exists due to channel crosstalk and Rayleigh back scattering is observed. Nevertheless, power penalty could be effectively reduced to 0.6 dB by using a tunable bandpass filter. This module may find an important role in the tri-directional lightwave system, especially when wavelengths interchange, optical cross connect and optical add/drop functions are necessary.

The rapid increase of IP service and large capacity in DWDM layer request the all-optical packet delineation and process. Optical burst switching is a promising technology in the future IP/DWDM networks, which provide a feasibility to realize the optical packet switching without optical buffer. Presently the QoS guarantee of OBS depends on the bandwidth control protocol. JET based on RFD generally applied in OBS network is considered in this letter. According to the queuing theoretical model, the performance of one node is studied using simulations. Besides the traditional exponential arriving and service traffic, we analysis the general and self-similar traffic under the same average service time. The analysis and simulating results for different class have important effects on networking performance by comparing the queuing service models.

Optical burst switching (OBS) is a viable solution to support next generation optical Internet. This paper addresses the problem of inter-class bandwidth management for OBS supporting multi-class services. Bandwidth management should be simple to implement and optimally designed to maximize the network performance. We propose here a wavelength based inter-class bandwidth management approach for optical burst switched networks (OBSN). Three different wavelength allocation schemes are presented respectively. In our schemes, each class of service is endued with a weight according to its importance, which is proportional to the charge on corresponding class of service by the network operator. The static scheme allocates wavelengths based on the weight of class, while the dynamic allocation scheme is based on the weighted traffic load with the help of a traffic predictor. The merit optimized allocation aims to achieve the maximized weighted throughput of the network based on a traffic predictor and a dynamic optimization algorithm. The performance of the proposed three schemes are evaluated and compared by computer simulations, where the weighted loss rate is chosen to be a merit used in performance evaluations. From the viewpoint of the network operator, the merit-based scheme provides the maximized profit since it improves the weighted throughput of the system, which is approved by simulation results.

This paper focuses on the problem of optical buffer control strategies for recirculation shared optical buffer in optical packet switching. Five control algorithms are presented, and their performances are evaluated and compared by computer simulations. Issues on the optimal configuration of optical buffers are also discussed. The numerical results show that the time delay resolution of fiber delay line has significant influence on the system performance in all three discussed switching mode. The results are valuable in designing the optical packet switch.

The effect of liquid-crystal spiral phase plates on the generation of doughnut laser beams is studied. Several parameters of the liquid-crystal (LC) are analyzed and carefully selected including birefringence of the LC, thickness, cell gap, type, topographic charge number, and applied voltage. The LC spiral phase plates have been fabricated.

An accurate calculating method for optical power attenuation in intersatellite laser link (ISL) is presented. Based on transmission characteristics of Gaussian beam, we analyze the effect of light wavelength, transmission distance and optical antenna upon optical power, and pointing error is also taken into consideration. Compared with the usually used method, which borrowed from microwave antenna theory, this calculation method is well-founded and possessing higher accuracy.

Free-space optical transmission can provide high-speed links for a variety of applications. However, inhomogeneities in the temperature and pressure of the atmosphere lead to variations of the refractive index along the transmission path. This atmospheric turbulence can deteriorate the quality of the image formed at the receiver, and can cause fluctuations in both the intensity and the phase of the received signal. Laser beams experience three effects under turbulence: beam wander, beam spreading and scintillation. Of the three turbulence effects, free space optical systems might be most affected by scintillation. Random interference with the wave front can cause peaks and dips, resulting in receiver saturation or signal loss. Based on Kolmogorov’s theory, the Signal-to-Noise Ratio (SNR) and Bit Error Rate (BER) from scintillation were drawn in this paper. For a FSO system, after choosing a certain beam and communication range, link performance characterized by its SNR or BER is related to the refractive index structure coefficient C_n² and turbulence depth L_t. Considering all the three effects, we find for a weak turbulence model, a FSO system should be applied within a communication range of 2.5km and above the ground at least 5m.