Search

This paper has been presented at 2020 European Conference on Networks and Communications (EuCNC). The fifth generation (5G) mobile network has paved the way for innovations across vertical industries. The integration of distributed intelligent edge into the 5G orchestrated architecture brings the benefits of low-latency and automation. A successful example of this integration is exhibited by the 5G-DIVE project, which aims at proving the technical merits and business value proposition of vertical industries such as autonomous drone surveillance and navigation. In this paper, and as part of 5G-DIVE, we present an aerial disaster relief system, called EagleEYE, which utilizes edge computing and machine learning to detect emergency situations in real-time. EagleEYE reduces training time by devising an object fusion mechanism which enables reusing existing datasets. Furthermore, EagleEYE parallelizes the detection tasks to enable real-time response. Finally, EagleEYE is evaluated in a real-world testbed and the results show that EagleEYE can reduce the inference latency by 90% with a high detection accuracy of 87%. This work has been partially funded by the H2020 EU/TW joint action 5G-DIVE (Grant #859881).

We present the architecture/testbed of an SDN-based optical-tunnel-network system (OPTUNS) for 5G edge-datacenters. Benchmarking results show that, compared to electrical-spine-leaf datacenter-network, OPTUNS achieves 82.6% power saving and significantly lower mean/p99 latency under high traffic load/locality.

Data center networks (DCNs) provide reliable and efficient infrastructures in support of a wide variety of emerging cloud and enterprise applications and services, such as grid/cloud computing, Big Data computing, and social networking. The proliferation of such data-rich delay-sensitive applications and services places a growing demand on DCNs that can deliver exceedingly high bandwidth, low latency, high reliability, and reduced power consumption at the same time. And yet, the existing state-of-the-art DCN technology that employs electrical switching of packets has been shown incapable of meeting such a demand.

An edge data center (EDC) network infrastructure is required to flexibly deliver massive bandwidth and ultralow latency for a 5G edge cloud. Existing electrical-switching-based infrastructure, however, has been shown incapable of meeting such requirements. In this paper, we present an EDC network architecture and prototype testbed, referred to as the intelligence-defined optical tunnel network system (OPTUNS). OPTUNS consists of a set of optical switching subsystems that operate collectively to facilitate packet transport through logical wavelength-based optical tunnels. These optical tunnels are governed by a software-defined-networking-based intelligent tunnel control system, in a proactive manner. As such, optical tunnels are always made available whenever needed. In essence, OPTUNS boasts several crucial features, including high scalability, massive wavelength reuse (yielding high bandwidth), proactive optical tunnel control (yielding ultralow latency), and fault tolerance. We have built an OPTUNS testbed, including 30 optical switching subsystem prototypes, that interconnect a total of 25 racks (400 servers). Benchmarking results show that OPTUNS achieves 82.6% power saving compared with electrical spine-leaf networks. Further, our NetPipe-based experimental results show that OPTUNS invariably achieves mean and p99 end-to-end latencies of less than 17 µs, regardless of traffic load and locality.

Optical WDM metro-ring networks have been considered to be promising solutions for wireless backhaul. In this paper, we propose an integrated traffic control scheme (ITCS), which facilitates circuit emulation service (CES) for wireless backhaul over a previously proposed experimental optical packet-switched WDM metro-ring network, HOPSMAN. The ITCS seamlessly integrates variable-bit-rate (VBR) CES wireless with connection-less best-effort data, satisfying stringent QoS requirements and retaining maximal network throughput. The ITCS ensures efficient setup of CES connections by employing a simple mean-rate-based distributed admission control followed by a novel slot-marking reservation. The total connection setup delay is comprised of the setup queueing delay and slot-marking delay. While the slot-marking delay is nearly a ring time, the mean setup queueing delay is formally computed through an accurate approximation based on an M/G/m queueing analysis. We show simulation results to demonstrate that the ITCS accommodates remarkably high CES traffic loads while satisfying a wide range of delay requirements for wireless backhaul. Taking the background best-effort traffic into account, ITCS achieves exceedingly low delay, jitter, and delay bound for CES under various traffic loads and burstiness.

Broadband access networking is an important form of basic infrastructure that supports competitive social and economic development. Emerging broadband applications such as IPTV, real-time network gaming, backhauling of broadband wireless connection data etc., demand new technological innovations to meet their ever-increasing performance-to-cost requirements. Optical fiber has been established as a future-proof medium capable of facilitating the next-generation network applications. For instance, in Feb. 2010 Google just announced the plans to build an experimental gigabit-per-second fiber-to-the-home (FTTH) network to households in North America for testing out new concepts in technologies and applications. Worldwide, FTTH deployment had surpassed thirty million users in 2009 and is still continuing to grow at a rapid pace.

Optical packet switching has been considered a prominent paradigm for future WDM networks to efficiently support a multitude of applications with diverse quality of service requirements. In this article we present the architectural design and experimental demonstration of a 10 Gb/s QoS optical packet switching system (QOPSS) for WDM networks. It embodies a set of many-to-one space switches, each of which handles the switching solely for a cluster of wavelengths. With the cluster-based optical switch design, QOPSS trades off limited statistical multiplexing gains for higher system scalability. By many-to-one, multiple packets that are carried by different internal wavelengths are scheduled to switch to the same output port but receive different delays afterward. QOPSS adopts downsized feed-forward optical buffers, yielding drastic reduction in packet loss probability in an economical manner. Significantly, through using four-wave-mixing wavelength converters at the output section, QOPSS permits optical packet preemption, thus achieving effectual QoS differentiation. The article presents both simulation and experimental testbed results to demonstrate the feasibility and superior packet loss/QoS performance of the system.

Advance lightpath reservation is a new research topic for connecting high-speed computer servers in lambda grid applications and for dynamic lightpath provisioning in the future optical internet. In such networks, users make call requests in advance to reserve network resources for communications. The challenge of the problem comes from how to jointly determine call admission control, lightpath routing, and wavelength assignment. In this paper, we propose an efficient Lagrangean relaxation (LGR) approach to resolve advance lightpath reservation for multi-wavelength optical networks. The task is first formulated as a combinatorial optimization problem in which the revenue from accepting call requests is to be maximized. The LGR approach performs constraint relaxation and derives an upper-bound solution index according to a set of Lagrangean multipliers generated through subgradient-based iterations. In parallel, using the generated Lagrangean multipliers, the LGR approach employs a new heuristic algorithm to arrive at a near-optimal solution. By upper bounds, we assess the performance of LGR with respect to solution accuracy. We further draw comparisons between LGR and three heuristic algorithms—Greedy, First Come First Serve, and Deadline First, via experiments over the widely-used NSFNET network. Numerical results demonstrate that LGR outperforms the other three heuristic approaches in gaining more revenue by receiving more call requests.

For future WDM MANs, optical packet-switching has been considered to be a promising paradigm that efficiently supports a wide range of Internet-based applications having time-varying and high bandwidth demands and stringent delay requirements. This article presents the design of an experimental testbed system for a high-performance optical packet-switched WDM metro ring network, HOPSMAN. HOPSMAN boasts three crucial features. First, it has a scalable architecture in which the number of nodes is unconstrained by the number of wavelengths. Second, HOPSMAN nodes are equipped with high-speed photonic hardware components, including fast tunable receivers and optical slot erasers, capable of performing speedy optical packet-switching operations. Third, HOPSMAN incorporates a MAC scheme that embodies efficient and dynamic bandwidth allocation, resulting in exceptional delay-throughput performance. The article presents the key hardware components by highlighting the challenging issues we faced and the solutions we proposed for the testbed implementation. Finally, to demonstrate the feasibility of HOPSMAN, the article describes the experimental setup and presents the results obtained from running a commercially available remote media player application on the system.

Optical Packet Switching (OPS) has been envisioned as a prominent future optical networking technology for datacentric IP over Wavelength Division Multiplexing (WDM) networks, or optical Internet. Such OPS technology however raises significant transport and Quality of Service (QoS) challenges due to technological limitations. To circumvent OPS limitations, we have proposed a new Optical Coarse Packet Switching (OCPS) paradigm, which uses in-band-controlled per-burst switching and advocates traffic control enforcement to achieve high packet-loss performance and differentiated services. Based on OCPS, we have constructed an experimental IP-over-WDM network, referred to as OPSINET. OPSINET consists of two major types of nodes-edge routers, and Optical Label Switched Routers (OLSRs), and is facilitated with an out-of-band Generalized Multi-protocol Label Switching (GMPLS) control network. In this paper, we first introduce the OCPS paradigm. We then present the architecture of OPSINET, describe the in-band header/payload modulation technique, and detail the operations of the edge routers, OLSRs, and GMPLS control

For IP-over-WDM networks, optical coarse packet switching (OCPS) has been proposed to circumvent optical packet switching limitations by using in-band-controlled per-burst switching and advocating traffic control enforcement to achieve high bandwidth utilization and quality-of-service (QoS). In this paper, we first introduce the OCPS paradigm. Significantly, we present a QoS-enhanced traffic control scheme exerted during packet aggregation at ingress nodes, aiming at providing delay and loss class differentiations for OCPS networks. Serving a dual purpose, the scheme is called (/spl psi/,/spl tau/)-Scheduler/Shaper, where /spl psi/ and /spl tau/ are the maximum burst size and burst assembly time, respectively. To provide delay class differentiation, for IP packet flows designated with delay-associated weights, (/spl psi/,/spl tau/)-Scheduler performs packet scheduling and assembly into bursts based on their weights and a virtual window of size /spl psi/. The guaranteed delay bound for each delay class is quantified via the formal specification of a stepwise service curve. To provide loss class differentiation, (/spl psi/,/spl tau/)-Shaper facilitates traffic shaping with larger burst sizes assigned to higher loss priority classes. To examine the shaping effect on loss performance, we analytically derive the departure process of (/spl psi/,/spl tau/)-Shaper. The aggregate packet arrivals are modeled as a two-state Markov modulated Bernoulli process (MMBP) with batch arrivals. Analytical results delineate that (/spl psi/,/spl tau/)-Shaper yields substantial reduction, proportional to the burst size, in the coefficient of variation of the burst interdeparture time. Furthermore, we conduct extensive simulations on a 24-node ARPANET network to draw packet loss comparisons between OCPS and just-enough-time (JET)-based OBS. Simulation results demonstrate that, through burst size adjustment, (/spl psi/,/spl tau/)-Shaper effectively achieves differentiation of loss classes. Essentially, compared to JET-based OBS using out-of-band control and offset-time-based QoS strategy, OCPS is shown to achieve invariably superior packet loss probability for a high-priority class, facilitating better differentiation of loss traffic classes.

Optical wavelength-division multiplexed (WDM) networks often include optical cross-connects with multigranularity switching capability, such as switching on a single lambda, a waveband, or an entire fiber basis. In addition, it has been shown that routing and wavelength assignment (RWA) in an arbitrary mesh WDM network is an NP-complete problem. In this paper, we propose an efficient approximation approach, called Lagrangean relaxation with heuristics (LRH), aimed to resolve RWA in multigranularity WDM networks particularly with lambda and fiber switches. The task is first formulated as a combinatorial optimization problem in which the bottleneck link utilization is to be minimized. The LRH approach performs constraint relaxation and derives a lower-bound solution index according to a set of Lagrangean multipliers generated through subgradient-based iterations. In parallel, using the generated Lagrangean multipliers, the LRH approach employs a new heuristic algorithm to arrive at a near-optimal upper-bound solution. With lower and upper bounds, we conduct a performance study on LRH with respect to accuracy and convergence speed under different parameter settings. We further draw comparisons between LRH and an existing practical approach via experiments over randomly generated and several well-known large sized networks. Numerical results demonstrate that LRH outperforms the existing approach in both accuracy and computational time complexity, particularly for larger sized networks.

In this letter, we propose a novel passive optical network (PON) architecture supporting radio-over-fiber signals and orthogonal frequency-division multiple access (OFDMA) PON signals. The architecture transports upstream multiple remote antenna's wireless signals using only one upstream wavelength. In addition, the windowed-orthogonal frequency-division-multiplexing technique is employed to mitigate interference during the integration process. Experimental results shows that 10-Gb/s OFDMA and three radio-frequency signals at 2.1 GHz are successfully transmitted over 20-km single-mode fiber in a 32-optical-network-unit PON.

In this paper, we propose an energy and cost efficient WDM/OFDMA PON system architecture, referred to as NEWOPS. Based on a two-tiered semi-tree topology, NEWOPS allows a large number of ONUs to be connected to OLT using only a handful of colorless transceivers. NEWOPS' unique architecture not only facilitates high spectral efficiency and flexible sub-channel bandwidth allocation, but also enables the system to achieve high energy and cost efficiency. We conduct a theoretical study to identify and resolve a new data-rate performance issue due to the colorless and wavelength re-use design of the system. Experimental results show that when only 5-GHz bandwidth is allotted for each wavelength, NEWOPS successfully achieves symmetric 20-Gb/s downstream and upstream OFDM-16QAM transmissions via the same fiber trunk. As a demonstration of our system, we show how NEWOPS achieves energy efficiency that is two to twelve times higher than that of a typical tree-based WDM/OFDM PON when supporting 40 ONUs.

Protocol testing leads to the synchronization problem should test sequences be applied to multiple distanced testers, namely under the multiparty configuration. This paper presents a novel synchronization paradigm which seamlessly unifies two synchronization techniques, self-synchroniable sequences and external synchronization operations, by means of the state expansion transformation . In the paradigm, the protocol specification is first transformed into a state expansion digraph with two pieces of datum augmented. They are: 1. (a) a zero cost assigned to each synchronization-problem-free crossing from one transition to another transition, and 2. (b) a weighted cost assigned to each external synchronization operation whenever the synchronization is deemed necessary. On the basis of the state expansion transformation, synchronizable, optimal sequences for testing can be efficiently derived. To demonstrate the viability of the proposed paradigm, we present the generations of two synchronizable sequences, namely the synchronizable preamble and the synchronizable distinguishing sequence, which have previously been used for the testing of the correctness of a protocol's transition. The paper also shows that the complexities of the two sequences generations are also polynomial-bounded, i.e., O (( np 2 ) log( np )) and O (( n 2 p 2 ) log( np )), respectively, where n and p are the numbers of states and input symbols of the protocol specification.

Multimedia applications including video data require the smoothing of video playout to prevent potential discontinuity. In this paper, we propose a dynamic video playout smoothing method, called the Video Smoother, which dynamically adopts various playout rates in an attempt to compensate for high delay variance of networks. Specifically, if the number of frames in the buffer exceeds a given threshold (TH), the Smoother employs a maximum playout rate. Otherwise, the Smoother uses proportionally reduced rates in an effort to eliminate playout pauses resulting from the emptiness of the playout buffer. To determine THs under various loads, we present an analytic model assuming the Interrupted Poisson Process (IPP) arrival. Based on the analytic results, we establish a paradigm of determining THs and playout rates for achieving different playout qualities under various loads of networks. Finally, to demonstrate the viability of the Video Smoother, we have implemented a prototyping system including a multimedia teleconferencing application and the Video Smoother performing as part of the transport layer. The prototyping results show that the Video Smoother achieves smooth playout incurring only unnoticeable delays.

Multimedia systems combine a variety of information sources, such as voice, graphics, animation, images, and full-motion video, into a wide range of applications. The paper initially categorizes existing multimedia applications into three classes: noninteractive-oriented, interactive-oriented client-server-based, and interactive-oriented peer-party-based. In particular, the paper examines interactive-oriented applications and provides an in-depth survey of the media synchronization problem for the design of these applications. The paper then presents our prototyping system, called the Distributed Multimedia Teleworking System (DMTS), which allows two or more remote systems in collaboration to access and modify multimedia data through a network in a fully synchronous fashion. The system has been developed over TCP/IP and an FDDI network, using an XVideo D/A card. The media supported by DMTS include text, graphics, voice, and video. DMTS employs a master-slave collaboration model to maintain the coherence of the text and graphics data being simultaneously modified. Moreover, DMTS also adopts effective mechanisms to reduce skew (asynchrony) and jitter delays between video and voice streams. Finally, the paper demonstrates that DMTS achieves a maximum throughput of 13 frames per second, and reveals that the throughput bottleneck resides in the hardware capture and D/A processing of video frames.

Call Admission Control (CAC) has been accepted as a potential solution for supporting diverse, heterogeneous traffic sources demanding different Quality of Services (QOSs) in ATM networks. Also, CAC is required to consume a minimum of time and space to make call acceptance decisions. In this paper, we present an efficient neutral-network-based CAC (NNCAC) mechanism for ATM networks with heterogeneous arrivals. All heterogeneous traffic calls are initially categorized into various classes. Based on the number of calls in each class, NNCAC efficiently and accurately estimates the cell delay and cell loss ratio of each class in real time by means of a pre-trained neutral network. According to our decent study which exhibits the superiority of the employment of analysis-based training data over simulation-based data, we particularly construct the training data from a heterogeneous-arrival dual-class queueing model M^[^N^"^1^] + I^[^N^"^2^]/D/1/K, where M and I represent the Bernoulli and interrupted Bernoulli processes, and N"1 and N"2 represent the corresponding numbers of calls, respectively. Analytic results of the queueing model are confirmed by simulation results. Finally, we demonstrate the profound agreement of our neural-network-based estimated results with analytic results, justifying the viability of our NNCAC mechanism.

Multimedia communications often require intramedia synchronization for video data to prevent potential playout discontinuity resulting from network delay variation (jitter) while still achieving satisfactory playout throughput. In this paper, we propose a neural network (NN) based intravideo synchronization mechanism, called the intelligent video smoother (IVS), operating at the application layer of the receiving end system. The IVS is composed of an NN traffic predictor, an NN window determinator, and a window-based playout smoothing algorithm. The NN traffic predictor employs an on-line-trained back-propagation neural network (BPNN) to periodically predict the characteristics of traffic modeled by a generic interrupted Bernoulli process (IBP) over a future fixed time period. With the predicted traffic characteristics, the NN window determinator determines the corresponding optimal window by means of an off-line-trained BPNN in an effort to achieve a maximum of the playout quality (Q) value. The window-based playout smoothing algorithm then dynamically adopts various playout rates according to the window and the number of packets in the buffer. Finally, we show that via simulation results and live video scenes, compared to two other playout approaches, IVS achieves high-throughput and low-discontinuity playout under a mixture of IBP arrivals.

In this paper, we propose an integrated traffic control scheme (ITCS), which facilitates circuit emulation service (CES) for wireless backhaul over a previously proposed experimental optical packet-switched WDM ring network, HOPSMAN. To ensure efficient setup of CES connections, ITCS employs a simple mean-rate-based distributed admission control followed by a novel slot-marking reservation. The total connection setup delay is comprised of the setup queueing delay and slot-marking delay. While the slot-marking delay is nearly a ring time, the mean setup queueing delay is formally computed through an accurate approximation based on an M/G/m queueing analysis. Simulation results show that ITCS accommodates remarkably high CES traffic loads while satisfying a wide range of delay and jitter requirements for wireless backhaul.

On behalf of the IEEE Communications Society and members of the organizing committee, it is our pleasure to welcome you to the 2013 IEEE 14th International Conference on High Performance Switching and Routing (IEEE HPSR 2013) in Taipei, Taiwan. The conference continues to be a leading forum for practitioners and researchers in this area.

In this paper, we propose an input access scheme for input-queued ATM multicast switches, achieving high system throughput, low packet delay and packet loss probability. Multicast and unicast packets of each input port are separately queued. Multicast queues take priority over the unicast queues, and both types of queues are fairly served in a cyclic-priority access discipline. In particular, each unicast queue is handled on a window-service basis, and each multicast packet is switched in a one-shot scheduling manner. To evaluate the performance of the access scheme, we propose an approximate analysis based on a simplified cyclic-priority model for anN×N finite-buffer multicast switch possessing Bernoulli multicast and unicast arrivals, with window-service (for unicasting) and one-shot scheduling (for multicasting) both taken into account. Finally, we show simulation results to demonstrate the accuracy of the approximate analysis and the superiority of the scheme over existing schemes with respect to normalized system throughput, mean packet delay, and packet loss probability.

Terminal-pair reliability determines the probabilistic reliability between two nodes of a network, given the failure probabilities of all links. The problem has been successfully solved by the path-based partition method, network reduction technique, and the combination of the two. The path-based partition algorithm, however, requires repeated execution of a path-search procedure in an attempt to locally minimize the total number of subproblems generated. We propose a cut-based partition algorithm based on the minimum cut separating the source from the remaining network nodes. This cut-based algorithm makes no attempt to locally minimize the number of subproblems. Instead, the algorithm leads to a great reduction of the computation time for the partitioning of each subproblem. Experimental results show the superiority of the cut-based algorithm over the path-based algorithm especially for complex networks. By further employing the network reduction technique, our cut-based partition algorithm outperforms the path-based partition algorithm for all benchmarks and random sparse or dense networks. Based on the cut-based partition algorithm, we further present an incremental method of computing the terminal-pair reliability for hierarchical networks. The incremental method results in minimal recomputation of the reliability should the network configuration change.

In this paper, we propose a QoS parallel incremental scheduling (QPIS) algorithm for a variant of an earlier-proposed 10-Gb/s QoS optical packet switching system. The variant system consists of optical pseudo-Banyan space switches, a handful of fiber-delay-line (FDL)-based optical buffers, and four-wave-mixing (FWM) wavelength converters. The QPIS minimizes the loss probability for high-priority packets while maximizing system throughput and satisfying two constraints (switch-contention free, and buffer-contention free). Most notably, we prove that QPIS is incremental, i.e., the computed packet sets within each time slot are monotonically non-decreasing, and then derive the computational complexity. From simulation results that pit the QPIS algorithm against two other sequential algorithms, we show that QPIS outperforms these algorithms on packet loss probability, QoS differentiation, and computational complexity.

In this paper, we propose a novel integrated traffic control scheme, called ITCS, which seamlessly supports connection-oriented and connectionless services for a previously proposed experimental optical packet-switched metro WDM network, HOPSMAN. ITCS aims to efficiently satisfy the diverse quality-of-service (QoS) requirements while retaining maximal network throughput. Through a flexible reservation scheme, ITCS avoids the receiver-contention problem inhered in current WDM network, achieving high statistical multiplexing gain for isochronous traffic. By employing constant mean rate reservation and with an effective distributed call admission control (CAC), ITCS establishes isochronous traffic connections only within a single ring time under normal loads and achieves exceedingly low VBR delay. Furthermore, we develop a novel approximation to acquire the accurate results of the expected connection setup delay by means of an M/G/m queueing analysis. Extensive simulation results show that the mean setup queueing time is in profound agreement with the analytic results, and that ITCS accommodates maximal isochronous traffic with remarkable QoS performance and still retaining exceptional aggregate system throughput.

We experimentally demonstrate a superior performance of 2.1-Tb/s·km OFDM-signal transmission over 100-km long-reach PONs, which is achieved by using an EAM and direct-detection, and adopting the 128-QAM format and adaptive subcarrier power weighting.

Distributed queue dual bus (DQDB) has been adopted as the IEEE 802.6 standard for metropolitan area networks (MANs). The access mechanism is based on an implicit distributed queue maintained at each node. DQDB has been shown to achieve maximal aggregate throughput and minimal access delay independent of the size and speed of the network. However, as the size and load of the network increase, DQDB gives rise to throughput unfairness problems due to the long propagation delay. This paper initially presents a survey and assessment of recently proposed remedial DQDB mechanisms. These approaches are categorized as time division multiplexing (TDM)-oriented versus non-TDM-oriented, and exhaustive versus non-exhaustive operations. Regarding TDM as the most promising mechanism, the paper then provides the performance analysis of one TDM-oriented method, called multiple segment control (MSC), proposed by the authors. In the analysis, the network is modelled as a set of identicalM/G/1 queues, each with different services for transmitting the first and remaining segments in a busy period. The average aggregate access delay is then analyzed via an embedded Markov model. The paper shows simulation results to demonstrate the precision of the performance analysis and the performance superiority of MSC over non-TDM-oriented DQDB mechanisms. Furthermore, as will be shown, the network using the TDM-oriented method may incur bandwidth waste should active stations fail. The paper finally proposes a guardian reliability scheme to ensure the robustness of networks.

Signaling control schemes for wireless multimedia services are expected to maximize the number of newly-accepted mobile terminals (MTs), i.e., system throughput, while maintaining the quality of service (QoS) requirements for the existing MTs in the system. In this paper, we propose a γ-percent time-bound signaling control (γ-TBSC) scheme that resolves a maximum number of MTs mostly within a given bounded time, subject to assuring γ-percent soft guarantee for existing MTs. The γ-TBSC scheme performs signaling control in two phases: the call-invitation phase, and the contention-resolution phase. In the first phase, an access point invites a number of so-called potential MTs to make reservations during a call invitation interval (CII). If any collision occurs, γ-TBSC performs contention resolution in the second phase until all potential MTs are resolved successfully. Most importantly, prior to the two-phase operation, γ-TBSC analytically and statistically determines the potential MT number and the CII length such that the system throughput is maximized and the QoS is soft guaranteed. From simulation results that pit γ-TBSC against several variants of carrier sense multiple access with collision avoidance (CSMA/CA) schemes, we delineate that γ-TBSC outperforms these schemes with respect to throughput and delay performance.

In this paper, we present the design of an experimental optical-header processing and access control system (OPACS) for an optical packet switched (OPS) WDM metro slotted-ring network. Each control header is in-band time-division-multiplexed with its payload. OPACS enables the optical headers across all parallel wavelengths to be efficiently received, modified, and re-transmitted based on a wavelength-time conversion technique. Bandwidth access in OPACS is governed by a versatile access control scheme, called the distributed multi-granularity and multiwindow reservation (DMGWR) mechanism. With DMGWR, the OPACS network is shown to outperform two existing networks with respect to throughput, access delay, and fairness under various traffic patterns. Experimental results demonstrate that all optical headers are removed and combined with the data in a fully synchronous manner, justifying the viability of the system.

This paper demonstrates a novel distributed control PON system (DCPON). Equipped with a distributed adaptable DBA algorithm that enables real-time precise bandwidth allocation, DCPON achieves high utilization, fairness, low request/response time, and superior service differentiation.

We propose a novel OFDMA-PON architecture enabling seamless integration of broadband data and wireless signals from multiple remote antennas. We demonstrate that three 20MHz wireless channels at 2.1GHz-band are trimly overlaid with 10Gbps OFDMA-PON signal.

We present the design/experimentation of a medium-access control and processing system (MACOPS) for an optical packet-switched WDM network. MACOPS enables optical parallel headers to be efficiently received/modified via wavelength-time conversion, and achieves exceptional throughput/delay/fairness performance.

In this paper, we present the access and control design of a high-performance optical packet-switched WDM metro ring network (HOPSMAN). HOPSMAN has been designed for networks and nodes to be unconstrained by the number of wavelengths. It includes a handful of nodes that are equipped with fast optical slot erasers making bandwidth reusable and achieving greater bandwidth efficiency. In essence, HOPSMAN incorporates a versatile medium access control (MAC) scheme, which embodies efficient and fair bandwidth allocation in accordance with a quota being exerted probabilistically. The quota is analytically derived with the number of slot-eraser-nodes taken into account. The scheme also employs a new notion of credit to regulate flexible access of remaining bandwidth that is suitable for the metro environment with bursty traffic. With the MAC scheme, HOPSMAN is shown to achieve exceptional throughput, delay, and fairness performance under a wide range of traffic settings via simulation results.

The paper presents the architecture and experimentation of a 10-Gb/s QoS-enabled almost-all-optical packet switching system (QOPS) for metro WDM networks. By applying cluster-based wavelength sharing and downsized single-staged optical buffers, QOPS is featured by its highly scalable and cost-effective design. In this paper, we first introduce the switch architecture, system operation, and the key techniques. We describe the in-band header/payload modulation and optical label swapping that is suitable for high-speed optical packet switching. We also present the design of the highly efficient Four-Wave Mixing wavelength converters for packet preemption. We then present an adaptive bifurcated routing (ABR) that directs same-connection packets to different switch clusters according to optimal bifurcation probabilities. Experimental and simulation results demonstrate that QOPS can achieve superior packet-loss performance, QoS differentiation, and minimize traffic blocking probability.

The paper presents the design and experimentation of a high-performance optical packet-switched metro WDM ring network (HOPSMAN). Equipped with novel medium access control, HOPSMAN achieves superior bandwidth efficiency, access delay, fairness, and bursty traffic adaptation.

In this paper, we first introduce an optical coarse packet switching (OCPS) paradigm. In principle, OCPS advocates the enforcement of traffic engineering and control to realize bandwidth-on-demand on sub-wavelength basis while circumventing optical packet switching limitations. Based on OCPS, we have constructed an experimental optical IP-over-WDM network testbed, referred to as OPSINET. In the paper, we present the architectural design of OPSINET and its Hybrid Contention Control (HCC) mechanism aiming to satisfy various classes of loss QoS guarantees. The mechanism preventively adopts appropriate burst sizes for different traffic classes during packet burstification at ingress routers. It also reactively performs prioritized contention resolution at OLSRs in the presence of contention. Experimental results show that HCC invariantly achieves low loss probability as a result of burstification-based traffic shaping. Moreover, it provides satisfied loss guarantees for high-priority classes while incurring minimal loss degradation for lower-priority classes.

In this paper, we propose an efficient approximation approach, called Lagrangean relaxation with heuristics (LRH), aimed to resolve routing and wavelength assignment (RWA) for multi-granularity WDM networks facilitating fiber, waveband, and lambda switching capabilities. The task is first formulated as a combinatorial optimization problem in which the bottleneck link utilization is to be minimized. The LRH approach performs constraint relaxation and derives a lower-bound solution index according to a set of Lagrangean multipliers generated through subgradient-based iterations. In parallel, using the generated Lagrangean multipliers, the LRH approach employs a new heuristic algorithm to arrive at a near-optimal upper-bound solution. With lower and upper bounds, we delineate the performance of LRH with respect to accuracy and convergence speed under different parameter settings. We further draw comparisons between LRH and a typical linear programming (LP) approach via experiments over the widely-used NSFNET and three randomly generated networks. Numerical results demonstrate that LRH outperforms the LP approach in both accuracy and computational time complexity particularly for larger sized networks.

Most existing contention access schemes are inherently unstable resulting in exponentially deteriorating throughput under increased traffic loads. In this paper, we propose a wide-sense stable, efficient hexanary-feedback contention access (HFCA) scheme, capable of providing signaling traffic high performance while retaining maximal throughput for wireless access networks. HFCA performs incremental contention resolution via a two-phase process. The two-phase process is augmented with hexanary feedback control facilitated by a pdf-based multiuser estimator (PMER) implemented at the physical layer. We present throughput and stability analyses in which HFCA is shown wide-sense stable, and the strict-sense stability condition is derived. Finally, analytic and simulation results delineate that, HFCA achieves high performance with respect to maximum stable and saturated throughputs, access delay, and blocking probability.

In this chapter, we present a network framework in which Variable Bit Rate (VBR) voice is transported over a wireless Local Area Network (LAN) using two novel intelligent control systems. First, the Intelligent Multiple Access Control System (IMACS) at the base station governs the medium access of the uplink channel. In particular, IMACS supports both Available Bit Rate (ABR) data traffic and in-band signaling control (SCR) traffic for VBR voice based on a TDM-based contention access scheme. Significantly, dynamic allocation of contention bandwidth between ABR and SCR traffic is facilitated through predicting ABR self-similar traffic characteristics based on a neural-fuzzy approach. As a result, IMACS offers various QoS guarantees and a maximum of network throughput irrelevant to traffic variation. With such guarantees, the second system- Intelligent Voice Smoother (IVoS) at the application layer of each mobile terminal facilitates intramedia synchronization of voice data streams. The traffic predictor of IVoS predicts three traffic characteristics of every newly encountered talkspurt period. Based on the predicted characteristics, IVoS determines the corresponding buffering delay to be imposed on the first frame. All subsequent frames of the talkspurt can be playout in a quasi-ConstantBit-Rate (CBR) manner. Finally, we demonstrate via experimental results that with such intelligent control, the playout Quality of Service (QoS) of VBR voice can be guaranteed irrespective of any traffic and load variation.

The random access channel (RACH) incurs a collision if more than one user make signaling requests over the same time slot, spreading code, or antenna element, in TDMA, CDMA, or SDMA wireless networks, respectively. We propose a hexanary-feedback collision resolution paradigm (HFCRP), capable of leveraging access efficiency of the RACH for all above wireless networks. HFCRP performs incremental contention resolution, managing a small subset of users at a time via a two-phase process. In the first phase, a group of users is probabilistically admitted, with a negligible probability of the group size greater than five. In the second phase, all users in the group are efficiently resolved. The two-phase process is augmented with hexanary feedback control facilitated by an envelope-PDF-based multi-user estimator (MR) implemented at the physical layer. Significantly, the MR measures the exact number of transmitting users (zero to five) in a contention slot by matching the envelope PDFs histograms of received signals to a library of pre-constructed PDFs. These six outcomes comprise the hexanary feedback being broadcast to users during the real-time operation of the two-phase process. Analytic and simulation results delineate that, performing in conjunction with the MR, HFCRP achieves high performance with respect to maximum stable throughput, saturated throughput, access delay, and blocking probability.

Prevailing weight-based scheduling algorithms advocate static or coarse-grained simplification of timestamp computation, resulting in performance bottlenecks and unnecessary downgrades under heavy and normal computation loads, respectively. Our goal is to design a scheduler that enables quantitative trade-off balance between performance and complexity. In this paper, we propose a stepwise QoS scheduler (SQS), enabling packet to burst scheduling under normal to heavy computation loads, based on a new notion of window. Window sizes of unity and greater than one correspond to packet and burst scheduling, respectively. SQS exerts simple FIFO service within the window and guarantees stepwise weight-proportional service at the window boundary. As a result, quantitative trade-off balance can be facilitated via the adjustment of the window size. We further formally specify the SQS service guarantee for a flow by a stepwise service curve, which is a function of W and /spl theta//sub min/-the least-upper-bound delay incurred by the first packet of the busy period of the flow. Finally, we demonstrate through experimental results that, by applying a small window (W/spl les/10), SQS outperforms weighted fair queueing (WFQ) and performs as superior as worst-case fair weighted fair queueing (WF/sup 2/Q) with respect to throughput fairness, mean delay, and worst-case delay fairness.

In this paper, we propose a QoS burstification (QBT) mechanism on the basis of a new notion window. The QBT mechanism efficiently emulates the burst behavior by exerting simple FIFO service within the window and weight-proportional service at the window boundary. As will be shown, QBT achieves fairness and mean-burstification-delay guarantees among various traffic classes.

We propose a synchronization paradigm, called SYNC*, offering two quality of service (QoS) requirements, skew and throughput, for multimedia communications, SYNC* is composed of three components: an intelligent video smoother (IVS), synchronizer, and multi-thread priority manager (MPM). While IVS and synchronizer achieves intramedia and intermedia synchronization, respectively, MPM ensures satisfactory throughput before playout. The main goal of the paper is to present in detail the design and performance analysis of the second component, the synchronizer. The synchronizer is aimed at achieving bounded skew between different media through periodic marking of correlated frames. The minimum marking interval yielding a given skew is analytically determined based on a Markov batch Bernoulli process (MBBP)/D/K/1 queueing model. We demonstrate the profound agreement between analytic and simulation results. The results also indicate that, the playout quality with respect to skew is highly relevant to the traffic burstiness and decoder buffer size. On the basis of a fixed marking interval, the greater the burstiness and buffer size, the poorer the playout quality.

Quality of service (QOS) for diverse multimedia applications can be satisfied by the provision of synchronization platforms furnishing intramedia/intermedia synchronization and efficient process multitasking at end systems. As a result, the design of viable platforms should confront two concerns: network delay and jitter, and system multitasking. In this paper, we propose an intelligent synchronization platform, called SYNC*, satisfying given QOSs in terms of skew and throughput. SYNC* is composed of three components: intelligent video smoother (IVS), synchronizer, and multi-thread priority manager (MPM). While IVS and synchronizer deal with the network delay/jitter concern, MPM manages the challenge of system multitasking. We demonstrate via prototyping results that SYNC* achieves superior playout with respect to skew and throughput. We also observe that, the absence of either IVS/synchronizer or MPM results in unsatisfactory playout.

Marginal reliability importance (MRI) of a link is a quantitative measure reflecting the significance of the link in contributing to the terminal-pair reliability (TR) of a given network. The computation of MRI for general networks has been shown to be an NP-complete problem. Moreover, attention has been drawn to a particular set of networks, called reducible networks, which can be simplified to source-sink (two-node) networks via six simple reduction axioms. The computational complexity of the MRI problem for such networks has been shown to be polynomial bounded. In this paper, we first propose a new reduction axiom, referred to as triangle reduction. Networks which can be fully reduced to source-sink networks by the triangle reduction axiom, in addition to the six reduction rules, are further defined as reducible/sup +/ networks. For efficient computation of MRI for reducible/sup +/ networks, we further propose a two-phase algorithm. The algorithm performs network reduction in the first phase, and backtracks the reduction steps and computes MRIs in the second phase. The two-phase algorithm, as shown, yields a linearly bounded complexity for the computation of MRI for reducible/sup +/ networks.

The authors propose a new dual-mode-control local area network/metropolitan area network, called DDMNET, based on a single-folded bus topology. Stations are connected to communication channels by means of three passive taps. At light loads, each station transmits packets on the outbound channel as long as the channel is sensed as being idle. As the load of the network increases, the occurrence of collisions will lead the network to be controlled under an implicit token-passing scheme. Transmissions at this mode are thus guaranteed collision-free and will be of bounded delay. It is shown that DDMNET differs from other dual-mode networks in several ways. Throughput superiority of DDMNET is demonstrated over four existing schemes through performance analysis and simulation results. >

MRACMAN (multislot reservation alternating cycle control metropolitan area network) is a new high-throughput backbone MAN aimed at interconnecting a collection of local area networks (LANs) via simplified bridges. The backbone network and LANs are all based on a dual unidirectional slotted bus architecture. The simplified bridges (the backbone nodes) perform only simple buffering and routing logic. In addition, the high-throughput scheme (MRAC) is used as the medium access control protocol on both the backbone network and each interconnected LAN. MRACMAN, as shown, offers high throughput by means of efficient and parallel transmissions among all local LANs and the backbone network. With a reasonable degree of traffic locality assumed, simplified bridges are guaranteed congestion-free. The authors provide a performance evaluation and comparison via simulation results to demonstrate the feasibility of congestion-free bridges and throughput superiority of MRACMAN over other MANs. >

A prototype workstation, called communication protocol analyzer and simulator system (CPASS), aimed at the modeling, verification, analysis, and simulation for real-time networks by means of a formal and automated approach is presented. The system is based on an extended finite state machine model, called SDL/sup N/ (a variant of SDL). It is composed of six components: network topology editor, SDL/sup N/GR editor, protocol verifier and analyzer, SDL/sup N/-to-C compiler, simulator, and statistical generator and comparator. The functionality of each component of CPASS is described. An example of a token ring network is used to illustrate the complete processes of modeling, analysis, and simulation. >

Fiber distributed data interface-II (FDDI-II) is a hybrid high speed local area network (HSLAN) providing circuit switched and packet switched services. Although the packet switched service remains connectionless, the circuit switched service is connection-oriented. The weaknesses of the media access control (MAC) protocol of FDDI-II is discussed. A new access scheme, called FDDI-II*, aimed at providing distributed control and access of bandwidth using connectionless service for both packet and isochronous data transmissions is proposed. Simulation results that demonstrate the performance superiority of FDDI*-II to FDDI-II in terms of isochronous data blocking probability and packet data delay are presented.

Multimedia communications require intra-media synchronization for video data to prevent potential playout discontinuity resulting from network delay variation (jitter) while still achieving satisfactory playout throughput. In this paper, we propose a neural-network-based intra-media synchronization mechanism, called neural network smoother (NNS). NNS is composed of a neural network (NN) traffic predictor, an NN window determinator, and a window-based playout smoothing algorithm. The NN traffic predictor employs an online-trained backpropagation neural network (BPNN) to periodically predict future traffic characteristics. The NN window determinator determines the corresponding optimal window by means of an off-line-trained BPNN in an effort to achieve a maximum of the playout quality value. According to the window, the window-based playout smoothing algorithm then dynamically adopts various playout rates. Compared to two other playout approaches, simulation results show that NNS achieves high-throughput and low-discontinuity playout under a variety of traffic arrivals.