TCP performance in multi-EPON access networks under different optical core switching paradigms

Optical burst switching (OBS) is a switching technology that can efficiently operate in the optical core network using WDM technology and can also develop the future optical internet. The OBS switching technology presents a trade-off between the two switching technologies: optical circuit switching (OCS) and optical packet switching (OPS). This switching approach increases resource utilization compared with OCS technology and avoids the technological barriers of OPS networks. In OBS networks, many packets are assembled into one data burst at the edge node and a burst header packet (BHP) is sent before the burst transmission by an offset time in order to reserve the required resources in core nodes. This can cause security issues in the network and more specifically denial of service attacks that can occur if a source node is compromised. In this paper, we study a specific denial of service attack which we refer to as "BHP flooding attack", which prevents legitimate traffics from reserving the required resources at intermediate core nodes. We also propose the design of a reconfigurable BHP flooding countermeasure module that allows to counter against this type of attacks in an OBS network.

In optical switching paradigm, there are three approaches; optical circuit switching, optical packet switching, and optical burst switching (OBS). OBS is a technology that takes the best features of packet switching and circuit switching by providing a flexible infrastructure for carrying IP traffic and at the same time ensuring efficient utilization of the fiber link. In this paper we propose an effective contention resolution scheme to achieve minimum burst loss at core nodes. Algorithm of zero burst loss for bimodal burst switching is extended from single node star network topology to two node star topology. Scheduling at the core node, typically an optical cross connect (OXC), is performed in two modes i.e., Mode 0 & Mode 1. Mode 0 is for burst transmission between all directly connected edge nodes whereas Mode 1 is for burst transmission through remote core node to a particular destination edge node. Synchronization of the reservation request between the core nodes is a critical task in such a topology. In order to achieve minimum burst loss, using a small length of fiber delay line in recirculation mode results in an effective strategy.

We have already proposed a Gigabit Ethernet-Optical Switched Access Network (GE-OSAN) architecture that realizes longer transmission distances and achieves higher security than the conventional Passive Optical Network (PON). To confirm the technical feasibility of the architecture, we introduce here a GE-OSAN prototype system for down-stream transmission. We present the Optical Switching Module (OSM), which uses (Pb,La)(Zr,Ti)O3 (PLZT) optical packet switches, and show that it realizes switching within 10ns of the designed position in the Inter-Frame Gap (IFG). We also introduce an Optical Network Unit (ONU) with optical burst receiver that uses off-the-shelf commercial devices to reduce its cost; two types of an optical to electrical signal (O/E) converter are implemented for performance comparison. After testing both of them, we select the one that satisfies our acquisition time requirement of 64ns.

An ILP-based CapEx and OpEx efficient multi-stage TDM/TWDM PON design methodology

Power budget of direct-detection ultra-dense WDM-Nyquist-SCM PON with low-complexity SSBI mitigation

The limited scheme with excess distribution (LS-ED), being a superior grant-sizing protocol in terms of average delay minimization while providing fairness and quality of service (QoS) satisfaction in Ethernet passive optical networks (EPONs), has already been deployed in diverse technologies. An analysis of the protocol is of paramount importance for obtaining deeper insight into several network design and deployment aspects. Thus, the primary objective of this paper is to perform a rigorous mathematical analysis of LS-ED, for the first time to the best of our knowledge, with the help of a discrete time Markov chain (DTMC). Owing to the dependency of the grant size of a particular optical network unit (ONU) on the queue length of other ONUs, the exact queuing analysis of this scheme necessitates solving a Markov model of the same dimension as the number of ONUs, which makes the analysis complicated and mathematically intractable. A lower-dimensional valid approximation model may allow us to perform such an analysis. In this paper, for the first time to the best of our knowledge, we present an analysis of the scheme, wherein the system dynamics are captured from the perspective of the scheduler leading to an approximate model. The model aids in obtaining a closed-form expression for the average cycle time and throughput along with closed-form expressions for average buffer occupancy and average delay at low load. Finally, we provide useful design insights for EPONs employing LS-ED.

Optical Circuit Switching (OCS) provides a good support for the services requiring high bandwidth and Quality of Service (QoS). When the character of network traffic changes from voice-oriented to data-oriented, OCS must cooperate with new switching mechanisms such as Optical Burst Switching (OBS) to improve performance. So a Hybrid Optical Switching (HOS) system is proposed. In HOS, different traffics are classified and both OBS and OCS can be processed. At the same time, delay buffer mechanisms, which are Fiber Delay Line (FDL) and electronic buffering, are introduced in HOS to solve the contention of OBS bursts. And a wavelengths classification mechanism, which assigns some wavelengths dedicated to OCS traffics, is provided to guarantee the QoS of OCS. To analyze the performance in theory, models based on Markovian theory and iterative method are provided. The results got by theory are accurate compared with simulations. The use of delay buffer can decrease the blocking probability of OBS and increase resource utilization. Wavelength classification mechanism benefits to improve OCS's performance.

The potential for sharing infrastructure costs between a large number of customers and the high data rates allowed by optical fibres make passive optical networks (PONs) an attractive solution to the problem of upgrading current copper-based access networks. Optically-amplified, long reach, time division multiple access (TDMA) PONs or 'SuperPONs' offer the potential to further reduce bandwidth transport costs by enabling the direct connection of access networks and inner core networks, thereby eliminating the costs of the outer core/metro backhaul network. The use of dense wavelength division multiplexing (DWDM) could also allow sharing the same feeder fibre and PON head end equipment between a number of such TDMA SuperPONs, each working at different ITU-grid wavelengths. However, a cost effective access solution should employ a customer optical network unit (ONU), which is independent of the PON wavelength, or colorless, in order to reduce the high inventory and deployment costs of using expensive, wavelength-specified sources at the customer. In this paper we demonstrate for the first time the use of a monolithically-integrated, electroabsorption modulator-semiconductor optical amplifiers (EAM-SOAs) as a colorless ONU in a high performance DWDM SuperPON system. These compact devices offer the potential for low-cost optoelectronic integration with other ONU components together with the ability to modulate at rates up to 10Gbps and beyond. We have used this approach to investigate the feasibility of supporting up to 17 SuperPONs from a single feeder fibre and PON head end, each of 100km-reach accommodating 512 users at 2.5Gb/s or 128 at 10Gb/s.

Modeling and performance analysis of a data center network (DCN) combining optical packet switching (OPS) and agile optical circuit switching (OCS) in a unified platform are presented. Hybrid optoelectronic packet routers which combine optical switching fabric and electronic buffers are used, supporting 100 Gbit/s (25 Gbit/s × 4 wavelengths) optical packets. OCS transmission is implemented by a novel “express path,” (ExP) an instant wavelength path on demand, which enables certain flows following the ExP with no packet contention, whereas ordinary OPS packets are switched with a deflection routing algorithm and buffering for contention resolution. This architecture can support OCS for large and reliability-sensitive data transferring and OPS for latency-sensitive applications. The performance of the network is investigated via numerical simulations of top-of-rack switch (ToR)-to-ToR traffic in an N-dimensional (up to 4096 node) torus topology. We have made the following observations: 1) torus topology results in high throughput and can be scaled out by increasing the dimension; 2) contention resolution between OPS and agile OCS is needed and our proposed strategies enhance the overall performance; and 3) in a lightly to moderately loaded case of OPS traffic, more than 100 ExPs can be supported simultaneously with acceptable affect on OPS.

This paper describes a novel channel allocation and DBA (Dynamic Bandwidth Allocation) mechanism for ECDM-PON (Electric Code Division Multiplex — Passive Optical Network) systems. In the current ECDM-PON systems, each ONU (Optical Network Unit) is limited to 2 or 3 CDM channels. This is because (fixed channel) CDM transmitters are expensive, and tunable CDM transmitters even more expensive. With a small number of CDM channels, the bandwidth utilization ratio is restricted by channel blocking. Our proposed mechanisms can reduce the channel blocking ratio without increasing the number of CDM transmitters or using tunable CDM transmitters. To clarify the advantages of the proposed system performance, we have evaluated the channel non-blocking ratio (Rn) and wasted resource ratio (Rw) when some users request bandwidth more than 100%. Evaluation of the non-blocking ratio, Rn shows that the proposed mechanisms approach the performance of a system with tunable CDM transmitters when the number of ONUs with over 100% traffic load is small. We have also simulated throughput for uniform traffic. In addition to these evaluations, we implemented our proposed mechanism on an FPGA (Field Programming Gate Array) and evaluated the calculation speed to allocate timeslots on CDM channels and a timeline.

We develop an integrated optical packet and circuit switching node in which signaling via optical packet switching is implemented. We demonstrate 80 (8λ × 10) Gbit/s optical packet switching with <;1E-4 packet-loss-rate, 8-lightpaths establishment, and dynamic wavelength-resources control.

Hybrid optical networks (HONs) are the combination of different switching mechanisms such as Optical Circuit Switching (OCS), Optical Burst Switching (OBS) and Optical Packet Switching (OPS). The basic idea behind HONs is to integrate the advantages of different switching mechanisms. HONs are classified into three classes as: (1) parallel hybrid optical networks, (2) integrated hybrid optical networks, (3) client-server hybrid optical networks. In this article, we describe these classes in detail and a new integrated hybrid optical network called EISM (Efficient Integration of Switching Mechanisms in all-optical networks) is introduced. The proposed EISM method could perform optimally while considering some implementation criteria; such criteria include network bandwidth utilization, implementation complexity and also the domain of network performance by combining four switching mechanisms-OCS, long-OBS, short-OBS and OPS.

Energy-efficient passive optical networks (PONs) have gained significant interest since they are estimated to be the largest energy consumers among the wired access networks for the next ten years. In Ethernet PONs (EPONs), the equipment placed at the customer premises, i.e., the optical network units (ONUs), has been shown to be responsible for almost 65% of the total EPON power consumption. Sleep mechanisms, implemented at ONUs’ side, can contribute to the EPONs’ energy-efficiency. However, the tradeoff between energy saving and quality of service (QoS) requirements should be carefully tuned, especially when the downstream transmission is considered, to achieve the desirable results. In this paper, we propose a general framework that exploits formal methods as an approach for EPONs’ energy-efficiency. The idea is to build a holistic model representing both the state machine of the ONU in its details as well as the ONU’s communication with the optical line terminal under EPONs’ specifications. Verification results reveal that, guided by the QoS constraints and considering the transitions’ cost, an “aggressive” policy defining long sleep periods of the 100 ms order can be up to 24.9% beneficial for energy-efficiency compared to a “conservative” policy with sleep periods of the 10 ms order.

The service evolution and the rapid increase in traffic levels fuel the interest toward switching paradigms enabling the fast allocation of Wavelength Division Multiplexing WDM channels in an on demand fashion with fine granularities (microsecond scales). For this reason, in the last years, different optical switching paradigms have been proposed (Sabella et al., 2000): optical-packet switching (OPS), optical-burst switching (OBS), wavelength-routed OBS, etc. Among the various all-optical switching paradigms, OPS attracts increasing attention. Owing to the high switching rate, Semiconductor Optical Amplifier (SOA) is a key technology to realize Optical Packet Switches. We propose some Optical Packet Switch (OPS) architectures and illustrate their realization in SOA technology. The effectiveness of the technology in reducing the power consumption is also analyzed. The chapter is organized in three sections. The main blocks (Switching Fabric, Wavelength Conversion stage, Synchronization stage) of an OPS are illustrated in Section 2 where we also show some examples of realizing wavelength converters and synchronizers in SOA technology. Section 3 introduces SOA-based single-stage and multi-stage switching fabrics. Finally the SOA-based OPS power consumption is investigated in Section 4.

On burst assembly in optical burst switching networks—A performance evaluation of just-enough-time