AI-based near-optimal placement of wavelength-conversion nodes in optical networks

Fiber optics has developed so rapidly during the last decades that it has be- come the backbone of our communication systems. Evolved from initially static single-channel point-to-point links, the current advanced optical backbone net- work consists mostly of wavelength-division multiplexed (WDM) networks with optical add/drop multiplexing nodes and optical cross-connects that can switch data in the optical domain. However, the commercially implemented optical net- work nodes are still performing optical circuit switching using wavelength routing. The dedicated use of wavelength and infrequent recon¯guration result in relatively poor bandwidth utilization. The success of electronic packet switching has inspired researchers to improve the °exibility, e±ciency, granularity and network utiliza- tion of optical networks by introducing optical packet switching using short, local optical labels for forwarding decision making at intermediate optical core network nodes, a technique that is referred to as optical label switching (OLS). Various research demonstrations on OLS systems have been reported with transparent optical packet payload forwarding based on electronic packet label processing, taking advantage of the mature technologies of electronic logical cir- cuitry. This approach requires optic-electronic-optic (OEO) conversion of the op- tical labels, a costly and power consuming procedure particularly for high-speed labels. As optical packet payload bit rate increases from gigabit per second (Gb/s) to terabit per second (Tb/s) or higher, the increased speed of the optical labels will eventually face the electronic bottleneck, so that the OEO conversion and the electronic label processing will be no longer e±cient. OLS with label processing in the optical domain, namely, all-optical label switching (AOLS), will become necessary. Di®erent AOLS techniques have been proposed in the last ¯ve years. In this thesis, AOLS node architectures based on optical time-serial label processing are presented for WDM optical packets. The unicast node architecture, where each optical packet is to be sent to only one output port of the node, has been in- vestigated and partially demonstrated in the EU IST-LASAGNE project. This thesis contributes to the multicast aspects of the AOLS nodes, where the optical packets can be forwarded to multiple or all output ports of a node. Multicast capable AOLS nodes are becoming increasingly interesting due to the exponen- tial growth of the emerging multicast Internet and modern data services such as video streaming, high de¯nition TV, multi-party online games, and enterprise ap- plications such as video conferencing and optical storage area networks. Current electronic routers implement multicast in the Internet protocol (IP) layer, which requires not only the OEO conversion of the optical packets, but also exhaus- tive routing table lookup of the globally unique IP addresses. Despite that, there has been no extensive studies on AOLS multicast nodes, technologies and tra±c performance, apart from a few proof-of-principle experimental demonstrations. In this thesis, three aspects of the multicast capable AOLS nodes are addressed: 1. Logical design of the AOLS multicast node architectures, as well as func- tional subsystems and interconnections, based on state-of-the-art literature research of the ¯eld and the subject. 2. Computer simulations of the tra±c performance of di®erent AOLS unicast and multicast node architectures, using a custom-developed AOLS simulator AOLSim. 3. Experimental demonstrations in laboratory and computer simulations using the commercially available simulator VPItransmissionMakerTM, to evaluate the physical layer performance of the required all-optical multicast technolo- gies. A few selected multi-wavelength conversion (MWC) techniques are particularly looked into. MWC is an essential subsystem of the AOLS node for realizing optical packet multicast by making multiple copies of the optical packet all-optically onto di®er- ent wavelengths channels. In this thesis, theMWC techniques based on cross-phase modulation and four-wave mixing are extensively investigated. The former tech- nique o®ers more wavelength °exibility and good conversion e±ciency, but it is only applicable to intensity modulated signals. The latter technique, on the other hand, o®ers strict transparency in data rate and modulation format, but its work- ing wavelengths are limited by the device or component used, and the conversion e±ciency is considerably lower. The proposals and results presented in this thesis show feasibility of all-optical packet switching and multicasting at line speed without any OEO conversion and electronic processing. The scalability and the costly optical components of the AOLS nodes have been so far two of the major obstacles for commercialization of the AOLS concept. This thesis also introduced a novel, scalable optical labeling concept and a label processing scheme for the AOLS multicast nodes. The pro- posed scheme makes use of the spatial positions of each label bit instead of the total absolute value of all the label bits. Thus for an n-bit label, the complexity of the label processor is determined by n instead of 2n.

DWDM networks make use of optical switching networks that allow light waves of multiple lengths to be serviced and provide the possibility of converting them appropriately. Research work on optical switching networks focuses on two main areas of interest: new non-blocking structures for optical switching networks and finding traffic characteristics of switching networks of the structures that are already well known. In practical design of switching nodes in optical networks, in many cases, the Clos switching networks are successfully used. Clos switching networks are also used in Elastic Optical Networks that can effectively manage allocation of resources to individual multi-service traffic streams. The research outcomes presented in this article deal with the problems of finding traffic characteristics in blocking optical switching networks with known structures. This article aims at presenting an analysis of the influence of traffic management threshold mechanisms on the traffic characteristics of multi-service blocking Clos switching networks that are used in the nodes of elastic optical networks as well as their influence on the traffic efficiency of network nodes. The analysis was carried out on the basis of research studies performed in a specially dedicated purpose-made simulation environment. The article presents a description of the simulation environment used in the experiments. The study was focused on the influence of the threshold mechanism, which is one of the most commonly used and elastic traffic management mechanisms, and on the traffic characteristics of switching networks that service different mixtures of multi-service Erlang, Engset and Pascal traffic streams. The conducted study validates the operational effectiveness and practicality of the application of the threshold mechanism to model traffic characteristics of nodes in an elastic optical network.

This paper addresses the problem of optimally placing a limited number of wavelength converting nodes in multicast optical WDM networks. The problem is motivated by the high cost incurred in the usage of converters. We propose a Mixed Integer Linear Programming (MILP) formulation algorithm with the objective function of minimizing the total costs for a given set of multicast sessions in WDM networks. Numerical results show that the algorithm is tractable in networks with dozens of nodes.

Switching/merging node placement in survivable optical networks with SSP

Next-generation optical networks are expected to provide tremendous capacity in order to support upcoming traffic increases. Many technologies are currently being developed for optical transport networks in order to increase throughput, improve energy efficiency and simplify network deployment. The most important problem in current optical networks is transmission of Internet protocol (IP) traffic. Regardless of the tremendous throughput with optical fibers, switching nodes still limit overall network performance. Recently, optical burst switching technology has been developed to overcome this problem. Optical burst switching combines the advantages of both circuit switching and packet switching networks and provides good performance in terms of packet data transmission. Even though optical burst switching networks provide a good mechanism for IP traffic transmission, overall performance is still limited because of access networks. Existing passive optical networks based on Ethernet technology are not fully compatible with optical burst switching, which results in bottlenecks on the border between transport and access networks. In this paper, we present a new method of optical wavelength time-division multiple access (OWTDMA) for passive optical networks. The proposed approach can provide outstanding scalability of network resources and can increase throughput of the optical access network. In addition, we propose implementation of OWTDMA in edge nodes of optical burst switching networks to eliminate bottlenecks between transport and access networks. Simulation results prove the advantage of our proposed approach.

Clustering of nodes in optical networks has been proven to be an efficient way to serve end-to-end connectivity. However, clustering requires specific topological characteristics, or alternatively the introduction of significant alterations of an existing topology to achieve the expected performance improvements. The comparison of future dynamic optical networking technologies should therefore include in the set of initial assumptions, apart from the statistical properties of the traffic load, the network topology to draw conclusions regarding the efficiency as well as feasibility and scalability of the proposed solutions. In this article, we show how node clustering under the CANON architecture can be applied in real-life core networks and provide superior performance compared to conventional burst switching techniques in terms of blocking, resource utilization and power consumption.

Cascaded optical add-drop multiplexers (OADM) and optical cross connects (OXC) are key components in optical wavelength-division multiplex networks. OADMs with filtering of the passing signals and OXCs can be constructed by the use of wavelength-division multiplexers. Cascadability of multiplexers is therefore vital for the network performance. The actual transfer function of multiplexers is important, because the available end-to-end bandwidth between connected nodes in optical networks with cascades of OXCs and OADMs is given by the product of the transfer functions. In this paper, we demonstrate the cascadability of arrayed waveguide (AWG) multiplexers in a recirculating loop experiment at 10 Gbit/s and show that up to 40 multiplexers each with a 3-dB bandwidth of 1 nm and sharp roll-off characteristics can be passed penalty free.

This paper proposes a replica caching scheme according to status of neighboring nodes in optical grid networks. In optical grid networks, data files for job execution are replicated at multiple servers in order to distribute loads. Clients download replicas via lightpaths and store those replicas as necessary. The blocking probability of lightpath establishments depends on location of replica files. The case that a replica of a data file is stored in a client is ideal because blocking does not occur. However, the storage size of the client is limited. In order to efficiently use storage space of clients, the proposed scheme focuses on the fact that lightpath establishments are rarely blocked if data files are stored at neighboring nodes. In the proposed scheme, clients preferentially store data files which are not stored at neighboring nodes. By doing so, each client and its neighboring nodes store replicas of more different data files. As a result, when a client downloads a data file, at least one replica of the data file tends to be stored at the client or neighboring nodes. Through simulation experiments, we show that the proposed scheme improves the blocking probability of lightpath establishments efficiently.

In order to overcome the problems of poor real-time performance and low fitting degree of real-time detection methods for burst faults of key nodes in existing optical transmission networks, a new real-time detection method for burst faults of key nodes in optical transmission networks based on fireworks algorithm is proposed. The experimental results show that, compared with the traditional methods, the proposed real-time detection method for burst faults of key nodes in optical transmission network based on fireworks algorithm greatly improves the real-time performance and fitness, fully demonstrating that the proposed real-time detection method for burst faults of key nodes in optical transmission network based on fireworks algorithm has better detection effect.

In order to overcome the problems of poor real-time performance and low fitting degree of real-time detection methods for burst faults of key nodes in existing optical transmission networks, a new real-time detection method for burst faults of key nodes in optical transmission networks based on fireworks algorithm is proposed. The experimental results show that, compared with the traditional methods, the proposed real-time detection method for burst faults of key nodes in optical transmission network based on fireworks algorithm greatly improves the real-time performance and fitness, fully demonstrating that the proposed real-time detection method for burst faults of key nodes in optical transmission network based on fireworks algorithm has better detection effect.

Heuristic algorithms for efficient allocation of multicast-capable nodes in sparse-splitting optical networks

It is a significant challenge for telecommunication network operators to immediately restore communication services in the disaster area. To quickly recover telecommunication services in the affected area, this paper proposes a wired and wireless network cooperation system. When the wired communication for leaf nodes of optical tree networks is disrupted, surviving leaf nodes relay packets to and from these nodes via wireless bypass routes. The advantages of the proposed method are promptness and high-throughput, which is achieved with single-hop wireless bypass routes backhauled by wired networks. The optimal routes for wireless links are calculated to maximize the expected throughput by solving a binary integer programming problem. The proposed system is cost effective, because it can be deployed with minimum additional functions for leaf nodes of optical networks. To overcome the limitation of the proposed approach that the distribution of leaf nodes is determined by the demand distribution, additional recovery nodes can be deployed to improve the expected throughput. The numerical simulations including a medium access control level simulation conducting carrier sense multiple access with collision avoidance behavior showed that the proposed method can achieve a higher throughput than an existing bypass routing method, irrespective of the topology of the wired networks.

Utilizing Hilbert-pair-based digital filtering, intensity modulation and passive optical coupling, DSP-enabled optical filter- and O-E-O conversion-free ROADMs have been reported to offer excellent flexibility, colorlessness, gridlessness, contentionlessness, adaptability and transparency to physical-layer network characteristics. In this paper, DAC/ADC-induced impacts on ROADM performances are, for the first time, extensively explored in various IMDD-based optical network nodes. Numerical results indicate that for optical and electrical filters-free network nodes, ROADM has excellent performance robustness against the DAC characteristic variations, however high-resolution ADCs are required to improve the ROADM drop performances. While when equipping the network nodes with electrical baseband filters, ROADM drop operation performance robustness against the DAC/ADC characteristic variations is considerably improved and the low-cost ADCs with much lower resolutions and sampling speeds are also tolerable to be used in demodulating the dropped signals.

Secure networks composed of optical networks and quantum key distribution (QKD) are considered in this paper. A network layer structure is defined to focus on issues important with respect to the QKD network layer that controls QKD links and supplies a secure key for message transmission. A multiple-user QKD network layer can be constructed with timing synchronization, secure key sharing between indirectly connected nodes with switched QKD systems and key relays, and key management to control the volume of key generation and supply in response to the demand from optical network nodes.

Wavelength division multiplexing optical fiber networks rest mainly on two components, the network access station (NAS) and the optical network node (ONN). The design of the network depends heavily on the technology underlying these components, opto-electronics for NAS and photonics for ONN. These areas show large and fast development which will have an impact on network design. A series of characterization experiments were set up and conducted for characterization of the optical fiber and the laser diodes used. Among the important components in this project was a self study item of the student choice for one of the components of the link or the different aspects of the WDM systems and networks. The main features of the component with up to date information were required. The items ranged from fiber fabrication to soliton propagation in fibers and from WDM network layers to topologies as well as DFB and quantum well lasers.