Korea is small country but has recently dramatic growth in broadband internet subscribers. At the present of December in 2003, about 73% of total households are internet subscribers. The country is now pursuing new internet highway with faster speed and broadband services for future intelligent internet. The viable research targets are the switching technologies, the access network and convergence network technologies. The optical burst switching will be discussed as the intermediate solution before the coming sophisticated optical packet switching technologies. Another important issue is how we accept diverse service types in future network. There are always dilemmas to choose “optical” or “electrical” for the network in both cost and technology views. The WDM PON-based FTTH and the broadband convergence network (BcN) plans in Korea will also be discussed.

This paper highlights advantages of Optical Transmission Network (OTN) standards, explains its benefits in combination with IP networks and discusses potential introduction strategies. Furthermore a practical introduction example is presented.

For distributed control, network topology and link resource information is required at the ingress nodes to compute and route optical paths. It is the responsibility of routing protocols to disseminate this information. Generalized MPLS has extended traditional IP routing protocols such as OSPF for explicit path computation and traffic engineering (TE) in optical transport networks. However it is noted that the current version of GMPLS-extended OSPF-TE does not carry sufficient link state information to compute shared mesh restored paths. This paper describes the new extensions to the GMPLS OSPF-TE protocol to support the path computation for shared mesh restoration in optical networks. The efficient methods for aggregation and dissemination of link resource availability and sharing information are proposed. Especially these new extensions support both Shared Risk Link Group (SRLG) disjoint and node disjoint restoration paths, which provides more flexibility to the network operators. The network operators can choose to guarantee recovery from a SRLG failure or from a node/SRLG failure based upon the applications and the service level agreements with their customers. Furthermore, the paper presents an analytical model to estimate the performance of the GMPLS OSPF-TE protocol and the proposed extensions, including control bandwidth and memory requirements.

Over the last decade, deployed core telecom networks have migrated from being based on single-channel SONET regeneration links to multi-span, multi-channel optically amplified systems. More recently, the industry has been aggressively pursuing a natural extension of this philosophy towards all-optical “analog” core networks, with each traffic demand touching electrical digital circuitry only at the in/egress nodes. This trend produced a substantial elimination of regeneration costs, increase in network capacity, and notionally simpler operation and service turn-up. At the same time, the optical “analog” network requires a large amount of sophisticated hardware and software for monitoring and manipulating high bit rate optical signals. The primary goal for current equipment suppliers is to provide cost effective system designs that are simple to deploy and operate. This paper will examine the trade-offs inherent in the technology and architecture choices needed to reach this goal through the “analog” transmission/all-optical ideal and concludes that it is difficult to improve on the present approach which uses a mix of transparent and opaque network elements.

Current optical transport networks provide high bandwidth through the use of advanced WDM technology, but are difficult to adapt to the different statistical patterns and quality of service (QoS) demands of future traffic. There has been much debate whether the use of dynamically reconfigurable optical networks would have a number of advantages in accommodating the needs of future traffic demands. Dynamic networks would eliminate the need for frequent opto-electronic conversion in current networks, and may save resources through higher utilization and fast adaptation. Different architectures have been proposed to address this problem: Wavelength-routed optical networks (WRON), optical burst switching (OBS), and optical packet-switching (with increasing granularity and speed of reconfiguration). In this paper we investigate whether these architectures are suitable (necessary?) to meet the demands of future traffic, using an analysis focusing on both modeling and experimental aspects.

In the last decade Optical Time Division Multiplexed (OTDM) systems have attracted considerable interest, as an alternative to the existing commercially deployed Wavelength Division Multiplexed (WDM) systems. In this presentation, first the basic point-to-point bit interleaved OTDM system will be explained, with special focus on the transmitter and the receiver functionalities, demultiplexing and clock extraction. Some reported results are highlighted. In order to expand the system to include a number of nodes, channel identification is required, and a novel technique will be introduced. A simple network configuration is presented, exemplifying advantages and drawbacks of bit interleaved OTDM systems. Finally, a short introduction to reported results on packet based OTDM systems will be presented.

In the past the concept of normalized transmission sections for WDM systems, where each link of a network is divided into equally spaced amplifier sites of 80 km length and appropriate dispersion compensation, has been validated theoretically and experimentally. The advantage of such a concept compared to traditional link design is that from a network point of view the performance along any transparent transmission path within this network can be easily predicted since every section of a path is built by cascading the same building blocks e.g. amplifiers and dispersion compensating fibers. However, for a field deployment in a real network this concept is not very practical, since lengths distribution of the housing stations and spare use of the dispersion compensating fibers along the transmission path, are not considered. In this paper we will compare different approaches to reduce and simplify the design complexity of such optical transmission systems taking into account DCF modules with limited compensation range e.g. DCF-80, DCF-20 or DCF-5 to compensate for 80, 20 or 5 km of standard single-mode fiber and a typical field distribution for the span lengths. As results distributions of the system penalty are analyzed for the different cases and specified target lengths for e.g. a national European network of 800 and 1600 km length with field values for the amplifier sites to guarantee a robust system design taking into account dispersion, fiber nonlinearity and amplifier noise.

This paper discuss the different possible multiplexing structure for an optical network - in a broad sense ranging from all optical networks including optical switching to optical networks “as we know them” mainly using optics for the transmission.

Increasing demand for high transmission capacity and the decreasing revenues per bit, combined with the given economy of scale for optical networks, forces the network operators to enhance the channel data rates as well as the channel numbers. Higher channel data rates result in a lower footprint, energy consumption and a lower complexity in network management and operation support systems, due to lower channel numbers. In principle the enhancement of channel data rate leads to a reduction of system tolerance for chromatic dispersion, PMD and nonlinear effects. Furthermore higher order effects like dispersion slope and higher order polarization mode dispersion have to be taken into account. On the other hand the fast pulse broadening leads to a quasi linear behaviour of the systems, which relaxes the impact of fiber nonlinearities compared to 40 Gbit/s transmission. The lower tolerances can partially be mitigated by the implementation of compensators and more complex amplification schemes. Accounting for less tolerances, adaptive compensating modules and higher sophisticated amplification schemes, the complexity of system design is increased. We investigate theoretically and numerically the limiting physical effects and the impact on the signal performance, induced by chromatic dispersion and nonlinear impairments. We present derived engineering rules for all relevant effects and for various fiber types, based on channel data rates of 160 Gbit/s. These engineering rules enable design engineers to perform a fast system design and system degradation estimation, without time consuming full numerical simulations.

Wavelength-selective switching is generally considered to be an essential element for building agile “all optical” mesh telecommunications networks. This paper will provide an overview of the various node architectures suitable for implementing such networks. It will also provide an overview of the various technologies, that have been developed for such networks, including integrated optical waveguide circuits and free-space optical devices.

We present a new WADM network node with 50% optical component reduction, 30% smaller footprint, and 35% electrical power saving compared with conventional WADM for optical channel shared protection ring application. It has no single-point-of-failure. Since coherent crosstalk might be a potential issue in this design, we experimentally measured the Q-factor as function of Signal-to-Interference Ratio (SIR) to characterize its performance.

This paper explains benefits and challenges of future transparent optical networks. With a special focus on new transmitter/receiver designs, it discusses the key components enabling transparency. The challenges in the system design of such networks are also addressed.

This paper proposes an architecture of a next-generation Internet eXchange (IX) based on the Generalized Multi-Protocol Label Switching (GMPLS) technologies and the photonic cross-connect (PXC), hereafter referred to a Lambda-IX. At first, we investigate a basic Lambda-IX model where the PXC provides a GMPLS-controlled lambda label-switched path (LSP) to interconnect different ISPs' border routers. We verify that the proposed Lambda-IX model can achieve the lambda-based and resilient interconnection for the ISPs, thanks to the PXC's bit-rate insensitive operation as well as fast restoration operation. In addition, once GMPLS functionalities are introduced on the border routers as well as the PXC, very flexible interconnection can be achieved such as demand-based creation and deletion of lambda LSPs. Next, we initiatively demonstrate an experimental Lambda-IX using a PXC and IP/MPLS routers. A Lambda-LSP with OC-192 bandwidth can be successfully created by using the GMPLS RSVP-TE signaling protocol via a control plane, and an EGP session of Border Gateway Protocol 4 (BGP-4) can be established over the Lambda-LSP created between the GMPLS-enabled border routers via a data plane. We also evaluate the fault recovery operation in the case where such Lambda-IXs are consisted of several PXCs and demonstrate that the Lambda-LSP as well as the corresponding BGP session can be restored with a fast recovery time of less than 1s. Through these investigation and demonstration, it is revealed that the Lambda-IX can be put to practical use aiming at inter-exchanging a large traffic in a near future, while enriching the functions of IX.

Required limits of transmission distance, nodes cascadability and grade of transparency (number of regenerators) are presented to implement semi-transparent DWDM backbone networks. Results are based on OSNR performance of three realistic optical network scenarios. Optical transmission impairments are key issues for transparent optical networks. To improve transmission in an optical network one can use forward error correction (FEC), Raman amplification, robust modulation formats tolerant to non-linear effects and noise, optimised dispersion maps, semi-transparent OXC architectures with selective 3R or 2R regeneration and reduction of losses in the optical cross connector (OXCs) architectures. In a typical optical network with mesh topology, a transmitted optical signal is expected to traverse several nodes connecting any source-destination pair. The cascadability limit of transparent optical cross-connect (OXC) nodes and the transmission distance between OXC nodes are crucial network design parameters. The required limits of transmission distance, OXC nodes cascadability and number of regenerators per node are presented for DWDM backbone networks. OXC architectures with low loss components and cost effective distributive amplification is required to facilitate a high OSNR and reduce the regeneration rate of semi-transparent networks. OXC architectures with OSNR below 36.7 dB are capable of reducing the regeneration rate to 13% for realistic network topologies.

This paper presents different approaches to enable high-speed transmission of 10 Gbit/s and beyond on polarization mode dispersion (PMD) limited fibers. An introduction to the phenomenon of PMD and its impact on system performance is presented. An overview of common optical PMD mitigation methods and their basic concept is given, including the problem of multi-channel PMD compensation schemes for WDM systems. Furthermore alternative methods like polarization scrambling, forward error correction and electrical mitigation are considered. Bit error rate (BER) determined by error counting is used as quantification for PMD induced outages and a comparison to eye opening penalty (EOP) based performance evaluation is given. The performance of basic PMD mitigation schemes is compared by using the EOP and the BER based outage criterion.

We report for the first time a multichannel in-line PMD compensator which independently provides higher-order PMD compensation for each DWDM channel in the C band. Feedforward compensation is implemented through a robust optimization algorithm using high-speed, high-resolution spectral polarimetric data for each channel. Both 10 and 40 Gb/sec 40-channel 100-GHz compensators are reported, and extensive test results over a range of second-order PMD conditions are presented.

In high-speed and/or long-haul transmission systems with data rates of 40 Gbit/s and beyond, the system degradation due to 1st- and also higher-order polarization mode dispersion (PMD) is a significant issue. So far, PMD compensators have been proposed for 1st- and higher-order PMD. In this paper, we focus on two types of optical PMD compensators, single-stage first-order compensator and double-stage compensator with higher-order canceller, and estimate their performances. Moreover, we propose a single-stage first & higher-order compensator as a “degenerate” version of the double-stage one by reducing the number of monitoring parameters (circuit) and the number of controllers (PCs). The functions of 1st- and higher-order PMD compensators are integrated into a single-stage configuration, with signal distortion monitoring at the output of the polarizer. The compensation performance of three PMD compensators is investigated in a 10.7 Gbit/s NRZ transmission system. The results show that the proposed single-stage compensator can effectively suppress the degradation due to 2nd-order PMD.

Dynamic dispersion compensation based on non-linear self-phase modulation (SPM) in an all-fiber device is demonstrated. The basic design of the compensator is very simple, consisting only of a pre-compensating negative dispersion fiber, an optical amplifier, and a highly non-linear positive dispersion fiber. Multiple channel operation of the compensator is feasible and experimentally demonstrated. An increase of dispersion tolerance of at least a factor of 2 is shown with low penalty of less than 2 dB. Finally, device performance in a 2000 km fiber loop experiment is presented.

With increasing line rates system tolerances due to both, chromatic dispersion and dispersion slope, become more restrictive. By means of numerical simulations we investigate the influence of temperature fluctuations in cascaded multi-section dispersion-compensated standard single-mode fiber SMF based transmission systems. Eye opening penalty (EOP) for different modulation formats - non-return-to-zero NRZ as well as return-to-zero RZ with duty cycles varying from 0.2 to 0.6 and for different signal power levels varying from -9 to +3 dBm at a line rate of 160 Gbit/s are calculated. Performance degradation due to temperature-induced dispersion variations is investigated for three different scenarios: a) the temperature of the installed terrestrial fiber-optic cable changes, b) the temperature of the temperature-controlled indoor dispersion compensating module DCM changes, c) combined effects.

A joint field trial on 40G transmission was conducted, where Deutsche Telekom's and NTT's R&D groups got involved. 43-Gbit/s/ch CS-RZ signals of the NTT G.709-OTN system were transmitted over selected high-PMD fiber links in DT's actual field environment. The advantage of CS-RZ modulation format was confirmed in comparison with conventional NRZ against all-order (first- and higher-order) PMD. Moreover, we introduced a new scheme for an accelerated determination of outage probability due to PMD that is adapted to practical field conditions in an operated fiber network with buried fibers and in-line operational optical elements.

Polarization mode dispersion (PMD), polarization dependent loss (PDL) and chromatic dispersion (CD) of fiber optic components and fiber transmission links are key issues in fiber optical communication. Of interest are the respective static values as well as their variations in time which might be due to environmental changes or due to rerouting. Basic aspects will be briefly addressed, a variety of standard and state-of-the-art characterization and monitoring techniques will be reviewed and recent experimental results will be presented.

In this article a new quantitative method for polarization mode dispersion (PMD) tolerance testing is presented. Novel PMD sources, test procedures, and evaluation tools have been integrated into a comprehensive test methodology. The new PMD sources provide not only first order PMD, but also higher order PMD states. These sources are stable, able to maintain any desired PMD state for extended bit error tests. The basis of the new PMD tolerance methodology is to remove PMD statistics during the data acquisition, and reintroducing the PMD statistics in the data analysis. Total PMD outage time can then be determined for a fiber span with known mean PMD and defined system power penalty for PMD distortion. We will describe a new PMD source, test and evaluation method in detail, showing simulated test results and the ability to use this method to optimize optical transport systems and network design rules.

A simple, adaptive PMD compensator is demonstrated in a 1.76Terabit/s (44 x 43 Gbit/s) transmission experi- ment over 600km SMF with a mean PMD of 8 ps. The PMDC is composed of a LiNbO3 polarization-controller followed by a component with an adjustable differential group delay (DGD) in the range of 0-20 ps. Feedback control is provided by means of continuously monitoring and maximizing the degree of polarization (DOP). This PMDC is shown to automatically compensate well for 1st-order (0-28 ps) and one part - the depolarization - of 2nd-order (0-130 ps2) PMD. Polarization dependent chromatic dispersion (PCD), being the other part of 2nd-order PMD, is negligible as long transmission systems without a PMDC are considered. But in fact PCD plays a role when an optical PMDC is introduced. Because of the anyway tight tolerance of 40 Gbit/s optical signals to chromatic dispersion, the variable dispersion compensator VIPA (virtually imaged phased array) with a tuning range of +/-800 ps/nm is introduced besides the PMDC. Therefore, with VIPA not only residual chro- matic dispersion put also PCD is e®ectively compensated for. Using adaptive optical technologies compensating for PMD and chromatic dispersion, error-free transmission (BER< 10¡15 with FEC, Q-margin=3 dB) of 44 x 43 Gbit/s (1.76 Tbit/s) over 600km of high PMD (8 ps) SMF is demonstrated.

In this paper we examine the statistics of combined polarization mode dispersion (PMD) and polarization dependent loss (PDL) and the impact on pulses in optical fiber systems. The probability density functions of PDL, of the differential group delay's (DGD) real and imaginary part and of the angle between the principal states of polarization (PSPs) are investigated by numerical simulations. The PDL and the real part of the DGD turn out to be Maxwellian-distributed while the imaginary part of the DGD complies with a sech²-distribution. Moreover we take a look at the transmission of optical pulses with varying input states of polarization and show that the PSPs of the well-known PMD-theory are not necessarily the fastest and slowest states nor are they the least distorted ones.

Next generation automatic switched optical networks (ASON) show great promise in coping with increasing bandwidth demands, as they provide both on-demand bandwidth and improved switching flexibility. In a multi-layer data-centric network, such an ASON acts as a server layer and provides the topology for a client IP network. Multi-layer Traffic Engineering (MTE) enhances single-layer traffic engineering (i.e. adaptive routing) with the possibility to reconfigure the logical IP topology, utilizing the aforementioned optical flexibility in setting up and tearing down end-to-end lightpaths. To provide a robust network service, the multi-layer networks must be recoverable from different types of failures. Resilience mechanisms such as various forms of protection and restoration allow to recover from optical layer failures. In a multi-layer network however, some traffic will be forwarded through the IP layer routers, requiring a survivability scheme in this IP layer. This paper shows how MTE strategies that normally cope with changing traffic demands, can be used to provide this IP layer resilience. This is done through diverting affected IP traffic and replacing failing optical lightpaths, effectively leading to a dynamic restoration scheme. We will evaluate and compare failure performance of different types of resilience mechanisms, based on an existing MTE strategy.

A multilayer network is a complex system composed of many devices which in turn are composed of many subsystems. A failure event changes the configuration of the multilayer network, triggering recovery at more than one layer. Proving that a circuit will work after provisioning operations is an important step to preventing misconfiguration of network devices. We have been working on implementing the consistency checker for one layer. It creates a network representation and adds changes to the network. To verify operation of the system, graphical representation of packets are exchanged between different ports. To understand such a systems, we describe them with a block diagram representing important resources.

Photonic crystal fibers (PCFs) and photonic bandgap fibers (PBFs) for high performance systems and devices are reviewed. A PCF has an array of air holes surrounding a silica core region. The PCF has special characteristics compared with conventional single mode fibers. Theoretical analyses and measurements show that the zero dispersion wavelength of PCF is shorter than 1280 nm that is suitable for dispersion compensation at 1550 nm. PCFs with zero dispersion in the 1550 nm region were recently proposed. Another noteworthy characteristic of PCFs is their strong birefringence, which is induced by the size and arrangement of the air holes. A theoretical analysis and experiments showed high birefringence, three times larger than that of conventional polarization maintaining fibers. Therefore, optical components with better polarization maintaining characteristics are expected. The new topic is PBFs that have hollow core and a lattice in the surrounding region to form a photonic band gap. They are expected to realize low-loss, low-dispersion, and low nonlinerity fiber because the core is air. Recently we proposed a new type PBF to aim at wide bandgap opening by using superlattice structure. This paper describes the characteristics of dispersion controlled PCFs, polarization maintaining PCFs, and the band structure and experimental results of the new PBF. The potential of PCFs and PBFs will be discussed with reference to the next generation high performance networks.

Recently emerged photonic bandgap fibers with their extraordinary optical properties offer many interesting device applications. We present the status of our work on the use of this kind of a fiber in sensing and wavelength referencing both in the 1300 and 1500 nm wavelength regions. The photonic bandgap fibers are spliced to standard single-mode fibers at input end for easy coupling and filled with gas through the other end placed in a vacuum chamber. The technique is applied to measure absorption lines of strongly absorbing gases such as acetylene and hydrogen cyanide by employing tunable lasers and LEDs as light sources. The measurement of weakly absorbing gases such as methane and ammonia is also explored. To realize a permanent wavelength reference sealing of a photonic bandgap fiber using index-matching UV-curable adhesive is demonstrated.

Driven by the increasing data traffic and the increasing demand for bandwidth optical fiber technologies play a greater role in todays and future data-communication networks. Although the well-known silica fiber have the potential of achieving very large bandwidth, this fiber is not the ideal medium for high bit rate data-communication for office and home applications because its small dimension requires well sophisticated components as well as installation technologies. This increases the total system cost, inevitably. However, the technologies of plastic (polymer) optical fibers (POF) and the devices for POF nowadays show rapid process /2/. So, we could benefit from the special advantages of these fibers over a wide field of applications, from decoration to local networks, including lighting, image guides and sensor technique. Today, inexpensive and robust POF transmission systems are available on the market with high bit rate capacity. Bus-systems, e.g. MOST and Byteflight, are applied in the rough automotive environment.

Performance of in-line all-optical signal regenerators utilizing self-phase modulation (SPM) or four-wave mixing (FWM) in fibers is analyzed. First, two types of SPM-based regenerators are analyzed and compared: one utilizes soliton-like pulse compression in anomalous-dispersion fiber and subsequent filtering and the other utilizes spectrum broadening in normal-dispersion fiber and subsequent spectrum slicing. Although both types of regenerators show good regenerator performance, one based on spectrum broadening and slicing has better ability of stabilization of signal amplitude while requiring larger signal power launched into the nonlinear fiber. Then the performance of FWM-based regenerators, which have ability of amplitude stabilization while maintaining phase information carried by the signal is discussed. It is shown that return-to-zero differential phase-shift keying transmission performance can be improved by the regenerator.

We demonstrate an all-optical waveform sampling system with simultaneous sub-mW optical signal sensitivity (20 dB SNR) and sub-picosecond temporal resolution over more than 60 nm optical bandwidth. The optical sampling was implemented by four-wave mixing in a 10 m highly nonlinear fiber using a sampling pulse source with a sampling pulse peak power of only 16 W. The sampling performance was evaluated in terms of sensitivity, temporal resolution and optical bandwidth with respect to fiber length, sampling pulse source wavelength offset from the zero-dispersion wavelength of the highly nonlinear fiber, sampling pulse peak power and walk-off due to chromatic dispersion. We also present a summary of the available methods to achieve polarization-independent optical sampling.

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 systems. In terms of signal distortion besides advanced link design strategies optical regenerators are of high interest for signal conditioning along a fiber link. Furthermore the upcoming techniques of advanced modulation formats build a basis for further improvement of transmission quality that ends up with even higher reach, higher robustness against transmission impairments, lower signal to noise requirements etc. In this talk the current status of research in the field of optical signal regenerators suitable for signal regeneration for different modulation formats will be reviewed and presented, where we restrict our analysis to 2R regenerators. The focus will be put on regenerators taking advantage of non linear optical effects in order to improve signal quality. The main challenge is the treatment of phase modulated signals within regenerators. Especially the differential phase shift keying (DPSK) modulation format or the duobinary modulation format are attractive candidates for such kind of regenerators for further improvement of DPSK based transmission systems. The paper will go through several regenerative concepts like non linear optical loop mirror (NOLM) based and cross phase modulation based regenerators in order to figure out their potential for advanced modulation format based systems.

This paper will describe some of the latest developments in the area of optical sources for WDM systems, including tunable lasers and multiwavelength sources. We will consider both DWDM and CWDM. We will start by looking at the source requirements, such as wavelength switching time, for various applications. After that we will give an overview of the latest developments in source options and technologies. The source options under discussion will include external cavity lasers, wavelength selectable laser arrays, tunable VCSELs, monolithic tunable lasers, and non-semiconductor alternatives. Numerous examples will be shown, and the characteristics and performance of the various devices will be discussed. The key performance parameters, such as tuning range, power and switching time will be related to the expected areas of application. These areas include sparing, fixed wavelength transmitter replacement, and use in wavelength switched networks. We will pay particular attention to packaging and modules, and discuss the emergence of pluggable lasers. Finally we will consider the prospects for uncooled operation.

In this paper, we have analyzed the simulation results for 10 Gbit/s indirect modulated optical links with post-dispersion compensation for the signal formats, return to zero (RZ), non-return to zero (NRZ) and gaussian pulse using 100kms standard fiber and Fiber Bragg Grating (FBG). The performance characteristics like Q-factor from minimum bit error rate and eye diagrams at the output are obtained by simulating optical links using different signal formats. It is observed that the optical link with NRZ pulse offer maximum eye opening, Q-factor from minimum BER and the electrical SNR varies from 18.06 to 28.3 dB at -800 ps/km to -1600 ps/km respectively.

Based on the scenario that the number of wavelengths may be fixed by upgrading the WDM metro ring networks, this paper proposes a novel WDM slotted-ring architecture with nodes employing one fixed transmitter and a set of fixed receivers. It presents a detailed design and analysis of node architecture and MAC protocol, which aim at guaranteeing fairness to each node and achieving efficient bandwidth utilization. The main advantages of this proposed node architecture and its corresponding MAC protocol can be summarized as: 1) Electronic buffer and process delays can be minimized as the packets will not be processed until they arrive at the destination nodes. 2) No contention will occur at the destination nodes. 3) No tunable components used. 4) No dedicated control channel used to transfer the packets. 5) Packets can be stripped by both destination and source nodes, and there is no need to perform O-E-O conversions. It is shown through theoretical analysis and simulation results how the proposed architecture and MAC protocol can achieve a good throughput, low queuing delay and reasonable packet loss probability. In the case of a large number of wavelengths, or if the number of wavelengths increases by upgrading the networks, this paper further proposes an original design method: employing one fixed transmitter and a set of tunable receivers that can only be tuned among partial continuous wavelengths, we show this method can achieve a good trade-off in terms of performance metrics such as network’s throughput, cost and implementing complexity.

In IEEE 802.17 draft standard, bridging technique is suggested to connect multiple resilient packet rings (RPRs). However, we demonstrate that bridging technology may not be a feasible solution because too much of resources and overhead would be occupied or needed to transmit packets across multiple RPR sub-networks. To construct interconnected RPRs and thus solve bridging related issues, this paper introduces a new kind of RPR nodes, namely, RPR-X nodes, to interconnect multiple RPR sub-networks. In our solutions, standard RPR frame defined by IEEE 802.17 is used to transmit packets on the local RPR sub-network, the extended RPR frame (i.e., cross frame) is used to transmit the packets across different RPR sub-networks. For RPR-X node, it can identify and process both standard RPR frame and extended RPR frame. While for RPR-S node (standard RPR node), it can only identify and process standard RPR frame. Based on the extended definitions of standard RPR frame, we give a detailed design of MAC layer reference model of RPR-X node, and propose an original packet cross forwarding algorithm, with a view to transmitting the packets across multiple RPR sub-networks, and at the same time to guaranteeing the cross forwarding algorithm to be compatible with the standard forwarding procedure of RPR-S node. Theoretical analysis and simulation results demonstrate that the proposed methods perform well in terms of performance metrics such as network's throughput and end-to-end delay.

Atmospheric turbulence causes fluctuations in both the intensity and phase of the received signal in an optical wireless communication link. These fluctuations, often referred to as scintillation noise, lead to signal fading, which increase bit errors in digital communication links using intensity modulation and direct detection. The performance of an optical link can be improved by the use of a time delayed diversity technique, which takes advantage of the fact that the atmospheric path from transmitter to receiver is statistically independent for time intervals beyond the correlation time of the intensity fluctuations. We have designed and constructed a prototype optical wireless system using this scheme. Bit-error-rate measurements have been used to characterize the link performance for different delay periods under conditions of controlled simulated turbulence. It has been determined that link performance improves significantly, especially in strong turbulence. In addition, we have implemented orthogonal polarization modulation, which works especially well in optical wireless systems. In contrast to fiber optic communications, the polarization state of a laser beam is well preserved on a free space optical path.

The proliferation of high-bandwidth applications has created a growing interest in upgrading networks to deliver broadband services to homes and small businesses between network providers. There has to be a great efficiency between the total cost of the infrastructures and the services that can be offered to the end users. Coarse Wavelength Division Multiplexing (CWDM) is an ideal solution to the tradeoff between cost and capacity. This technology uses all or part of the 1270 to 1610 nm wavelength fiber range with optical channel separation about 20 nm. The problem in CWDM systems is that for a given reach the performance is not equal for all of transmitted channels because of the very different fiber attenuation and dispersion characteristics for each channel. In this work, by means of an Optical Communication System Design Software, we study a CWDM network configuration, for lengths of up to 100 km, in order to achieve low Bit Error Rate (BER) performance for all optical channels. We show that the type of fiber used will have an impact on both the performance of the systems and on the bit rate of each optical channel. In the study, we use both on the already laid and widely deployed singlemode ITU-T G.652 optical fibers and on the latest "water-peak-suppressed" versions of the same fiber as well as G.655 fibers. We have used two types of DML. One is strongly adiabatic chirp dominated and another is strongly transient chirp dominated. The analysis has demonstrated that all the studied fibers have a similar performance when laser strongly adiabatic chirp dominated is used for lengths of up to 40 Km and that fibers with negative sign of dispersion has a higher performance for long distance, at high bit rates and throughout the spectral range analyzed. An important contribution of this work is that it has demonstrated that when DML are used it produces a dispersion accommodation that is function of the fiber length, wavelength and bit rate. This could put in danger the quality of a system CWDM if it is not designed carefully.