<p>The task of the project was to evaluate a differential phase shift keying demodulation technique by replacing a Mach-Zehnder interferometer receiver with an optical filter (Fiber Bragg Grating). Computer simulations were made with single optical transmission, multi channel systems and transmission with combined angle/intensity modulated optical signals. The simulations showed good results at both 10 and 40 Gbit/s. Laboratory experiments were made at 10 Gbit/s to verify the simulation results. It was found that the demodulation technique worked, but not with satisfactory experimental results. The work was performed at Eindhoven University of Technology, Holland, within the framework of the STOLAS project at the department of Electro-optical communication.</p>

Recently there is an emergence of many Internet applications such as multimedia, video conferencing, distributed interactive simulations (DIS), and high-performance scientific computations like Grid computing. These applications require huge amount of bandwidth and a viable communication paradigm to coordinate with multiple sources and destinations. Optical networks are the potential candidates for providing high bandwidth requirement. Existing communication paradigms include broadcast, and multicast. Hence supporting these paradigms over optical networks is necessary. Multicasting over optical networks has been well investigated in the literature. QoS policies implemented in IP does not apply for Wavelength division multiplexed (WDM) or optical burst switched (OBS) networks, as the optical counterpart for store-and-forward model does not exist. Hence there is a need to provision QoS over optical networks. These QoS requirements can include contention, optical signal quality, reliability and delay. To support these diverse requirements, optical networks must be able to manage the available resources effectively. Destinations participating in the multicast session are fixed (or rather static). Due to the random contention in the network, if at least one or more destination(s) is not reachable, requested multicast session cannot be established. This results in loss of multicast request with high probability of blocking. Incorporating wavelength converters (WCs) at the core nodes can decrease the contention loss, however WCs require optical-electrical-optical (O/E/O) conversion. This increases the delay incurred by optical signal. On the other hand all-optical WCs are expensive and increase the cost of the network if deployed. Goal of this thesis is, to provide hop-to-hop QoS on an existing all-optical network (AON) with no WC and optical regeneration capability. In order to minimize the request vi Abstract vii lost due to contention in AON, we propose a variation of multicasting called Quorumcasting or Manycasting. In Quorumcasting destinations can join (or leave) to (or from) the group depending on whether they are reachable or not. In other words destinations have to be determined rather than knowing them prior, as in the case of multicasting. Quorum pool is minimum number of destinations that are required to be participated in the session for successful accomplishment of the job (k be the size of quorum pool). Providing QoS for manycasting over OBS has not been addressed in the literature. Given the multicast group (with cardinality m > k) and the number of destinations required to be participated, the contribution of this work is based on providing necessary QoS. In this thesis we study the behavior of manycasting over OBS networks. In OBS networks, packets from the upper-layer (such as IP, ATM, STM) are assembled and a burst is created at the edge router. By using O/E/O conversion at the edge nodes, these optical bursts are scheduled to the core node. Control header packet or burst header packet (BHP) is sent to prior to the transmission of burst. The BHP configures the core nodes and the burst is scheduled on the channel after certain offset time. In the first part of the thesis, we explain the different distributed applications with primary focus on Grid over OBS (GoOBS). We study the loss scenario due contention and inadequate signal quality for an unicast case in OBS network. We further extend this to manycasting. We modify the BHP header fields to make the burst aware of not only contention on the next-hop link, but also bit-error rate (BER). By using recursive signal and noise power relations, we calculate the BER (or q-factor) of the link and schedule the burst only if the required BER threshold is met. Thus all the bursts that reach the next-hop node ensure that contention and BER constraint are met. This are called “Impairment-Aware (IA) Scheduling”. Burst loss in the network increases due to BER constraint. Hence we propose algorithms to decrease the burst loss and simultaneously providing the sufficient optical signal quality. We propose three algorithms called IA-shortest path tree (IA-SPT), IA-static over provisioning (IA-SOP), and IA-dynamic membership (IA-DM). In IA-SPT destination set is sorted in the non-decreasing order of the hop-distance from source. First k of them are selected and bursts are scheduled to Abstract viii these destinations along the shortest path. In IA-SOP we select additional k0(_ m − k) destinations where k0 is the over provisioning factor. Over provisioning ensures that burst at least reach k of them, decreasing the contention blocking. However as the burst has to span more destinations, the fan-out of the multicast capable switch will be more and the BER could be high. In IA-DM destinations are dynamically added or removed, depending on contention and BER. Destination is removed and new destination is added based on the two constraints. Our simulation results shows that IA-DM out performs the other two algorithms in terms of request blocking. We show that IP-based many casting has poor performance and hence there is a need for supporting many casting over OBS networks. We verify our simulation results with the proposed analytical method. In the next part, we focus on provisioning QoS in many casting. QoS parameters considered for analysis include, signal quality i.e., optical signal to noise ratio (OSNR), reliability of the link and, propagation delay. In this work we consider application based QoS provisioning. In other words, given the threshold requirements of an application, our aim is to successfully schedule the burst to the quorum pool satisfying the threshold conditions. We use a de-centralized way of the scheduling the burst, using BHP. With the help of local-network state information, the burst is scheduled only if it satisfies multiple set of constraints. Corresponding reception of burst at the node ensures that all the QoS constraints are met and burst is forwarded to the next hop. QoS attributes are either multiplicative or additive. Noise factor of the optical signal and reliability factor are multiplicative constraints, where as propagation delay is additive. We define a path information vector, which provides the QoS information of the burst at every node. Using lattice theory we define an ordering, such that noise factor and propagation delay are minimum and reliability is maximum. Using path algebra we compute the overall QoS attributes. Due to multiple set of constraints, the request blocking could be high. We propose algorithms to minimize request blocking for Multiple Constrained Many cast Problem (MCMP). We propose two algorithms MCM-SPT and MCM-DM. We consider different set of service thresholds, such as real time and data service thresholds. Real time services impose restriction on signal quality and the propagation delay. On the other hand Abstract ix data services require high reliability and signal quality. Our simulation study shows that MCM-SPT performs better than MCM-DM for real-time services and the data services can be provisioned using MCM-DM.

The task of the project was to evaluate a differential phase shift keying demodulation technique by replacing a Mach-Zehnder interferometer receiver with an optical filter (Fiber Bragg Grating). Computer simulations were made with single optical transmission, multi channel systems and transmission with combined angle/intensity modulated optical signals. The simulations showed good results at both 10 and 40 Gbit/s. Laboratory experiments were made at 10 Gbit/s to verify the simulation results. It was found that the demodulation technique worked, but not with satisfactory experimental results. The work was performed at Eindhoven University of Technology, Holland, within the framework of the STOLAS project at the department of Electro-optical communication.

The objective of this thesis is to study of optical effects, arising in the form of non-uniform waveguide structure, complicated refractive index profiles or due to pulse propagation in dense wavelength division multiplexing (DWDM) optical communication systems. These effects are important and critically influence the performance of DWDM optical systems. A comprehensive survey of current literature on optical effects due to nonuniform optical structure and nonlinear optical effects is first done, showing their advantages and disadvantage in optical communication systems. A survey on methods of optical waveguide analysis is also done. The main contribution has been made to three main aspects of the problem : Accurate analysis of uniform/non-uniform optical waveguides with arbitary refractive index profiles Pulse propagation and distortion in DWDM Raman amplification systems. Use of non-uniform FBG to compensate for pulse distortion We study several existing analytical techniques developed so far for analyzing the mode of non-uniform optical waveguide structures. Later, we verify the analytical results by finite element method (FEM). The convergence study is also carried out. A new computational technique is proposed modifying the finite element method to analyze complex refractive index profiles required for the analysis, namely single mode step index profile, multi clad fiber, W -profile, chirp profile etc. An accuracy of 10−4 in the calculation of propagation constant/eigen-value is demonstrated. Dispersion characteristics of optical fibers w.r.t. different profile parameters is evaluated. A modification to scalar BPM is proposed and applied to study the effects of inhomogeneities along the propagation direction. The applicability and accuracy of the method is tested using integrated optic waveguide devices, namely, graded index slab waveguide. The proposed BPM uses Fourier decomposition of the transverse field. Coupled mode theory (CMT) of optical waveguides in non-homogeneous optical medium is applied to study the interaction of lightwaves propagation together such as in a DWDM system. The BPM results is verified by CMT. The inhomogeneous waveguide theory is extended to study pulse propagation in DWDM optical communication system. Nonlinear optical effects are an important aspects of DWDM systems with fiber Raman amplifier. Finite difference time domain (FDTD) method is necessary to study these nonlinear optical effects as other conventional methods are not suitable here. Here, we discuss DWDM optical communication systems due to nonlinearity in the form of SRS effect. In case of FRA, we study the various kinds of fiber profile design parameters, for the purpose to achieve and extend the flat gain bandwidth over the EDFA window. We also propose and study, a new bi-directional optical fiber transmission scheme with various constraints, using Raman amplification process with and without pump depletion. Our scheme, provides an advantage like high SNR, low pump induced noise, for long-haul communication link. We find that, there is a quite significant crosstalk and power coupling among the dense DWDM channels but earlier discussed BPM fails to account for possible interference effects among the channels. To reduce the harmful nonlinear optical effects like four wave mixing (FWM), we need to deploy a high chromatic dispersion fiber, which will ultimately lead to high pulse walk-off rate among the DWDM channels; hence for high bit rate long haul systems, walk-off effect can not be ignored. Application of FDTD provided an improved insight into the effect of GVD on stimulated Raman scattering crosstalk than different modulation techniques and line codes. It is shown through analysis that pulse walk-off phenomena may distort the data asymmetrically; especially for case of wide-band DWDM transmission system. Hence, the pulse walk-off effect should be considered in future systems containing optical amplifier. It is shown, that large walk-off rate may reduce the crosstalk among DWDM channels but tends to increase the asymmetric pulse distortion. Data may lose due to high walk-off effect. We also investigate channel addition/removal process in DWDM fiber Raman amplifier. We also demonstrate that the pulse walk-off effect tends to lead significantly to positive chirp for higher frequency channels. This feature can be exploited to overcome the chromatic dispersion effects in DWDM transmission systems. Pulse walk-off induced chirp, can be compensated by using the nonuniform fiber Bragg grating (NUFBG). The CMT due to periodic perturbation of the circular cylindrical waveguide structures is applied here. Here, we discuss the function of fiber Bragg grating as a transmission versus reflecting grating filter. We also discuss, FBG application to gain flattening of an EDFA window as well as how the group velocity dispersion (GVD) will be affected with bandwidth and coupling coefficient. We develop a new analytical technique to estimate the bandwidth of FBG based optical system. Finally, we investigate the dispersion compensation properties, pulse distortion, peak reflectivity analysis in uniform/non-uniform FBG due to an uniform/non-uniform incoming signal. More complicated refractive index profile can significantly reduce the GVD as well as side lobes intensity. Dispersion characteristic due to an arbitrary refractive index profile is discussed in details for the case of non-uniform FBG. Thus, we concluded that wide band DWDM optical communication system need to closely take into account various inhomogeneities and nonlinearities of optical fibers w.r.t. wave and pulse propagation.

This thesis presents an introduction to the design and simulation of a novel class of integrated photonic phased array switch elements. The main objective is to use nano-electromechanical (NEMS) based phase shifters of cascaded under-etched slot nanowires that are compact in size and require a small amount of power to operate them. The structure of the switch elements is organized such that it brings the phase shifting elements to the exterior sides of the photonic circuits. The transition slot couplers, used to interconnect the phase shifters, are designed to enable biasing one of the silicon beams of each phase shifter from an electrode located at the side of the phase shifter. The other silicon beam of each phase shifter is biased through the rest of the silicon structure of the switch element, which is taken as a ground. Phased array switch elements ranging from 2×2 up to 8×8 multiple-inputs/multiple-outputs (MIMO) are conveniently designed within reasonable footprints native to the current fabrication technologies. Chapter one presents the general layout of the various designs of the switch elements and demonstrates their novel features. This demonstration will show how waveguide disturbances in the interconnecting network from conventional switch elements can be avoided by adopting an innovative design. Some possible applications for the designed switch elements of different sizes and topologies are indicated throughout the chapter. Chapter two presents the design of the multimode interference (MMI) couplers used in the switch elements as splitters, combiners and waveguide crossovers. Simulation data and design methodologies for the multimode couplers of interest are detailed in this chapter. Chapter three presents the design and analysis of the NEMS-operated phase shifters. Both simulations and numerical analysis are utilized in the design of a 0º-180º capable NEMS-operated phase shifter. Additionally, the response of some of the designed photonic phased array switch elements is demonstrated in this chapter. An executive summary and conclusions sections are also included in the thesis.

O surgimento de novos serviços de telecomunicações tem provocado um enorme aumento no tráfego de dados nas redes de transmissão. Para atender a essa demanda crescente, novas tecnologias foram desenvolvidas e implementadas ao longo dos anos, sendo que um dos principais avanços está na área de transmissão óptica, devido à grande capacidade de transporte de informação da fibra óptica. A tecnologia que melhor explora a capacidade desse meio de transmissão atualmente é a multiplexação por divisão de comprimento de onda ou Wavelength Division Multiplexing (WDM) que permite a transmissão de diversos sinais utilizando apenas uma fibra óptica. Redes ópticas WDM se tornaram muito complexas, com enorme capacidade de transmissão de informação (terabits por segundo), para atender à explosão de necessidade por largura de banda. Nesse contexto, é de extrema importância que os recursos dessas redes sejam utilizados de forma inteligente e otimizada. Um dos maiores desafios em uma rede óptica é a escolha de uma rota e a seleção de um comprimento de onda disponível na rede para atender uma solicitação de conexão utilizando o menor número de recursos possível. Esse problema é bastante complexo e ficou conhecido como problema de roteamento e alocação de comprimento de onda ou, simplesmente, problema RWA (Routing and Wavelentgh Assignment problem). Muitos estudos foram realizados com o objetivo de encontrar uma solução eficiente para esse problema, mas nem sempre é possível aliar bom desempenho com baixo tempo de execução, requisito fundamental em redes de telecomunicações. A técnica de algoritmo genético (AG) tem sido utilizada para encontrar soluções de problemas de otimização, como é o caso do problema RWA, e tem obtido resultados superiores quando comparada com soluções heurísticas tradicionais encontradas na literatura. Esta dissertação apresenta, resumidamente, os conceitos de redes ópticas e de algoritmos genéticos, e descreve uma formulação do problema RWA adequada à solução por algoritmo genético. / The advent of new telecommunication services resulted in a huge increase of data traffic in the transmission networks. New technologies were developed and implemented over the years to attend to this growing demand, and the optical transmission technology stands. It has advanced greatly, due to the optical fibers large capacity of information transmission. Actually, the best technology to exploits the capacity of the fiber is the wavelength-division multiplexing (WDM), allowing the transmission of multiple signals over a single optical fiber. The WDM optical networks have become very complex, with huge capacity (terabits per second), to attend the ever growing need for bandwidth. In this context, it is extremely important to use the networks resources in an intelligent and optimized way. One of the biggest challenges in an optical network is choosing a route, and selecting a available wavelength on the network to attend a connection request using the least amount of resources. This problem is quite complex, and is known as the routing and wavelength assignment problem or simply RWA problem. Many studies were conducted in order to find an efficient solution to this problem, but it is not always possible to combine good performance with low execution time, a fundamental requirement in telecommunications networks. Genetic Algorithms have been used to solve hard optimization problems, as is the case of the RWA problem, and has produced remarkable results when compared to traditional heuristics found in the literature. This work presents an overview of the concepts of optical networks and genetic algorithms, and describes a formulation of RWA problem that is adequate for solution by genetic algorithm.

Optical WDM network is the suitable transport mechanism for ever increasing bandwidth intensive internet applications. The WDM technique transmits the data over several different wavelengths simultaneously through an opticalfiber and the switching is done at wavelength level. The connection between the source and destination is called the light path. Since the WDM network carries huge amount of tra c, any failure can cause massive data loss. Therefore protecting the network against failure is an important issue. Maintaining high level of service availability is an important aspect of service provider. To provide cost effective service, all-optical network is the suitable choice for the service provider. But in all optical network, the signals are forced to remain in optical domain from source to destination. In the firrst part of the thesis, we deal the physical layer impairments (PLIs) aware shared-path provisioning on a wavelength routed all-optical networks. As the signal travels longer distances, the quality of the signal gets degraded and the receiver may not be able to detect the optical signal properly. Our objective is to establish a light path for both the working path and protection path with acceptable signal quality at the receiver. We propose an impairment aware integer linear programming (ILP) and impairment aware heuristic algorithm that takes into account the PLIs. The ILP provides the optimal solution. It is solved using IBM ILOG CPLEX solver. It is intractable for large size net-work. Therefore we propose the heuristic algorithm for large size network. It is evaluated through discrete-event simulation. But the algorithm provides only the suboptimal solution. To know the performance of this algorithm, the simulation result is compared with the optimal solution. We compute total blocking probability, restoration delay, computation time, and connection setup delay with respect to network load for the heuristic algorithm. We compare the performance of shared-path protection with dedicated-path protection and evaluate the percentage of resource saving of shared-path protection over the dedicated-path protection. In the second and third part of the thesis, we address the issues related to flexible-grid optical networks. In wavelength routed optical network, the bandwidth of each wavelength is fixed and rigid. It supports coarse grained tra c grooming and leads to ancient spectrum utilization. To overcome this, flexible-grid optical networks are proposed. It supports flexible bandwidth, and ne grained tra c groom In the second part of the thesis, we address the routing and spectrum allocation (RSA) algorithm for variable-bit-rate data tra c for flexible-grid optical networks. The RSA problem is NP-complete. Therefore a two-step heuristic approach (routing and spectrum allocation) is proposed to solve the RSA problem. The first step is solved by using a classical shortest path algorithm. For the second step we propose two heuristic schemes for frequency-slot allocation: (i) largest number of free frequency-slot allocation scheme and (ii) largest number of free frequency-slot maintaining scheme. As the network load increases, the spectrum is highly fragmented. To mitigate the fragmentation of the spectrum, we propose a xed-path least-fragmentation heuristic algorithm which fragments the spectrum minimally. It also supports varying-bit-rate tra c and also supports dynamic arrival connection requests. Through extensive simulations the proposed algorithms have been evaluated. Our simulation results show that the algorithms perform better in terms of spectrum utilization, blocking probability, and fraction of fragmentation of the spectrum. The spectrum utilization can reach up to a maximum of 92% and that only 71% of the spectrum is fragmented under maximum network load condition. Finally in the third part of the thesis, we discuss PLIs-aware RSA for the transparent exible-grid optical network. In this network, not only the optical signal expected to travel longer distance, but also to support higher line rates, i.e., data rate is increased up to 1 Tb/s. In such a high data rate, the optical signals are more prone to impairments and noises. As the transmission distance increases, optical signals are subject to tra-verse over many bandwidth-variable wavelength cross connects (BV-WXC) and multiple fibber spans due to which the PLIs get accumulated and are added to the optical signal. These accumulated impairments degrades the signal quality to an unacceptable level at the receiver, the quality of transmission falls below the acceptable threshold value, and the receiver may not be able to detect the signal properly. Therefore our objective is to develop an impairment aware RSA algorithm which establishes the QoT satisfied empathy based on the available resources and the quality of the signal available at the receiver. We formulate the PLIs-RSA problem as an ILP that provides an optimal solution. The optimal solution is obtained by solving the ILP using IBM ILOG CPLEX optimization solver. Since ILP is not efficient for large-size networks, we propose a heuristic algorithm for such a large-size networks. The signal power is measured at the receiver and the connection is established only when the signal power lies above the threshold value. The heuristic algorithm is evaluated through discrete-event simulation. It gives the sub-optimal solution. The simulation result is compared with optimal solution. The result shows that heuristic algorithm performs closer to the ILP. We compute the total blocking probability versus the network load for different spectrum allocation schemes. Total blocking probability is the sum of frequency-slot blocking probability and QoT blocking probability. We compute spectrum efficiency for the proposed algorithm. We also compare our algorithm with the existing routing and spectrum allocation algorithm, and the result shows that our algorithm outperforms the existing algorithms in terms of blocking probability and spectrum utilization.

[pt] Neste trabalho realizamos uma prova da viabilidade de um sistema de comunicação óptica com capacidade de transmissão de 50 Gbit/s em uma largura de banda de 50 GHz utilizando o legado dos sistemas com taxas de 10 Gbit/s. Uma série de configurações experimentais foi testada em uma ordem de complexidade crescente para verificar separadamente as etapas e as técnicas aplicadas para o aumento da capacidade de transmissão de dados e a eficiência espectral. Em alguns casos, a curva característica resultante da configuração backto- back do analisador de taxa de erro de bit média foi utilizada como referência para comparação das configurações experimentais realizadas durante o trabalho. / [en] In this work we perform a proof of feasibility of 50 Gbit/s transmission within a 50 GHz optical bandwidth exploring the 10 Gbit/s legacy. A series of experimental configurations were tested in an order of increasing complexity to verify separately the steps and applied techniques for increasing data transmission capacity and spectral efficiency. In some cases, the comparison of experimental configuration was made using the back-to-back configuration of the Bit Error Rate Tester.

O surgimento de novos serviços de telecomunicações tem provocado um enorme aumento no tráfego de dados nas redes de transmissão. Para atender a essa demanda crescente, novas tecnologias foram desenvolvidas e implementadas ao longo dos anos, sendo que um dos principais avanços está na área de transmissão óptica, devido à grande capacidade de transporte de informação da fibra óptica. A tecnologia que melhor explora a capacidade desse meio de transmissão atualmente é a multiplexação por divisão de comprimento de onda ou Wavelength Division Multiplexing (WDM) que permite a transmissão de diversos sinais utilizando apenas uma fibra óptica. Redes ópticas WDM se tornaram muito complexas, com enorme capacidade de transmissão de informação (terabits por segundo), para atender à explosão de necessidade por largura de banda. Nesse contexto, é de extrema importância que os recursos dessas redes sejam utilizados de forma inteligente e otimizada. Um dos maiores desafios em uma rede óptica é a escolha de uma rota e a seleção de um comprimento de onda disponível na rede para atender uma solicitação de conexão utilizando o menor número de recursos possível. Esse problema é bastante complexo e ficou conhecido como problema de roteamento e alocação de comprimento de onda ou, simplesmente, problema RWA (Routing and Wavelentgh Assignment problem). Muitos estudos foram realizados com o objetivo de encontrar uma solução eficiente para esse problema, mas nem sempre é possível aliar bom desempenho com baixo tempo de execução, requisito fundamental em redes de telecomunicações. A técnica de algoritmo genético (AG) tem sido utilizada para encontrar soluções de problemas de otimização, como é o caso do problema RWA, e tem obtido resultados superiores quando comparada com soluções heurísticas tradicionais encontradas na literatura. Esta dissertação apresenta, resumidamente, os conceitos de redes ópticas e de algoritmos genéticos, e descreve uma formulação do problema RWA adequada à solução por algoritmo genético. / The advent of new telecommunication services resulted in a huge increase of data traffic in the transmission networks. New technologies were developed and implemented over the years to attend to this growing demand, and the optical transmission technology stands. It has advanced greatly, due to the optical fibers large capacity of information transmission. Actually, the best technology to exploits the capacity of the fiber is the wavelength-division multiplexing (WDM), allowing the transmission of multiple signals over a single optical fiber. The WDM optical networks have become very complex, with huge capacity (terabits per second), to attend the ever growing need for bandwidth. In this context, it is extremely important to use the networks resources in an intelligent and optimized way. One of the biggest challenges in an optical network is choosing a route, and selecting a available wavelength on the network to attend a connection request using the least amount of resources. This problem is quite complex, and is known as the routing and wavelength assignment problem or simply RWA problem. Many studies were conducted in order to find an efficient solution to this problem, but it is not always possible to combine good performance with low execution time, a fundamental requirement in telecommunications networks. Genetic Algorithms have been used to solve hard optimization problems, as is the case of the RWA problem, and has produced remarkable results when compared to traditional heuristics found in the literature. This work presents an overview of the concepts of optical networks and genetic algorithms, and describes a formulation of RWA problem that is adequate for solution by genetic algorithm.

The objective of this dissertation is to investigate the factors that influence the design of wide-area wavelength division multiplexing (WDM) optical networks. Wide-area networks are presented as communication networks capable of transporting voice and data communication over large geographical areas. These networks typically span a whole country, region or even continent.The rapid development and maturation of WDM technology over the last decade have been well-received commercially and warrants the development of skills in the field of optical network design.The fundamental purpose of all communication networks and technologies is to satisfy the demand of end-users through the provisioning of capacity over shared and limited physical infrastructure. Consideration of the business aspects related to communications traffic and the grooming thereof are crucial to developing an understanding of customer requirements in terms of the selection and quality of services and applications. Extensive communication networks require complex management techniques that aim to ensure high levels of reliability and revenue generation.An integrated methodology is presented for the design of wide-area WDM optical networks. The methodology harnesses physical, logical, and virtual topologies together with routing and channel assignment (RCA) and clustering processes to enhance objectivity of the design process. A novel approach, based on statistical clustering using the Ward linkage as similarity metric, is introduced for solving the problem of determining the number and positions of the backbone nodes of a wide-area network, otherwise defined as the top level hub nodes of the multi-level network model. The influence of the geographic distribution of network traffic, and the intra/inter-cluster traffic ratios are taken into consideration through utilisation of modified gravity models and novel network node weighting. / Dissertation (MEng)--University of Pretoria, 2005. / Electrical, Electronic and Computer Engineering / unrestricted