Abstract

Toward the forthcoming spatial division multiplexing (SDM) era, a spatial channel network (SCN) was recently proposed in which the optical layer is explicitly decoupled from the hierarchical SDM and wavelength division multiplexing (WDM) layers, and an optical node evolves into a spatial cross-connect (SXC) and wavelength cross-connect (WXC) to achieve a hierarchical optical cross-connect (HOXC). In this paper, we perform a technoeconomic analysis of SCNs comprising different HOXC architectures employing different planning policies on how the degree of spectral grooming by WXCs and spatial bypassing by SXCs in an SCN affects the total required number of spatial lanes (SLs), whose physical entity is the core in parallel single-mode fibers or a multicore fiber and the total node cost in the SCN. Toward this end, we develop a routing and SDM/WDM multilayer resource assignment (RSWA) heuristic in which spatial bypassing and spectral grooming are performed such that the required number of SLs is minimized. Using RSWA with four planning policies, i.e., express-only, express/local-hybrid (spatial-bypass-oriented), express/local-hybrid (spectral-grooming-oriented), and local-only, we compare the performance levels of a high-port-count matrix-switch-based HOXC and core selective switch (CSS) based HOXC with those of the baseline-stacked conventional WXCs as the network traffic load increases. Here, a CSS is a new type of spatial switch, which is the counterpart to a wavelength selective switch in a current WDM network. We clearly show that hierarchical spatial bypassing and spectral grooming are beneficial in terms of the required number of SLs and network-total node cost when the required number of SLs between optical nodes in an SCN is equal to or greater than roughly 10.

© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1. INTRODUCTION

The history of transmission technology is also the history of the development of multiplexing technology, and optical fiber transmission technology is no exception. In this decade, considerable research efforts have been focused on space division multiplexing (SDM) technology as the last remaining multiplexing domain in optical fiber transmission [1,2]. Per Fig. 1, we reflect on the history of the development of optical network technology from multiplexing technology and corresponding optical node architecture viewpoints and discuss the future of optical transport network architecture toward the forthcoming SDM era. In early optical transmission systems, time division multiplexing (TDM) technology was employed to increase the system capacity. With the advent of erbium-doped fiber amplifiers, which are capable of amplifying over a wide wavelength range, wavelength division multiplexing (WDM) technology was put into practical use at the beginning of the 2000s. Numerous WDM channels are multiplexed into a fiber in the spectral domain. Rapid growth in the WDM capacity has necessitated the deployment of a tremendous amount of electrical terminating and switching equipment such as optical transponders, synchronous digital hierarchy cross-connects (SDH-XCs), and Internet protocol (IP) routers, where all network traffic entering the node is processed in the electrical domain.

This has incurred serious cost and power-consumption problems. Optical bypass technology based on optical switches has developed in order to reduce the burden of SDH-XCs and IP routers [25]. Nowadays, the most popular transport network architecture is the hierarchical IP [over an optical transport network (OTN)] over a WDM network that comprises IP routers [and optical channel data unit (ODU) XCs] and reconfigurable optical add/drop multiplexers (ROADMs) based on wavelength selective switches (WSSs). In such a network, an optical channel (OCh) is established between a source/destination pair, where a sufficient amount of traffic to fill almost the entire capacity of the wavelength exists, and the OCh bypasses the overlying OTN or IP layers in the optical domain. Simple two-degree ROADMs were first introduced to WDM ring networks, and they have evolved to multidegree ROADMs, which are referred to as wavelength cross-connects (WXCs), to be applied in mesh optical networks. Digital coherent transmission technology with its associated sophisticated modulation schemes and digital signal processing (DSP) has enhanced network efficiency by introducing elasticity and adaptation into WDM networks [68].

 

Fig. 1. History and perspective of optical network evolution.

Download Full Size | PPT Slide | PDF

In the same way as when WDM technology was introduced to optical networks, currently-in-progress advances in SDM technology present a challenge in that they necessitate the deployment of large WXCs with tremendous numbers of conventional WSSs. In order to address this problem, larger-scale WDM/SDM cross-connects based on, for example, the joint switching of spatial superchannels [9] and a subsystem-modular WXC architecture [10] have been investigated. It should be noted that, in such WDM/SDM cross-connects, all network traffic entering the node is processed in the fine-granular wavelength domain.

On the other hand, recalling that the introduction of WDM required an optical bypass in the WDM layer based on ROADMs/WXCs, introducing a “spatial bypass” in the coarser granular SDM layer using a spatial cross-connect (SXC) would be a natural approach toward addressing the above-mentioned problem. The concepts of coarser granular optical switching in, for example, the wavelength-band layer and fiber layer were previously proposed at the end of the 1990s [1113]. Since then, there have been considerable research efforts; however, in our opinion, they have remained rather notional and lacked a practical perspective such as in terms of reliability, scalability, long-distance transmission, and economics, even after SDM technology began to gain much attention [1418].

Toward the forthcoming SDM era, we recently reevaluated the hierarchical optical network architecture and redefined it as a spatial channel network (SCN), which is based on a growable, reliable, low-loss, and cost-effective optical node architecture [1922]. In an SCN, the optical layer is explicitly decoupled to the hierarchical SDM and WDM layers, and an optical node evolves into an SXC and WXC to achieve a hierarchical optical cross-connect (HOXC). Here, adjacent SXCs are connected by an SDM link, which can be a multicore fiber (MCF) or parallel single-mode fibers (SMFs). We define a spatial channel (SCh) as an ultrahigh-capacity optical data stream, which is allowed to occupy the entire available spectrum of a spatial lane (SL). Here, the physical entity of an SL is a core in an SMF or an MCF. An SCh is optically routed end-to-end as a single entity through low-loss SXCs spatially bypassing the overlying WDM layer. Conventional OChs are spectrally multiplexed into an SCh and may be groomed by WXCs if necessary. The SCN yields an extended optical reach for spatially bypassed OChs, which originates from the inherent low-loss switching feature of SXCs [19].

As reliable, scalable, and low-loss SXC architectures, two modular SXC architectures based on submatrix switches (sub MSs) and core selective switches (CSSs) [1921] were proposed. In addition, the feasibility of spatial channel networking, including spatial bypassing, spectral grooming, SL change, and spatial-channel protection was experimentally demonstrated over a spatial channel network testbed comprising the two types of modular SXC prototypes [22]. Although the proposed modular SXC architectures provide a limited SL change capability, a single SDM layer dimensioning simulation showed that this limited capability exerts little influence on the total number of required SLs, while the architectures themselves yield significant total node cost reduction when compared with the traditional nonblocking SXC architecture based on an ultra-high-port-count optical MS [20]. A brief overview of the SCN technology reported thus far can be seen in a recent conference paper [23].

The hierarchical spatial bypass and spectral grooming in an SCN are expected to enable us to accommodate a wide variety of traffic demands from the wavelength level to spatial level in a spectrally and spatially efficient manner. In addition, coarse granular spatial switching is potentially cost-effective in terms of the switching cost per bit. Traditionally, the traffic grooming problem in optical transport networks has been extensively investigated in the context of an electrical grooming node over a WXC [24,25]. This paper presents a technoeconomic analysis on how the degree of spectral grooming by WXCs and spatial bypassing by SXCs in an SCN affects the total required number of SLs (cores) and the total node cost in an SCN by applying a routing and SDM/WDM multilayer resource assignment (RSWA) heuristic to three SDM-compatible optical node architectures: (1) stacked conventional WXCs, (2) high-port-count MS-based HOXC, and (3) CSS-based HOXC.

The rest of this paper is organized as follows. Section 2 describes the details of the three SDM-compatible optical node architectures. Section 3 explains how optical signals are hierarchically routed in an SCN. Section 4 defines the routing and SDM/WDM multilayer resource assignment problem. Section 5 proposes a heuristic that we developed in order to address the RSWA problem. Section 6 presents network, traffic, and node-cost models used in simulations for the technoeconomic analysis. Section 7 presents simulation results and discusses the benefits of spatial bypass and spectral grooming in an SCN. Section 8 describes the conclusions and future work.

2. SDM-COMPATIBLE OPTICAL NODE ARCHITECTURES

This section presents optical node architectures that are used in the technoeconomic analysis simulation of SCNs in this paper. We first describe the stacked conventional WXC architecture as a baseline for SDM-compatible node architectures and then present two HOXC architectures. One is based on a traditional high-port-count MS-based SXC; the other is based on a CSS, which is a new type of optical spatial switch recently proposed and demonstrated in [1922].

A. Stacked Conventional WXCs

Figure 2(a) shows an SDM-compatible node architecture that comprises stacked conventional WXCs for a site whose node degree $D$ is 3. In this architecture, $C$ WXCs with a size of $D \times D$ are sandwiched by $1 \times C$ spatial demultiplexers (SDEMUXs), which have an input MCF, $C$ SMFs, and $C \times 1$ spatial multiplexers (SMUXs), which have $C$ input SMFs and an output MCF. Using an SDEMUX, SLs of an input MCF supporting $C$ cores are spatially demultiplexed. Each SL is connected to a WXC that has the same index as that of the SL. Output ports of the $i$th WXC are connected to the $i$th input port of an SMUX. There are several technologies, which are based on free-space optics, a thin-cladding SMF bundle, and a 3D waveguide, that achieve an SMUX/DEMUX for an MCF.

 

Fig. 2. SDM compatible optical node architectures. (a) Stacked conventional WXCs. (b) High-port-count MS-based HOXC. (c) CSS-based HOXC.

Download Full Size | PPT Slide | PDF

An optical network that comprises SDM-compatible nodes based on the stacked conventional WXC architecture is equivalent to $C$ stacked single-layer WDM networks. If we use parallel SMFs as transmission media in a link instead of an MCF, no SMUX and SDEMUX are needed. We assume the use of a WXC that employs $D$ $1 \times D$ WSSs arranged in the broadcast and select (B&S) configuration and $D$ wavelength multiplexers/demultiplexers (WMUXs/WDEMUXs). This architecture has connectivity constraints regarding dedicated core access and a no-lane-change (LC) capability for OChs. It provides fine granular spectral grooming at the expense of requiring a huge number of conventional WSSs, which is represented by $\textit{CD}$.

B. Hierarchical Optical Cross-Connects

1. High-Port-Count MS-Based HOXC

Figure 2(b) shows an HOXC architecture that employs an SXC based on traditional high-port-count MSs. Similar to the stacked conventional WXC architecture, this architecture requires a $1 \times C$ SDEMUX and a $C \times 1$ SMUX for each node degree if an MCF is employed for an SDM link. Preparing for an MS failure, a secondary high-port-count MS that is connected in parallel to the primary MS with optical splitters and optical ${2} \times {1}$ switches is implemented (not shown). Large $N \times N$ optical switches ($N \gt {\sim}300$) with a sufficiently low insertion loss (${\lt}{\sim} 2\,\,{\rm dB}$) are commercially available.

In addition to input and output ports for line-side connections, the MS equips add/drop ports where SChs are terminated. WMUXs/WDEMUXs and/or WXCs are connected to add/drop ports in order to aggregate/disaggregate and groom OChs.

The advantage of this architecture is that it provides fully nonblocking connection flexibility including the LC capability.

2. Core Selective Switch-Based HOXC

Figure 2(c) shows HOXC architecture based on a CSS [5,7,8]. Here, a $1 \times n$ CSS is the SDM counterpart of a WSS in a WDM network, which has an input MCF and $n$ output MCFs where each MCF has $C$ cores in its cladding. A $1 \times n$ CSS has functionalities for spatially demultiplexing SChs from an input MCF and switching and multiplexing any of them into any of $n$ output MCFs. In order to achieve low-loss SXCs, CSSs should be used in the route and select (R&S) configuration to avoid splitting loss, which implies that $2D$ $1 \times D$ CSSs are needed that support $C$ SLs. The line-side CSS used in this architecture is a $1 \times D$ CSS whose $D - 1$ ports are for through SChs, and the remaining port linked to an SDEMUX (SMUX) is for local drop (add) SChs.

This SXC architecture is growable in terms of the nodal degree and provides dedicated core access and no LC capability. If adjacent SXCs are connected via parallel SMFs instead of an MCF, an SMUX (SDEMUX) is implemented between an ingress (egress) MCF of a CSS and parallel SMFs.

There is currently no commercially available CSS. A CSS can be constructed by combining discrete optical devices of a $1 \times C$ SDEMUX, $C$ $1 \times N$ switches, and $N$ $C \times 1$ SMUXs [22]; however, it would be bulky and most likely expensive. One way to achieve a compact and cost-effective CSS will be to employ free-space optics. A free-space-based $1 \times 6$ CSS supporting five-core MCFs was recently designed and prototyped in [21].

3. HIERARCHICAL ROUTING IN AN SCN

In SCNs based on the HOXCs described in Section 2.B, there must be several approaches toward deciding where to implement a WXC for spectral grooming and where to groom the traffic practically. That is, WXC(s) can be implemented in all nodes, a subset of the nodes, or even none of them. Spectral grooming can be performed at every grooming node (GN) that is equipped with a WXC(s) or selectively.

In the approach reported herein, we first select a set of $K$ nodes among nodes in a given network to be GNs. The remaining nodes, which have no WXC and are referred to as nongrooming nodes (NNs), backhaul the traffic that they generate to nearby GNs. In order to balance the node simplicity and link resource usage efficiency, demands between a source/destination node pair are grouped into two demands, i.e., express and groomed, according to the filling ratio of the frequency bin in which the demand is virtually packed, as explained in Section 5.

Optical signals for express demands are aggregated in the spectral domain using a WMUX at the source node. They are routed end-to-end by SXCs through a dedicated SCh (referred to as an express SCh), without being groomed by a WXC on the route, and finally demultiplexed by a WDEMUX at the destination node (end-to-end aggregation). On the other hand, optical signals for groomed demands are aggregated in the spectral domain using a WXC or WMUX at the source node depending on if the source node is a GN or NN. They are carried in a shared SCh(s) [referred to as a local SCh(s)] while being spectrally groomed by a WXC(s) on the route. They are finally demultiplexed using a WXC or WMUX at the source node. Among the GNs on the route, a limited number of GNs are selected as the accrual grooming nodes according to the employed planning policy. The necessary number of WXCs is deployed at the GNs. One motivation for limiting the number of times grooming on the route is to reduce the total node cost at the expense of a reasonable increase in the required number of SLs. Another motivation is to enjoy the benefits of optical reach extension, which are enabled by spatially bypassing the WXC with a low-loss SXC [19].

As the first step in the technoeconomic analysis of SCNs, we simply assume that all SChs are unprotected, operate in a single modulation format (distance-adaptive modulation format selection is not employed), and are transparently routed end-to-end (no transmission impairment is considered, and no optical-electrical-optical regeneration is utilized).

4. PROBREM STATEMENT

The definition of the RSWA problem for a set of static traffic demands in an SCN is described below.

Given

  • • Physical network topology ${{\boldsymbol G}_{\boldsymbol p}}({{\boldsymbol V},{\boldsymbol E}}),$ where ${\boldsymbol V}$ denotes a set of nodes and ${\boldsymbol E}$ a set of SDM links connecting two nodes in ${\boldsymbol V}$.
  • • Ordered set of SLs in each link, ${\boldsymbol C} = \{{{c_1},{c_2}, \ldots ,{c_{| {\boldsymbol C} |}}} \},$ where the subscript represents an SL index.
  • • Ordered set of frequency-slot units (FSUs), ${\boldsymbol F} = \{{{f_1},{f_2}, \ldots ,{f_{| {\boldsymbol F} |}}} \}$ for each SL where the subscript represents an FSU index.
  • • Traffic demand set ${\boldsymbol \Lambda}$, in which a traffic demand can be defined as ${\boldsymbol T}({s\!,d,q}),$ where $s$ is a source node, $d$ is a destination node, and $q$ is a requested capacity expressed in Tb/s.
  • • Traffic demand set ${{\boldsymbol \Lambda}_{i\!,j}}({\in {\boldsymbol \Lambda}})$ between a source and destination node pair, $({i,j})$.
  • • Required size of a frequency slot (FS) for a traffic demand, which is represented by the required number of contiguous FSUs.
  • • Allowable number of times grooming on the route, $\alpha$.
  • • Threshold for filling ratio of an SL, $h$, expressed in percent.

Find

  • • Spatial topology ${{\boldsymbol G}_{\boldsymbol s}}({{\boldsymbol V},{\boldsymbol S}})$ where ${\boldsymbol S}$ denotes a set of spatial links connecting two nodes in ${\boldsymbol V}$. Each spatial link is provided by an SCh established in ${{\boldsymbol G}_{\boldsymbol p}}$, which is identified by the route, and an SL used in each SDM link on the route in ${{\boldsymbol G}_{\boldsymbol p}}$.
  • • Placement of WXCs for spectral grooming.
  • • Routing and FSU assignment of traffic demands over ${{\boldsymbol G}_{\boldsymbol s}}$ (establishing OChs).

Objectives

  • • Minimize the required number of SLs in the network while subject to the following constraints where the required number of SLs is represented by the highest SL index among the indexes assigned to the used SLs in the network.

Constraints

  • • Along with each SCh, an SL for each link must not overlap that of other SChs.
  • • For an SCN employing a CSS-based SXC, each SL comprising an SCh must have the same SL index.
  • • Along with each OCh, FSUs must be contiguous to each other, must be the same for each link on the route, and must not overlap with those of other OChs.

5. RSWA HEURISTIC

In the first stage of network planning, we first select nodes with a larger node degree as GNs while minimizing the maximum distance between any NNs and the nearest GN using a $K$-center algorithm [26]. Then we address the RSWA problem described in the previous section using the heuristic presented below.

  • Step 1: Calculate the shortest routes on physical network topology
    • (1) For each node pair $({i,j})$ on ${{\boldsymbol G}_{\boldsymbol p}}$, calculate the $k$-shortest routes.
  • Step 2: Classify demands
    • (1) For each node pair $({i,j})$ on ${{\boldsymbol G}_{\boldsymbol p}}$, order traffic demands in traffic demand set ${{\boldsymbol \Lambda}_{i,j}}$ in descending order of their required FS size. Pack each FS for a traffic demand into a finite number of frequency bins with a size of $| {\boldsymbol F} |$ such that the number of bins is minimized using a bin-packing heuristic.
    • (2) Express traffic demand: The aggregate traffic demand set accommodated in the $n$th frequency bin having a filling ratio greater than $h$ is notated as ${\boldsymbol E}_{i,j}^n$, which is classified to be carried end-to-end through a dedicated express SCh without being groomed by a WXC(s).
    • (3) Local traffic demand: The traffic demand set that comprises the remaining traffic demands, which are tentatively accommodated in the frequency bins having a filling ratio of less than $h$, is notated as ${{\boldsymbol L}_{i,j}}$, which is classified to be carried through a shared local SCh(s) while being spectrally groomed by a WXC(s) on the route.
  • Step 3: Establish express SChs
    • (1) Order aggregate traffic demand set ${\boldsymbol E}_{i,j}^n$ in descending order of their shortest path lengths.
    • (2) Select an aggregate traffic demand set from the ordered set. Select a route for the aggregate traffic demand set from the $k$-shortest routes on ${{\boldsymbol G}_{\boldsymbol p}}$, so that the required number of SLs in the network is the lowest based on the first-fit approach.
  • Step 4: Establish a local SCh and place a WXC
    • (1) Order all local traffic demands in the network, which belong to ${{\boldsymbol L}_{i,j}}$ for all node pairs, in descending order of the product of their shortest path lengths and FS sizes.
    • (2) Select a local traffic demand from the ordered set. Along one of the $k$-shortest routes on ${{\boldsymbol G}_{\boldsymbol p}}$ for the demand, create a grooming graph, ${{\boldsymbol G}_{\boldsymbol g}}$, by connecting all combinations of nodes from a node set comprising the source node, destination node, and GNs on the shortest route.
    • (3) On ${{\boldsymbol G}_{\boldsymbol g}}$, create a set of candidate routes between the source and destination nodes whose number of hops equals or is less than $\alpha + 1$. Along a candidate route, if two adjacent nodes are not connected by an SCh, a new SCh is tentatively established between the nodes by selecting an unused SL(s) based on the first-fit approach. Make sure that the OCh carrying the local demand can be accommodated using the available FSU resources in the existing and newly established SChs. If not, add a new SCh on the bottleneck link(s).
    • (4) Select the route on ${{\boldsymbol G}_{\boldsymbol g}}$ for the local traffic demand such that the number of SLs required in the network is the lowest and the required number of FSUs along the SChs on the route is the lowest. Place a new WXC(s) at the end(s) of the newly established SCh(s) if necessary. Here, the required number of FSUs is represented by the highest FSU index among the indexes assigned to the used FSUs in the SChs on the route.

Figure 3 illustrates the routine of the RSWA algorithm in a network with eight nodes. We assume that nodes A, G, H, and D are NNs; nodes B, C, E, and F are GNs; and $\alpha = 1$. If a demand between nodes A and H is an express traffic demand, an express SCh is established end-to-end, as shown in Fig. 3(c). On the other hand, if it is a local traffic demand, a GN is selected from two candidate nodes (node E in this example). The traffic demand is accommodated by two local SChs and spectrally groomed by a WXC at the GN.

 

Fig. 3. Routine of the RSWA algorithm. (a) Physical network topology and shortest routes from A to H. (b) SCh and OCh created on shortest route 4 in the case of an express demand. (c) Grooming graph for shortest route 4 and candidate routes (blue, green, and black when $\alpha = {1}$) in the case of a local demand.

Download Full Size | PPT Slide | PDF

Tables Icon

Table 1. Traffic Model

6. NETWORK, TRAFFIC, AND NODE COST MODELS

A. Network Model

We conduct SCN dimensioning simulations for network models DT14 (14 nodes, 23 links, average and maximum node degrees of 3.29 and 5, respectively) and NSF15 (15 nodes, 23 links, average and maximum node degrees of 3.07 and 4, respectively).

B. Traffic Model

We employ a mixed bit rate traffic model in the SCN dimensioning simulations, in which the bit rate required by each demand is an integer multiple of the base bit rate, and the proportion of demand at a certain bit rate decreases with the bit rate as shown in Table 1. We assume that the total traffic load in the network and the base bit rate (except in the eight- and nine-year later traffic models) increase exponentially every year. In Table 1, each bit rate is accompanied by the necessary number of FSUs to accommodate a Nyquist WDM superchannel having the bit rate. Here, we assume that the width of an FSU is 50 GHz and the comprising subchannel employs a modulation format of 32-Gbaud dual-polarization quadrature phase shift keying. The distribution in the last row of Table 1 shows the ratio of the number of demands for each bit rate to the total number of demands. We assume a uniform demand distribution for all the source/destination node pairs in the network.

For example, in the base year, we assume the network total traffic load is 0.1 Pb/s, in which traffic demands requiring a bit rate of 100 and 200 Gb/s are the majority accounting for 30% and 20% of the total, respectively. Although the ratio is very small at 1%, there exists a traffic demand requiring a bit rate of 1 Tb. If we assume the compound annual growth rate of 40% for the total traffic load as well as the base bit rate, the total traffic load and base bit rate become 1 Pb/s and 1 Tb/s, respectively, seven years after the base year.

C. Node Cost Model

The node cost for each site of an SCN employing stacked conventional WXCs, ${C_{{\rm WXC}}}$, a high-port-count MS-based HOXC, ${C_{{\rm ms}}}$, and CSS-based HOXC, ${C_{{\rm css}}}$, are, respectively, calculated as

$${C_{{\rm wxc}}} = {S_{{\rm wxc}}}{D_n}\cdot{c_{{\rm wss}}},$$
$${C_{{\rm ms}}} = 2\cdot{{ C}_{{\rm ms}}}({{S_{{\rm ms}}}\cdot{D_n} + {a_n}} ) + {b_n}{D_n}\cdot{c_{{\rm wss}}},$$
$${C_{{\rm css}}} = 2{D_n}\cdot{c_{{\rm css}}} + {b_n}{D_n}\cdot{c_{{\rm wss}}}.$$
Here, ${S_{{\rm wxc}}}$ and ${S_{{\rm ms}}}$ are the number of used cores in the network for the stacked conventional WXC and high-port-count MS-based HOXC approaches, respectively. In addition, ${D_n}$, ${a_n}$, and ${b_n}$ are the node degree, the number of add ports, and the required number of WXCs of the $n$th node, ${{C}_{{\rm ms}}}(N)$ is a linear function of $N$ that gives the costs of an $N \times N$ MS, ${c_{{\rm css}}}$ is the cost of a $1 \times 8$ CSS supporting 37 cores [27], and ${c_{{\rm wss}}}$ is the cost of a conventional $1 \times 9$ WSS supporting the C-band. Based on private communications with some venders, we assume that ${{C}_{{\rm ms}}}({32})$ is 3.75 times ${c_{{\rm wss}}}$ and ${c_{{\rm css}}}$ is equal to ${c_{{\rm wss}}}$, which varies with the maturity of manufacturing and the vender pricing strategy. We also assume the number of FSUs per SL is 80. The costs for a WMUX, a WDEMUX, an SMUX, and an SDEMUX are assumed to be negligibly small compared with those for a WSS, a CSS, and an MS, respectively, and are not included in the node costs.

7. SIMULATION RESULTS

A. Simulation Conditions

We calculate the required number of SLs and total node cost when the traffic model given in Table 1 is applied to network models DT14 and NSF15 for three SDM-compatible node architectures: stacked conventional WXCs, high-port-count MS-based HOXC, and CSS-based HOXC. To construct an CSS-based HOXC, we assume the use of a CSS that supports 37 hexagonal closely packed cores [27] in its cladding. Using the $K$-center algorithm, we select six GNs (${\sim}{40}\%$ of all nodes), as indicated in blue in Fig. 4. The allowable number of times grooming on the route, $\alpha$, is set to 3. This value is determined based on the preliminary simulation results that the reduction in the required number of SLs tends to saturate when α is greater than 3, which is consistent with the results for electrical subrate traffic grooming in an optical-bypass-enabled WDM network, as discussed in [25]. Simulation results are averaged over 100 randomly generated traffic matrices for source/destination node pairs from the traffic model for each year.

 

Fig. 4. Network models used in the simulation. Nodes indicated in blue are GNs, and the remaining nodes are NNs. (a) DT14, (b) NSF15.

Download Full Size | PPT Slide | PDF

An optical network comprising SDM-compatible nodes based on the stacked conventional WXCs is equivalent to multiple conventional WDM networks, which are independent of each other. We calculated the required number of SLs and total node cost for the case of the baseline stacked conventional WXCs by using a simple routing and spectral assignment (RSA) algorithm based on the $k$-shortest path and the first-fit approaches. First, we order all traffic demands in the network in descending order of the product of their shortest path lengths and FS sizes. Then, we accommodate each traffic demand into a finite number of WDM networks on ${{\boldsymbol G}_{\boldsymbol p}}$ so that the number of WDM networks is minimized, which is equivalent to attempting to minimize the number of used SLs.

On the other hand, for the high-port-count MS-based HOXC and CSS-based HOXC, we apply the RSWA algorithm with the following four different planning policies, which are characterized by the threshold for the filling ratio of frequency bin $h$ to be classified as an express demand set.

  • Express-only ($h = 0\%$): This is an extreme planning policy where all traffic demands are carried by express SChs end-to-end without being spectrally groomed at any GN. An SCN operating under this policy is equivalent to a single-layer SDM network comprising SXCs only.
  • Express/local-hybrid (spatial-bypass oriented) ($h = 40\%$): This is a spatial-bypass-oriented planning policy where the threshold for filling ratio $h$ is set to a low level. An aggregate demand set having a filling ratio of a frequency bin greater than 40% is carried by an express SCh end-to-end without being spectrally groomed by a WXC(s).
  • Express/local-hybrid (spectral-grooming oriented) ($h =$ $ 80\%$): This is a spectral-grooming oriented planning policy where the threshold for filling ratio $h$ is set to a high level. Traffic demands are carried by a local SCh(s) while being groomed by a WXC(s), unless a filling ratio of an aggregate demand set is greater than 80%.
  • Local-only ($h = 100\%$): This is another extreme planning policy where the RSWA attempts to groom all demands at as many GNs as possible on the route, while spatial bypass is only performed at NNs on the route.

B. Required SLs

Figures 5(a) and 5(b) show simulation results of the required number of SLs for network model DT14 when employing the high-port-count MS-based HOXC and CSS-based HOXC, respectively. We compare results for the four planning policies—express-only, express/local-hybrid (spatial-bypass-oriented), express/local-hybrid (spectral-grooming-oriented), and local-only—to that for the baseline stacked conventional WXCs as the network traffic load increases from 0.1 to 2 Pb/s.

 

Fig. 5. Required number of SLs for the DT14 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral-grooming-oriented), and express/local-hybrid (spatial-bypass-oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.

Download Full Size | PPT Slide | PDF

The required number of SLs monotonically increases with the network traffic load for all cases. The baseline stacked conventional WXCs always require the fewest SLs, as plotted with black circles in Fig. 5. This is because it takes advantage of the highest chance of spectrally aggregating the low-rate traffic demands onto every SL using a WXC at every node on the route leading to efficient network resource utilization. The HOXC with the express-only policy cannot utilize the spectral grooming; thus, the SL capacity is underutilized in most cases. So, this policy always requires the most SLs, as plotted with white circles in Fig. 5. The numbers of SLs required by the HOXCs with the local only (plotted with squares), the express/local-hybrid (spectral-grooming-oriented) (plotted with triangles), and the express/local-hybrid (spatial-bypass-oriented) (plotted with diamonds) policies take an intermediate value between those in the above two cases and increases in this order as expected.

The difference in the required numbers of SLs between the baseline-stacked conventional WXCs and the HOXCs becomes smaller as the network traffic load increases. In particular, the HOXCs with the local-only and the express/local-hybrid (spectral-grooming-oriented) policies require almost the same number of SLs as that for the stacked conventional WXCs when the network traffic load is equal to or greater than 0.5 Pb/s. This is because, in this situation, most express SChs are well-packed, and the remaining demands are efficiently groomed by WXCs. For example, the CSS-based HOXC with express/local-hybrid (spectral-grooming-oriented) policy requires 15% and 6% more SLs than the stacked conventional WXCs when the network traffic loads are 0.5 and 1 Pb/s, respectively. If we employ the express/local-hybrid (spatial-bypass-oriented) policy, the ratio of additional required SLs becomes larger at 47% and 21%, respectively. On the other hand, the number of SLs required by the HOXC with the express-only policy never approaches that of the stacked conventional WXCs even at a high network traffic load. For example, the CSS-based HOXC with the express-only policy requires 83% more SLs than the stacked conventional WXCs when the network traffic load is 0.5 Pb/s, and necessitates 24% more SLs even when the network traffic load reaches 2 Pb/s. This is due to the existence of traffic demands that are carried by an underutilized SCh without being spectrally groomed. This means that a single-layer SCN comprising only SXCs never achieves the same level of network resource utilization as a single-layer WDM network even at a high network traffic load. These results clearly show the importance of an SCN with hierarchical SDM and WDM layers comprising SXCs and WXCs.

In the same way that a conventional WSS-based WXC does not provide the wavelength conversion functionality, a CSS-based SXC does not provide the SL change (core interchange) functionality. Fortunately, although the high-port-count MS-based HOXC, which has the SL change capability, requires fewer SLs than that required by the CSS-based HOXC, the difference is small and becomes negligible when the load is greater than 1 Pb/s, as can be seen when comparing Figs. 5(a) and 5(b). These results are consistent with the fact that many studies on the impact that wavelength contention has on the performance of optical-bypass-enabled networks concluded that there is only a small impact if good algorithms are used [24].

A similar simulation is conducted for network model NSF15, which has a slightly lower average and maximum node degree than those for DT14. The simulation results are shown in Figs. 6(a) and 6(b). They indicate that the same argument described above regarding the comparison of the required number of SLs between the HOXCs and the stacked conventional WXCs holds. The required number of SLs increases in the order of stacked conventional WXCs, HOXCs with the local only, express/local-hybrid (spectral-grooming-oriented), express/local-hybrid (spatial-bypass-oriented), and express-only policies, and the difference between the stacked conventional WXCs and the HOXCs becomes smaller as the network traffic load increases.

 

Fig. 6. Required number of SLs for the NSF15 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral-grooming-oriented), and express/local-hybrid (spatial-bypass-oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.

Download Full Size | PPT Slide | PDF

C. Total Node Cost

Figures 7(a) and 7(b) show simulation results of the total node cost normalized by the cost of a conventional $1 \times 9$ WSS for network model DT14 when employing a high-port-count MS-based HOXC and CSS-based HOXC, respectively. We compare the results of the four planning policies—express-only, express/local-hybrid (spatial-bypass-oriented), express/local-hybrid (spectral-grooming-oriented), and local-only—with those for the baseline-stacked conventional WXCs, as the network traffic load is increased from 0.1 to 2 Pb/s.

 

Fig. 7. Total node cost normalized by the cost of a conventional $1 \times 9$ WSS for the DT14 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral-grooming-oriented), and express/local-hybrid (spatial-bypass-oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.

Download Full Size | PPT Slide | PDF

The total node cost of the baseline-stacked conventional WXCs (plotted with black circles) requires the lowest cost in most cases at a low traffic load, but it increases with the network traffic load more rapidly than those for the high-port-count MS-based HOXC and CSS-based HOXC; at some point, it overtakes the MS-based and CSS-based HOXC architectures. This clearly shows that the HOXC approach could be a cost-effective solution for a high network traffic load. Since the high-port-count MS-based HOXC requires a secondary high-port-count MS that is connected in parallel to the primary MS in preparation for an MS failure, it requires a high total node cost even at a low traffic load. This is why the network traffic load at which the total node cost of the stacked conventional WXCs overtakes the CSS-based HOXC is much lower (${\sim}0.{2}\;{\rm Pb/s}$) than that for the high-port-count MS-based HOXC (${\sim}{1}\;{\rm Pb/s}$).

The growth rate in the total node cost for the HOXCs decreases in the order of the local-only (plotted with squares), the express/local-hybrid (spectral-grooming-oriented) (plotted with triangles), the express/local-hybrid (spatial-bypass-oriented) (plotted with diamonds), and the express-only (plotted with white circles) policies. This is because the required number of WXCs decreases in this order as the network traffic load increases. The same tendency is observed from the simulation results on the total node cost normalized by the cost of a conventional $1 \times 9$ WSS for network model NSF15, as shown in Fig. 8.

 

Fig. 8. Total node cost normalized by the cost of a conventional $1 \times 9$ WSS for the NSF15 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral grooming oriented), and express/local-hybrid (spatial bypass oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.

Download Full Size | PPT Slide | PDF

D. Required SLs versus Total Node Cost Trade-Off

If we compare Figs. 5 and 7 (DT14) or Figs. 6 and 8 (NSF15), we first notice that the node-cost curves for the CSS-based HOXC employing the spectral grooming and the stacked conventional WXCs intersect when the network traffic load is 0.2 Pb/s. However, at this traffic load, the former requires twice as many SLs as the latter, so the network traffic load at which the CSS-based HOXC presents the technoeconomic validity is higher. For example, when the network traffic load is 0.5 Pb/s, the CSS-based HOXC employing the express/local-hybrid (spectral-grooming-oriented) requires 15% more SLs (10.8 SLs on average) than the stacked conventional WXCs but yields a 33% node-cost savings. This may be a good compromise. It should be noted that, if the stacked conventional WXCs employ the R&S configuration instead of the B&S configuration used in our simulation in order to avoid the excessive splitting loss of optical splitters for a high-port-count WSS, the cost savings increases to 66%.

We also notice a trade-off between the required number of SLs and the total node cost in the selection of the planning policy. For example, the CSS-based HOXC architecture employing the express-only policy provides an extreme solution. The total node cost for this architecture is always the lowest, and it provides the node cost savings of 87% for network model DT14 when the network traffic load is 1 Pb/s. However, it requires 42% more SLs than the baseline stacked conventional WXCs. Over-the-top-players may prefer this express-only policy, in other words, employing the simple IP-over-SDM network architecture, since their inter-datacenter networks may be rather homogeneous and connected by cables consisting of hundreds or thousands of parallel SMFs, and aggregate traffic amounts of their router braids may reach the full capacity that a single-mode core can support. The CSS-based HOXC architecture employing the local-only policy provides another extreme solution. It requires only 5% more SLs than the baseline-stacked conventional WXCs at the traffic load of 1 Pb/s, while the node cost saving remains at 37%, although this is still a considerable amount. The CSS-based HOXC architecture employing the express/local-hybrid (spectral-grooming-oriented) policy, whose threshold for filling ratio $h$ is 80%, provides a more moderate and balanced solution. It requires only 6% more SLs than the baseline-stacked conventional WXCs at the traffic load of 1 Pb/s, while the node cost saving reaches 63%. Traditional telecom carriers may prefer the express/local-hybrid policy, in other words, employing the IP-over-WDM-over-SDM network architecture, to support a wide variety of traffic demands from the wavelength level to spatial level in a spectrally and spatially efficient manner.

On the other hand, the high-port-count MS-based HOXC architecture employing the express/local-hybrid (spectral-grooming-oriented) requires 2% fewer SLs but is 2.7 times costlier at the traffic load of 1 Pb/s than the CSS-based HOXC architecture employing the same planning policy.

If the assumptions described in Section 6 are reasonable, the results in our simulations lead us to conclude the following.

  • • Compared with the baseline-stacked conventional WXC architecture, the CSS-based HOXC architecture employing the express/local-hybrid (spectral-grooming-oriented) planning policy provides a more balanced solution between the required number of SLs and the network-total optical node cost when the required number of SLs between optical nodes in an SCN is equal to or greater than roughly 10.
  • • Although the CSS-based HOXC architecture has a non-SL changeable connectivity constraint, it provides practically the same performance as the high-port-count MS-based HOXC architecture in terms of the spatial resource utilization efficiency and outperforms it on the network-total node cost.

8. CONCLUSIONS

We conducted a technoeconomic analysis of SCNs comprising different HOXC architectures employing different planning policies on how the degree of spectral grooming by WXCs and spatial bypassing by SXCs in an SCN affects the total required number of SLs (cores) and the total node cost in the SCN. Toward this end, we developed an RSWA heuristic in which spatial bypassing and spectral grooming are performed in such a way that the required number of SLs is minimized. Using the RSWA with four planning policies—express-only, express/local-hybrid (spatial-bypass-oriented), express/local-hybrid (spectral-grooming-oriented), and local-only—we compared the performance levels of a high-port-count MS-based HOXC and CSS-based HOXC to those of the stacked conventional WXCs as the network traffic load increases. We clearly showed that hierarchical spatial bypassing and spectral grooming are beneficial in terms of the required number of SLs and network-total node cost when the required number of SLs between optical nodes in an SCN is equal to or greater than roughly 10.

In this paper, as the first step in the technoeconomic analysis of SCNs, we focused on a simple SCN model, i.e., all SChs are unprotected, operate in a single modulation format (distance-adaptive modulation format selection is not employed), and are transparently routed end-to-end (no optical-electrical-optical regeneration is utilized). Transmission performance impairments due to accumulated amplified-stimulated-emission noise of optical amplifiers, nonlinear interference noise generated in fiber cores, and intercore crosstalk noise in MCFs, which are not considered in the RSWA heuristic, may restrict routing and grooming flexibility. Although the description is omitted due to space limitations, ordering traffic demands in descending order, as in our RSWA heuristic, exhibited clearly better performance than the ascending ordering. However, there may be methods, simulated annealing techniques for example, that may further improve performance. The next step in our research is to investigate the impact of these features on the technoeconomic performance of SCNs.

Funding

National Institute of Information and Communications Technology (19302, 20401); Japan Society for the Promotion of Science (JP18H01443).

REFERENCES

1. T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag.50(20), s31–s42 (2012). [CrossRef]  

2. P. J. Winzer and D. T. Neilson, “From scaling disparities to integrated parallelism: a decathlon for a decade,” J. Lightwave Technol.35, 1099–1115 (2017). [CrossRef]  

3. A. A. M. Saleh, “Transparent optical networking in backbone networks,” in Optical Fiber Communication Conference (OFC), Baltimore, Maryland, March 7, 2000, paper ThD7.

4. E. Modiano and A. Narula-Tam, “Mechanisms for providing optical bypass in WDM-based networks,” SPIE Opt. Netw. Mag.1, 9–16 (2000).

5. J. Berthold, A. A. M. Saleh, L. Blair, and J. M. Simmons, “Optical networking: past, present, and future,” J. Lightwave Technol.26, 1104–1118 (2008). [CrossRef]  

6. M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag.47(11), 66–73 (2009). [CrossRef]  

7. M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag.48(8), 138–145 (2010). [CrossRef]  

8. M. Jinno, “Elastic optical networking: roles and benefits in beyond 100-Gb/s era,” J. Lightwave Technol.35, 1116–1124 (2017). [CrossRef]  

9. D. M. Marom, P. D. Colbourne, A. D’Errico, N. K. Fontaine, Y. Ikuma, R. Proietti, L. Zong, J. M. Rivas-Moscoso, and I. Tomkos, “Survey of photonic switching architectures and technologies in support of spatially and spectrally flexible optical networking [Invited],” J. Opt. Commun. Netw.9, 1–26 (2017). [CrossRef]  

10. Y. Iwai, H. Hasegawa, and K. Sato, “A large-scale photonic node architecture that utilizes interconnected OXC subsystems,” Opt. Express21, 478–487 (2013). [CrossRef]  

11. K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.

12. M. Jinno, J. Kani, and K. Oguchi, “Ultra-wide-band WDM networks and supporting technologies,” in Core Networks and Network Management (NOC’ 99) (1999), pp. 90–97.

13. A. A. M. Saleh and J. M. Simmons, “Architectural principles of optical regional and metropolitan access networks,” J. Lightwave Technol.17, 2431–2448 (1999). [CrossRef]  

14. X. Cao, V. Anand, and C. Qiao, “Framework for waveband switching in multigranular optical networks: part I-multigranular cross-connect architectures,” J. Opt. Netw.5, 1043–1055 (2006). [CrossRef]  

15. K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

16. M. Cvijetic, I. B. Djordjevic, and N. Cvijetic, “Dynamic multidimensional optical networking based on spatial and spectral processing,” Opt. Express20, 9144–9150 (2012). [CrossRef]  

17. N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express21, 8865–8872 (2013). [CrossRef]  

18. G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

19. M. Jinno, “Spatial channel network (SCN): opportunities and challenges of introducing spatial bypass toward massive SDM era,” J. Opt. Commun. Netw.11, 1–14 (2019). [CrossRef]  

20. M. Jinno, “Spatial channel cross-connect architectures for spatial channel networks,” IEEE J. Sel. Top. Quantum Electron.26, 3600116 (2020). [CrossRef]  

21. M. Jinno, T. Kodama, and T. Ishikawa, “Principle, design, and prototyping of core selective switch using free-space optics for spatial channel network,” J. Lightwave Technol.38, 4895–4905 (2020). [CrossRef]  

22. M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol.38, 2577–2586 (2020). [CrossRef]  

23. M. Jinno and T. Kodama, “Spatial channel network (SCN): introducing spatial bypass toward the SDM era,” in Optical Fiber Communications Conference (OFC) (2020), paper M2G. 1.

24. R. Dutta, A. E. Kamal, and G. N. Rouskas, Traffic Grooming for Optical Networks: Foundations, Techniques and Frontiers (Springer, 2008).

25. J. M. Simmons, Optical Network Design and Planning, Optical Networks (Springer, 2014).

26. B. Chen, G. N. Rouskas, and R. Dutta, “Clustering methods for hierarchical traffic grooming in large scale mesh WDM networks,” J. Opt. Commun. Netw.2, 502–514 (2010). [CrossRef]  

27. Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Masahiko Jinno (M'90–SM'12–F'20) received B.E. and M.E. degrees in electronics engineering from Kanazawa University, Ishikawa, Japan in 1984 and 1986, respectively, and a Ph.D. degree in engineering from Osaka University, Osaka, Japan, in 1995 for his work on ultrafast optical signal processing based on nonlinear effects in optical fibers. He currently serves as a professor of the Faculty of Engineering and Design at Kagawa University, Takamatsu, Japan. His current research interests include architecture, design, management, and control of optical networks, optical transmission systems, optical cross-connects, optical switches, and rate- and format-flexible optical transponders. Prior to joining Kagawa University in October 2012, he was a senior research engineer and supervisor at Nippon Telegraph and Telephone (NTT) Network Innovation Laboratories, NTT Corporation, conducting pioneering research on spectrum- and energy-efficient elastic optical networks (EONs). From 1993 to 1994, he was a guest scientist at the National Institute of Standards and Technology (NIST), Boulder, Colorado, USA. He authored or co-authored over 200 peer-reviewed journal and conference papers in the fields of ultrafast optical signal processing for high-capacity optical time division multiplexed transmission systems, optical sampling and optical time-domain reflectometry, ultra-wideband WDM transmission systems in the L-band and S-band, ROADM systems, GMPLS and application-aware optical networking, EONs, and spatial channel networks (SCNs). Prof. Jinno is a fellow of the Institute of Electronics, Information and Communication Engineers (IEICE) and a member of The Optical Society (OSA). He received the Young Engineer's Award in 1993, the Best Tutorial Paper Award in 2011, the Best Paper Award in 2012, the Achievement Award in 2017, and the Milestone Certificate in 2017 from the IEICE; the Best Paper Awards from the 1997, 1998, 2007, and 2019 Optoelectronics and Communications Conferences (OECC); the Best Paper Award from the 2010 ITU-T Kaleidoscope Academic Conference; and the Outstanding Paper Award in 2013 from the IEEE Communications Society Asia-Pacific Board.

Yu Asano received B.E. and M.E. degrees in electronics and information engineering from Kagawa University, Takamatsu, Japan, in 2018 and 2020, respectively.

Yoshiki Azuma received a B.E. degree in electronics and information engineering from Kagawa University, Takamatsu, Japan, in 2019. He is currently a graduate student at Kagawa University.

Takahiro Kodama (S'11-M'12) received a B.E. degree from Ritsumeikan University, Shiga, Japan, in 2008 and M.E. and Dr. Eng. degrees from Osaka University, Osaka, Japan, in 2010 and 2012, respectively. In 2012, he was selected as a research fellow of the Japan Society for the Promotion of Science (JSPS). In 2014, he joined Mitsubishi Electric Corporation, Kanagawa, Japan. In 2016, he was a research assistant professor with the Graduate Faculty of Interdisciplinary Research, University of Yamanashi, Yamanashi, Japan, and was selected as an excellent young researcher of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan. Since 2019, he has been a lecturer in the Faculty of Engineering and Design, Kagawa University. His research interests are in the areas of optical access, metro and core networks, optical packet switching networks, digital signal processing, and optical signal processing. He has published over 50 papers in refereed journals and international conference papers. Dr. Kodama is a member of the Institute of Electronics, Information and Communication Engineers (IEICE). He was the recipient of the 2011 IEEE Kansai Section Student Paper Award.

Riku Nakai received a B.E. degree in electronics and information engineering from Kagawa University, Takamatsu, Japan, in 2020. He is currently a graduate student at Kagawa University.

References

  • View by:
  • |
  • |
  • |

  1. T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
    [Crossref]
  2. P. J. Winzer and D. T. Neilson, “From scaling disparities to integrated parallelism: a decathlon for a decade,” J. Lightwave Technol. 35, 1099–1115 (2017).
    [Crossref]
  3. A. A. M. Saleh, “Transparent optical networking in backbone networks,” in Optical Fiber Communication Conference (OFC), Baltimore, Maryland, March7, 2000, paper ThD7.
  4. E. Modiano and A. Narula-Tam, “Mechanisms for providing optical bypass in WDM-based networks,” SPIE Opt. Netw. Mag. 1, 9–16 (2000).
  5. J. Berthold, A. A. M. Saleh, L. Blair, and J. M. Simmons, “Optical networking: past, present, and future,” J. Lightwave Technol. 26, 1104–1118 (2008).
    [Crossref]
  6. M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
    [Crossref]
  7. M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
    [Crossref]
  8. M. Jinno, “Elastic optical networking: roles and benefits in beyond 100-Gb/s era,” J. Lightwave Technol. 35, 1116–1124 (2017).
    [Crossref]
  9. D. M. Marom, P. D. Colbourne, A. D’Errico, N. K. Fontaine, Y. Ikuma, R. Proietti, L. Zong, J. M. Rivas-Moscoso, and I. Tomkos, “Survey of photonic switching architectures and technologies in support of spatially and spectrally flexible optical networking [Invited],” J. Opt. Commun. Netw. 9, 1–26 (2017).
    [Crossref]
  10. Y. Iwai, H. Hasegawa, and K. Sato, “A large-scale photonic node architecture that utilizes interconnected OXC subsystems,” Opt. Express 21, 478–487 (2013).
    [Crossref]
  11. K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.
  12. M. Jinno, J. Kani, and K. Oguchi, “Ultra-wide-band WDM networks and supporting technologies,” in Core Networks and Network Management (NOC’ 99) (1999), pp. 90–97.
  13. A. A. M. Saleh and J. M. Simmons, “Architectural principles of optical regional and metropolitan access networks,” J. Lightwave Technol. 17, 2431–2448 (1999).
    [Crossref]
  14. X. Cao, V. Anand, and C. Qiao, “Framework for waveband switching in multigranular optical networks: part I-multigranular cross-connect architectures,” J. Opt. Netw. 5, 1043–1055 (2006).
    [Crossref]
  15. K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.
  16. M. Cvijetic, I. B. Djordjevic, and N. Cvijetic, “Dynamic multidimensional optical networking based on spatial and spectral processing,” Opt. Express 20, 9144–9150 (2012).
    [Crossref]
  17. N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
    [Crossref]
  18. G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.
  19. M. Jinno, “Spatial channel network (SCN): opportunities and challenges of introducing spatial bypass toward massive SDM era,” J. Opt. Commun. Netw. 11, 1–14 (2019).
    [Crossref]
  20. M. Jinno, “Spatial channel cross-connect architectures for spatial channel networks,” IEEE J. Sel. Top. Quantum Electron. 26, 3600116 (2020).
    [Crossref]
  21. M. Jinno, T. Kodama, and T. Ishikawa, “Principle, design, and prototyping of core selective switch using free-space optics for spatial channel network,” J. Lightwave Technol. 38, 4895–4905 (2020).
    [Crossref]
  22. M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol. 38, 2577–2586 (2020).
    [Crossref]
  23. M. Jinno and T. Kodama, “Spatial channel network (SCN): introducing spatial bypass toward the SDM era,” in Optical Fiber Communications Conference (OFC) (2020), paper M2G. 1.
  24. R. Dutta, A. E. Kamal, and G. N. Rouskas, Traffic Grooming for Optical Networks: Foundations, Techniques and Frontiers (Springer, 2008).
  25. J. M. Simmons, Optical Network Design and Planning, Optical Networks (Springer, 2014).
  26. B. Chen, G. N. Rouskas, and R. Dutta, “Clustering methods for hierarchical traffic grooming in large scale mesh WDM networks,” J. Opt. Commun. Netw. 2, 502–514 (2010).
    [Crossref]
  27. Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

2020 (3)

M. Jinno, “Spatial channel cross-connect architectures for spatial channel networks,” IEEE J. Sel. Top. Quantum Electron. 26, 3600116 (2020).
[Crossref]

M. Jinno, T. Kodama, and T. Ishikawa, “Principle, design, and prototyping of core selective switch using free-space optics for spatial channel network,” J. Lightwave Technol. 38, 4895–4905 (2020).
[Crossref]

M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol. 38, 2577–2586 (2020).
[Crossref]

2019 (1)

2017 (3)

2013 (2)

2012 (2)

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

M. Cvijetic, I. B. Djordjevic, and N. Cvijetic, “Dynamic multidimensional optical networking based on spatial and spectral processing,” Opt. Express 20, 9144–9150 (2012).
[Crossref]

2010 (2)

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

B. Chen, G. N. Rouskas, and R. Dutta, “Clustering methods for hierarchical traffic grooming in large scale mesh WDM networks,” J. Opt. Commun. Netw. 2, 502–514 (2010).
[Crossref]

2009 (1)

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

2008 (1)

2006 (1)

2000 (1)

E. Modiano and A. Narula-Tam, “Mechanisms for providing optical bypass in WDM-based networks,” SPIE Opt. Netw. Mag. 1, 9–16 (2000).

1999 (1)

Aikawa, K.

Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Amaya, N.

Anand, V.

Awaji, Y.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Berthold, J.

Blair, L.

Cao, X.

Chen, B.

Colbourne, P. D.

Cvijetic, M.

Cvijetic, N.

D’Errico, A.

Djordjevic, I. B.

Dutta, R.

B. Chen, G. N. Rouskas, and R. Dutta, “Clustering methods for hierarchical traffic grooming in large scale mesh WDM networks,” J. Opt. Commun. Netw. 2, 502–514 (2010).
[Crossref]

R. Dutta, A. E. Kamal, and G. N. Rouskas, Traffic Grooming for Optical Networks: Foundations, Techniques and Frontiers (Springer, 2008).

Fontaine, N. K.

Harada, K.

K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.

Hasegawa, H.

Y. Iwai, H. Hasegawa, and K. Sato, “A large-scale photonic node architecture that utilizes interconnected OXC subsystems,” Opt. Express 21, 478–487 (2013).
[Crossref]

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

Henning, I.

Hirano, A.

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

Ikuma, Y.

Irfan, M.

Ishii, K.

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

Ishikawa, T.

M. Jinno, T. Kodama, and T. Ishikawa, “Principle, design, and prototyping of core selective switch using free-space optics for spatial channel network,” J. Lightwave Technol. 38, 4895–4905 (2020).
[Crossref]

M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol. 38, 2577–2586 (2020).
[Crossref]

Iwai, Y.

Jinno, M.

M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol. 38, 2577–2586 (2020).
[Crossref]

M. Jinno, “Spatial channel cross-connect architectures for spatial channel networks,” IEEE J. Sel. Top. Quantum Electron. 26, 3600116 (2020).
[Crossref]

M. Jinno, T. Kodama, and T. Ishikawa, “Principle, design, and prototyping of core selective switch using free-space optics for spatial channel network,” J. Lightwave Technol. 38, 4895–4905 (2020).
[Crossref]

M. Jinno, “Spatial channel network (SCN): opportunities and challenges of introducing spatial bypass toward massive SDM era,” J. Opt. Commun. Netw. 11, 1–14 (2019).
[Crossref]

M. Jinno, “Elastic optical networking: roles and benefits in beyond 100-Gb/s era,” J. Lightwave Technol. 35, 1116–1124 (2017).
[Crossref]

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

M. Jinno, J. Kani, and K. Oguchi, “Ultra-wide-band WDM networks and supporting technologies,” in Core Networks and Network Management (NOC’ 99) (1999), pp. 90–97.

M. Jinno and T. Kodama, “Spatial channel network (SCN): introducing spatial bypass toward the SDM era,” in Optical Fiber Communications Conference (OFC) (2020), paper M2G. 1.

Kamal, A. E.

R. Dutta, A. E. Kamal, and G. N. Rouskas, Traffic Grooming for Optical Networks: Foundations, Techniques and Frontiers (Springer, 2008).

Kamei, S.

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

Kani, J.

M. Jinno, J. Kani, and K. Oguchi, “Ultra-wide-band WDM networks and supporting technologies,” in Core Networks and Network Management (NOC’ 99) (1999), pp. 90–97.

Klaus, W.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Kodama, T.

M. Jinno, T. Kodama, and T. Ishikawa, “Principle, design, and prototyping of core selective switch using free-space optics for spatial channel network,” J. Lightwave Technol. 38, 4895–4905 (2020).
[Crossref]

M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol. 38, 2577–2586 (2020).
[Crossref]

M. Jinno and T. Kodama, “Spatial channel network (SCN): introducing spatial bypass toward the SDM era,” in Optical Fiber Communications Conference (OFC) (2020), paper M2G. 1.

Kozicki, B.

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

Kudou, T.

K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.

Luis, R. S.

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Marom, D. M.

Matsuoka, S.

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

Miyamoto, Y.

Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Miyazawa, T.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Modiano, E.

E. Modiano and A. Narula-Tam, “Mechanisms for providing optical bypass in WDM-based networks,” SPIE Opt. Netw. Mag. 1, 9–16 (2000).

Morioka, T.

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Narula-Tam, A.

E. Modiano and A. Narula-Tam, “Mechanisms for providing optical bypass in WDM-based networks,” SPIE Opt. Netw. Mag. 1, 9–16 (2000).

Neilson, D. T.

Nejabati, R.

Oguchi, K.

M. Jinno, J. Kani, and K. Oguchi, “Ultra-wide-band WDM networks and supporting technologies,” in Core Networks and Network Management (NOC’ 99) (1999), pp. 90–97.

Okuno, M.

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

Ozeki, T.

K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.

Poletti, F.

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

Proietti, R.

Puttnam, B. J.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Qiao, C.

Richardson, D.

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

Rivas-Moscoso, J. M.

Rouskas, G. N.

B. Chen, G. N. Rouskas, and R. Dutta, “Clustering methods for hierarchical traffic grooming in large scale mesh WDM networks,” J. Opt. Commun. Netw. 2, 502–514 (2010).
[Crossref]

R. Dutta, A. E. Kamal, and G. N. Rouskas, Traffic Grooming for Optical Networks: Foundations, Techniques and Frontiers (Springer, 2008).

Ryf, R.

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

Sakaguchi, J.

Saleh, A. A. M.

Saridis, G. M.

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Sasaki, Y.

Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Sato, K.

Y. Iwai, H. Hasegawa, and K. Sato, “A large-scale photonic node architecture that utilizes interconnected OXC subsystems,” Opt. Express 21, 478–487 (2013).
[Crossref]

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

Shimizu, K.

K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.

Simeonidou, D.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Simmons, J. M.

Sone, Y.

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

Takahashi, H.

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

Takara, H.

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

Takenaga, K.

Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Tanaka, T.

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

Tomkos, I.

Tsukishima, Y.

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

Wada, N.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Watanabe, A.

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

Winzer, P.

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

Winzer, P. J.

Zervas, G.

N. Amaya, M. Irfan, G. Zervas, R. Nejabati, D. Simeonidou, J. Sakaguchi, W. Klaus, B. J. Puttnam, T. Miyazawa, Y. Awaji, N. Wada, and I. Henning, “Fully-elastic multi-granular network with space/frequency/time switching using multi-core fibres and programmable optical nodes,” Opt. Express 21, 8865–8872 (2013).
[Crossref]

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

Zong, L.

IEEE Commun. Mag. (3)

T. Morioka, Y. Awaji, R. Ryf, P. Winzer, D. Richardson, and F. Poletti, “Enhancing optical communications with brand new fibers,” IEEE Commun. Mag. 50(20), s31–s42 (2012).
[Crossref]

M. Jinno, H. Takara, B. Kozicki, Y. Tsukishima, Y. Sone, and S. Matsuoka, “Spectrum-efficient and scalable elastic optical path network: architecture, benefits, and enabling technologies,” IEEE Commun. Mag. 47(11), 66–73 (2009).
[Crossref]

M. Jinno, B. Kozicki, H. Takara, A. Watanabe, Y. Sone, T. Tanaka, and A. Hirano, “Distance-adaptive spectrum resource allocation in spectrum-sliced elastic optical path network,” IEEE Commun. Mag. 48(8), 138–145 (2010).
[Crossref]

IEEE J. Sel. Top. Quantum Electron. (1)

M. Jinno, “Spatial channel cross-connect architectures for spatial channel networks,” IEEE J. Sel. Top. Quantum Electron. 26, 3600116 (2020).
[Crossref]

J. Lightw. Technol. (1)

M. Jinno, T. Kodama, and T. Ishikawa, “Feasibility demonstration of spatial channel networking using SDM/WDM hierarchical approach for peta-b/s optical transport,” J. Lightw. Technol. 38, 2577–2586 (2020).
[Crossref]

J. Lightwave Technol. (5)

J. Opt. Commun. Netw. (3)

J. Opt. Netw. (1)

Opt. Express (3)

SPIE Opt. Netw. Mag. (1)

E. Modiano and A. Narula-Tam, “Mechanisms for providing optical bypass in WDM-based networks,” SPIE Opt. Netw. Mag. 1, 9–16 (2000).

Other (9)

A. A. M. Saleh, “Transparent optical networking in backbone networks,” in Optical Fiber Communication Conference (OFC), Baltimore, Maryland, March7, 2000, paper ThD7.

K. Harada, K. Shimizu, T. Kudou, and T. Ozeki, “Hierarchical optical path cross-connect systems for large scale WDM networks,” in Optical Fiber Communication Conference and the International Conference on Integrated Optics and Optical Fiber Communication (1999), Vol. 2, pp. 356–358.

M. Jinno, J. Kani, and K. Oguchi, “Ultra-wide-band WDM networks and supporting technologies,” in Core Networks and Network Management (NOC’ 99) (1999), pp. 90–97.

G. M. Saridis, B. J. Puttnam, R. S. Luis, W. Klaus, T. Miyazawa, Y. Awaji, G. Zervas, D. Simeonidou, and N. Wada, “Experimental demonstration of a flexible filterless and bidirectional SDM optical metro/inter-DC network,” in 42nd European Conference on Optical Communication (ECOC) (2016), paper M.1.F.3.

K. Ishii, H. Hasegawa, K. Sato, M. Okuno, S. Kamei, and H. Takahashi, “An ultra-compact waveband cross-connect switch module to create cost-effective multi-degree reconfigurable optical node,” in European Conference and Exhibition on Optical Communication (ECOC) (2009), paper 4.2.2.

M. Jinno and T. Kodama, “Spatial channel network (SCN): introducing spatial bypass toward the SDM era,” in Optical Fiber Communications Conference (OFC) (2020), paper M2G. 1.

R. Dutta, A. E. Kamal, and G. N. Rouskas, Traffic Grooming for Optical Networks: Foundations, Techniques and Frontiers (Springer, 2008).

J. M. Simmons, Optical Network Design and Planning, Optical Networks (Springer, 2014).

Y. Sasaki, K. Takenaga, K. Aikawa, Y. Miyamoto, and T. Morioka, “Single-mode 37-core fiber with a cladding diameter of 248  µm,” in Optical Fiber Communications Conference and Exhibition (OFC), Los Angeles, California (2017), paper Th1H.2.

Cited By

OSA participates in Crossref's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (8)

Fig. 1.
Fig. 1. History and perspective of optical network evolution.
Fig. 2.
Fig. 2. SDM compatible optical node architectures. (a) Stacked conventional WXCs. (b) High-port-count MS-based HOXC. (c) CSS-based HOXC.
Fig. 3.
Fig. 3. Routine of the RSWA algorithm. (a) Physical network topology and shortest routes from A to H. (b) SCh and OCh created on shortest route 4 in the case of an express demand. (c) Grooming graph for shortest route 4 and candidate routes (blue, green, and black when $\alpha = {1}$ ) in the case of a local demand.
Fig. 4.
Fig. 4. Network models used in the simulation. Nodes indicated in blue are GNs, and the remaining nodes are NNs. (a) DT14, (b) NSF15.
Fig. 5.
Fig. 5. Required number of SLs for the DT14 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral-grooming-oriented), and express/local-hybrid (spatial-bypass-oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.
Fig. 6.
Fig. 6. Required number of SLs for the NSF15 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral-grooming-oriented), and express/local-hybrid (spatial-bypass-oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.
Fig. 7.
Fig. 7. Total node cost normalized by the cost of a conventional $1 \times 9$ WSS for the DT14 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral-grooming-oriented), and express/local-hybrid (spatial-bypass-oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.
Fig. 8.
Fig. 8. Total node cost normalized by the cost of a conventional $1 \times 9$ WSS for the NSF15 network model. Exp, Loc, Exp/Loc-G, and Exp/Loc-B are the four planning policies, i.e., express-only, local-only, express/local-hybrid (spectral grooming oriented), and express/local-hybrid (spatial bypass oriented), respectively. (a) High-port-count MS-based HOXC, (b) CSS-based HOXC.

Tables (1)

Tables Icon

Table 1. Traffic Model

Equations (3)

Equations on this page are rendered with MathJax. Learn more.

$${C_{{\rm wxc}}} = {S_{{\rm wxc}}}{D_n}\cdot{c_{{\rm wss}}},$$
$${C_{{\rm ms}}} = 2\cdot{{ C}_{{\rm ms}}}({{S_{{\rm ms}}}\cdot{D_n} + {a_n}} ) + {b_n}{D_n}\cdot{c_{{\rm wss}}},$$
$${C_{{\rm css}}} = 2{D_n}\cdot{c_{{\rm css}}} + {b_n}{D_n}\cdot{c_{{\rm wss}}}.$$