Yejun Liu

Optimizing Backup Optical-Network-Units Selection and Backup Fibers Deployment in Survivable Hybrid Wireless-Optical Broadband Access Networks

Survivability is one of the key issues in hybrid wireless-optical broadband access networks (WOBAN) since the single segment failure can cause huge data loss. The single segment failure refers to a scenario where all optical-network-units (ONUs) are disconnected with the optical line terminal (OLT). Previous schemes focus on protecting WOBAN against single segment failure by deploying backup fibers. However, previous schemes suffer from two key problems. First, they ignore optimizing the selection of backup ONUs, which determines the cost of recovering the traffic interrupted by the failure. Second, they underutilize the residual capacity of segments, thus requiring higher backup fibers cost. In this paper, we propose a new and efficient scheme, called Optimizing Backup ONUs selection and backup Fibers deployment (OBOF), to enhance the survivability of WOBAN against the single segment failure. Our OBOF is composed of two consecutive steps, backup ONUs selection and backup fibers deployment. In the first step, aiming to minimize the cost of recovering the traffic interrupted by the failure, the simulated annealing (SA) algorithm is customized to optimize the selection of backup ONUs. In the second step, most importantly, an enhanced greedy cost-efficiency (EGCE) algorithm is proposed to optimize the deployment of backup fibers. Our EGCE consists of a novel remote backup segment (RBS) method, which can efficiently utilize the residual capacity of the segments, and a Bound on Length of Backup-optical-path (BLB) method, which limits the increase in recovery time induced by RBS. Extensive experimental results demonstrate that our OBOF scheme outperforms the previous schemes significantly, especially in the scenario of higher traffic demand.

Protection based on backup radios and backup fibers for survivable Fiber-Wireless (FiWi) access network

Abstract In Fiber-Wireless (FiWi) access network, the optical back-end is vulnerable to the network component failure due to its tree topology. Any failure at the optical back-end may cause huge data loss. Thus, the survivability in FiWi is an important issue, especially the protection for the optical back-end. Some works propose to protect the optical back-end in FiWi by means of the wireless rerouting in the wireless front-end. However, these works cannot guarantee that there are always the available wireless paths for the traffic rerouting. In this paper, we divide the failures at the optical back-end into ONU-level failure and OLT-level failure according to the failure severity. To tolerate the ONU-level failure, we allocate each ONU a partner ONU and establish the wireless-backup-path between them by deploying backup radios on the traversed wireless routers. To tolerate the OLT-level failure, we cluster all segments in the network and place the backup fibers among the segments in each cluster to establish the protection ring. Thus, each pair of segments in the same cluster can backup for each other along the optical-backup-paths on the protection ring. We propose the heuristic algorithms to minimize the cost of backup radios and the cost of backup fibers. The simulation results demonstrate that the proposed algorithms are effective in enhancing the survivability of FiWi, while requiring less cost than the previous works.

Virtual network planning for converged optical and data centers: ideas and challenges

With the development of WDM techniques and cloud computing, incorporating the optical backbone into data centers has received extensive attention. Virtual network (VN) planning has been proposed for this new networking paradigm by taking advantage of network virtualization. VN planning is essentially a united virtualization of optical and server resources in a collaborative manner. In this article, we first elaborate the advantages and problems of existing VN planning schemes through a comprehensive survey. Next, we provide a VN planning scheme. The simulation results demonstrate the effectiveness of our strategy, particularly in the context of power outage and evolving recovery. Finally, the future opportunities and challenges are also discussed.

Planning of survivable long-reach passive optical network (LR-PON) against single shared-risk link group (SRLG) failure

Long-Reach Passive Optical Network (LR-PON) is known as a promising and economical solution for Next-Generation PON (NG-PON). Survivability is one of the key issues in the planning of LR-PON because massive high-rate traffic flows may be interrupted in case of network component failure. However, the survivability issue for LR-PON is addressed in little works. More importantly, most of the previous works focus on single distribution fiber failure and remain untouched the simultaneous failure of multiple distribution fibers, which is a possible failure scenario in LR-PON. In this paper, we focus on the survivability of LR-PON against single Shared-Risk Link Group (SRLG) failure. A reliability model is proposed to represent the disconnection probability of the ONUs. Based on this reliability model, we propose a novel Backup Fibers Protection (BFP) scheme. In the BFP scheme, we deal with the optimization problem of allocating backup capacity and deploying backup fibers. Under the constraint of reliability requirement, our objective is to fully protect all traffic demand in the network with the minimum deployment cost of backup fibers. Both ILP-based approach and heuristic approach are proposed to solve the optimization problem in the BFP scheme. To the best of our knowledge, this paper is the first work regarding the survivability of LR-PON against single SRLG failure. Through extensive simulation, we investigate the performance of BFP and demonstrate its effectiveness in different scenarios.

Markov-based vertical handoff decision algorithms in heterogeneous wireless networks

Our work has considered the QoS factors and constructed a reward function to model the QoS properties.Our work has considered MTs velocity, network access cost and switching cost for vertical handoff decision.We have applied G1 and entropy methods to calculate the weight of each QoS factor in the reward function.By taking the MTs velocity into consideration, our work can avoid unnecessary handoffs. In this paper, we propose a received signal strength (RSS)-based single-attribute handoff decision algorithm at first, and investigate handoff decision model based on connection lifetime, which can keep mobile terminals (MTs) staying long enough in the preferred network. Since the preferred quality of service (QoS) parameters may be distinct among different MTs, we then formulate the vertical handoff decision problem as a Markov decision process, with the objectives of maximizing the expected total reward and minimizing average number of handoffs. A reward function is constructed to assess the QoS during each connection, and the G1 and entropy methods are applied in an iterative way, by which we can work out a stationary deterministic handoff decision policy. Numerical results demonstrate the superiority of our proposed schemes compared with other existing algorithms.

Resource management and control in converged optical data center networks

In addition to the optical interconnection among servers for the Intra Data Center (IDC), the optical interconnection of geographically distributed data centers also becomes increasingly important since data centers are geographically distributed so that one data center maybe far away from another. So the converged Optical and Data Center Network (ODCN) emerges as the times require. In the ODCN, each optically interconnected IDC locates at the edge of the optical backbone. In this article, we first make an extensive survey on the resource management and control in the ODCN, and we find that: (1) for this new network paradigm, the intelligent coexistence of heterogeneous technologies should be considered because the ODCN will be required to satisfy diverse and highly dynamic network services; (2) the integrated virtualization of backbone bandwidth and computing resources should be performed for the resource management in ODCNs, with the objective to improve the underlying infrastructure utilization; (3) after performing the integrated virtualization, a set of virtual networks are generated, and each of them has virtual lightpaths and virtual machines. But a static virtual network merely satisfies a certain range of Service Level Agreements (SLAs), and it merely adapts to a particular network status. When the SLA significantly varies or the Quality of Transmission (QoT) gets worse, it is necessary to trigger the dynamic planning for the virtual network reconfiguration; (4) to decrease the control overhead and the delay of making decisions for the dynamic planning, it is practical to embed a highly effective network control plane into intelligent ODCN. Consequently, we make a blueprint where we execute the intelligent coexistence, integrated virtualization and dynamic planning for the resource management in ODCNs. Some preliminary works and simulation results will guide the future work.

Novel temporal/spectral coding technique based on fiber Bragg gratings for fiber optic CDMA application

A novel hybrid temporary spectral coding technique based on fiber Bragg gratings for fiber-optic CDMA application is proposed and demonstrated experimentally for the first time to our knowledge. Significant improvement in fiber-optic CDMA networks at high bit rates is expected.

Ring-based protection scheme for survivable Fiber-Wireless (FiWi) access network considering multiple failures

Survivability is one of the key issues in Fiber-Wireless (FiWi) access network since network component failure may interrupt many high-rate traffic flows, especially in the segment failure scenario. Previous works propose to protect FiWi against segment failure by deploying backup fibers among different segments. However, these works consider only single segment failure. Also, they underutilize the capacity of backup fibers and thus require huge cost of backup fibers. To conquer these challenges, we propose an efficient protection scheme called Ring-based Protection considering Multiple Failures (RPMF) in this paper. In RPMF, we build the protection rings by deploying backup fibers among the segments, aiming at the survivable FiWi against multiple failures with the minimum cost of backup fibers. Simulation results demonstrate that our RPMF outperforms the previous works significantly in terms of cost saving and survivability enhancement.

Optimization mechanisms in multi-dimensional and flexible PONs

The current Passive Optical Network (PON) has faced many challenges when heterogeneous access techniques and multiple resource multiplexing ways are involved. Thus, we require a unified control and management of time, spectrum and mode, which is a necessary standardization for PONs. For this end, the multi-dimensional and flexible PON (Mflexi-PON) will emerge as an ideal solution of supporting the heterogeneous access with different resource multiplexing ways. But a high-efficient Mflexi-PON has its own problem. In term of the physical layer of Mflexi-PONs, some signals include a series of orthogonal subcarriers, or transmission modes should have the orthogonal property. Once multiple subcarriers have the same phase, the instantaneous power of a signal will exceed the average value, thus leading to a high Peak-to-Average Power Ratio (PAPR). Furthermore, the signal or transmission mode must be kept strictly orthogonal so that it can be correctly demodulated by the receiver side, which results in the high sensitivity of frequency offset and phase noise. In addition, we need to solve these problems under various resource multiplexing approaches. Thus in this paper, we first review previous solutions for the next-generation PON and discuss some challenging issues for Mflexi-PONs. Next, some promising solutions are proposed for optimization mechanisms in this new networking paradigm, mainly including system architecture, MAC (Media Access Control) layer allocation of multi-dimensional resources, physical-layer performance improvement, and cross-layer optimization.

Cluster-based protection for survivable fiber-wireless access networks

Survivability is one of the key issues in fiber-wireless (FiWi) access networks, since network component failure may cause huge data loss. Especially in the scenario of segment failure, all optical network units (ONUs) lose their connections with the optical line terminal. Previous works focus on the protection of FiWi against segment failure by deploying backup fibers among the backup ONUs in different segments. However, these works consider only single segment failure. They also ignore the issue of traffic recovery efficiency. More importantly, they underutilize the backup fibers and ignore optimizing the selection of backup ONUs. Therefore, these works usually require a considerable deployment cost for backup fibers. To conquer these challenges, we propose an efficient protection scheme called cluster-based protection (CBP) in this paper. In CBP, we first partition the segments in the network into clusters in order to reduce the overhead for the management of traffic recovery. Then, we select one of the ONUs in each segment as the backup ONU and deploy backup fibers among the backup ONUs of different segments in the same cluster. Under a constraint on the maximum number of segments in each cluster, the CBP scheme aims to protect FiWi against the simultaneous failures of multiple segments with a minimum deployment cost for backup fibers. We propose both an integer-linear-programming-based approach and a heuristic approach to solve the joint optimization problem in CBP of selecting backup ONUs, clustering segments, and deploying backup fibers. To the best of our knowledge, this paper is the first work regarding the survivability of FiWi against the simultaneous failures of multiple segments. Through extensive simulation, we demonstrate that our CBP scheme outperforms the previous works significantly in terms of survivability enhancement and cost savings.