Tülin Atmaca

B-Planner: Night bus route planning using large-scale taxi GPS traces

Taxi GPS traces provide us with rich information about the human mobility pattern in modern cities. Instead of designing the bus route based on inaccurate human survey regarding peoples mobility pattern, we intend to address the night-bus route planning issue by leveraging taxi GPS traces. In this paper, we propose a two-phase approach based on the crowd-sourced GPS data for night-bus route planning. In the first phase, we develop a process to cluster “hot” areas with dense passenger pick-up/drop-off, and then propose effective methods to split big “hot” areas into clusters and identify a location in each cluster as a candidate bus stop. In the second phase, given the bus route origin, destination, candidate bus stops as well as bus operation time constraints, we derive several effective rules to build bus routing graph and prune the invalid stops and edges iteratively. We further develop two heuristic algorithms to automatically generate candidate bus routes, and finally we select the best route which expects the maximum number of passengers under the given conditions. To validate the effectiveness of the proposed approach, extensive empirical studies are performed on a real-world taxi GPS data set which contains more than 1.57 million passenger delivery trips, generated by 7,600 taxis for a month in Hangzhou, China.

Packet Filling Optimization in Multiservice Slotted Optical Packet Switching MAN Networks

Many aspects about Optical Packet-Switching (OPS) networks have been studied in the last few years and it has been shown that their performance depends on several determinant factors, such as the transmission mode and the optical packet format A synchronous slotted mode is more adapted to MAN network specifications, since it produces higher throughput comparing to the asynchronous mode. In this paper, we propose an optimized mechanism for creation of fixed-size optical payloads filled with variable size electronic client packets. Our algorithm is called GPFO for graduated packet filling optimization. It avoids the need for complex segmentation methods. It also provides high network throughput and good performance results in terms of packet loss ratio and mean access delay. We investigated the performance of our algorithm throw simulation works. Using the ns simulator, our experiments are performed on an optical MAN network with a ring topology. Numerical results have shown that, compared to existing solutions, the GPFO enhances the network throughput and optimizes the global network performance in terms of queuing delay and bit loss ratio.

Performance of OBS techniques under self-similar traffic based on various burst assembly techniques

In this paper, the performance of some well-known Optical Burst Switching (OBS) techniques is analyzed under self-similar traffic using the time threshold based burst assembly and the hybrid burst assembly techniques. The characteristics of the optical bursts formed at ingress nodes by the burst assembly techniques are studied extensively by changing input traffic load and the time threshold value. The performance of the OBS techniques, which are the horizon scheduling technique, two void filling techniques, and the group scheduling technique, are evaluated in terms of burst loss rate and byte loss rate. Although, in the OBS literature, burst loss rate is a commonly employed metric, it is observed that the burst loss rates obtained are different than the corresponding byte loss rates. Such a result is caused by variable burst sizes. It is also seen that the performance ranking using burst loss rate is not the same as the performance ranking using byte loss rate for some cases due to the characteristics of the OBS techniques. Moreover, the traffic characteristics of the optical bursts at ingress and egress nodes are examined in terms of the Hurst parameter values. It is observed that the self-similarity of the incoming traffic is mainly affected by the burst assembly techniques. It is shown that when the TTh based burst assembly is used, the Hurst parameter of the traffic degrades drastically. However, when the hybrid burst assembly is employed, the Hurst parameter of the traffic remains closer to the Hurst parameter of the original self-similar IP traffic.

Performance Study of OBS Reservation Protocols

This paper presents a comparison of reservation protocols used in optical burst switches (OBS) with quality of service (QoS). In terms of performance criteria, byte loss rate, end-to-end delay and access delay are considered. A 2-state MMPP(Markov modulated Poisson process) traffic generator on a 14-node mesh network topology illustrating NSFNET is chosen as the testbed. This topology is simulated using NS2 network simulation platform. Bursts are created using a hybrid model that takes into account both timeout and maximum length methods. In edge and core nodes, there are different QoS mechanisms such as explicit tear down and priority based queuing. Simulation results have shown that JIT (just in time) algorithm is not as successful as JET (just enough time) and horizon when the network is highly loaded. However, JET algorithm produces better performance than other reservation protocols while considering QoS.

Impacts of packet filling in an optical packet switching architecture

Increase in network traffic and emergence of a variety of applications requiring different QoS necessitate scalable and efficient resource utilization. An appealing solution for backbone networks is the optical packet switching architecture. However technological constraints prevent the deployment of such all-optical solutions. Instead, feasible optic-electrical hybrid systems are gaining importance. In this paper, first we briefly describe such an architecture and then focus on the edge part. More specifically, we focus on the optical packet creation process in the edge. The impacts of this optical packet creation mechanism on the incoming traffic profile are investigated. Finally we show, by exploiting service differentiation, that we can improve the optical packet creation mechanisms and use bandwidth more efficiently.

Markovian model of RED mechanism

The IP networks are faced today with difficult task of satisfying the needs of connections requiring different QoS by sharing the same physical resources, e.g. bandwidth and buffers. Buffers are a key component of packet-switched network, as they absorb burst arrivals of packets and hence reduce losses. Larger buffers can absorb larger bursts but they tend to build up high load and increase queueing delays. The traditional technique for managing delay is to set a maximum length for each buffer queue, accept packets in the queue until the maximum length is reached. Distributed computational grids depend on TCP to ensure reliable end-to-end communication between nodes across the wide area network (WAN). In the core of network (routers), it is believed that random early detection (RED) will alleviate problems related to synchronization of flows and also provide some notion of QoS by intelligent dropping. In this paper we develop simple Markovian model for the RED buffer management schemes, and use these models to quantify the benefits brought about by RED. In particular we examine the impact of RED on the loss rate and the mean delay. We analyse the transient and stationary states, using Markovian model. We show that the loss probability is the same for TCP and UDP traffics if RED algorithm is used. This loss rate of a flow going through a RED router does not depend of the burstiness of this flow, but only on the load it generates.

A new dynamic bandwidth allocation algorithm based on online-offline mode for EPON

In parallel with the enormous spread of Internet usage, the tendency of users for high capacity applications (HDTV, Online Games, P2P File Transfer) are rising. This increase in needs of data communication, compel the service providers to act to end users better than before. At this point, Passive Optical Networks (PONs) distinguish in access network solutions due to its capability of carrying high amount of data over longer distances. Although they are still in development, existing electronic based data communication techniques (ADSL, double twisted copper telephony systems like VDSL, cable modem coaxial cable CATV systems) can operate over short distances and have large amount of energy consumption. In this paper, basic schemes of dynamic bandwidth allocation algorithms on EPON have been presented and a novel online-offline based method called hcDBA (Half Cycling Dynamic Bandwidth Allocation) has been introduced. In addition, hcDBA algorithm has been improved with early prediction mechanism (p-hcDBA). By the simulation studies, it has been investigated that hcDBA algorithm performs better than existing mechanism in terms of byte loss ratio and access delays criteria. p-hcDBA also improved the performance in terms of access delay. In this study, we stated that when the bandwidth increases the prediction approached has more impact of the overall performance.

Performance Study of New OBS Channel Scheduling Algorithms in a Multiservice Network

This paper presents a comparative study between JET (Just Enough Time) and a number of new preemptive channel scheduling algorithms in Optical Burst Switching (OBS) with QoS (Quality of Service) requirement. In terms of performance criteria, loss rate in byte, access delay and end-to-end delay are considered. A 2-state MMPP (Markov Modulated Poisson Process) traffic generator on a 14-node mesh network topology illustrating NSFNET is designed as the testbed. NS2 Network Simulation tool is used for our tests. Bursts are created using a hybrid model that takes into account both timeout and maximum length threshold mechanisms. In core and edge nodes, in order to satisfy QoS requirements two mechanisms such as Regulative Wavelength Grouping (RWG) and priority based queuing are used. RWG mechanism, where the number of wavelengths for each traffic class is arranged for adjusting the burst drop probability of traffic classes under a specific threshold value, is used for providing priority levels in core nodes. In priority based queuing, bursts are sent according to their priority order, in edge nodes. Simulation works have shown that our preemptive channel scheduling algorithms give better results compared to the other studied algorithms while considering QoS.

Impact of Fixed-Size Packet Creation Timer and Packet Format on the Performance of Slotted and Unslotted Bus-Based Optical MAN

In a typical slotted bus-based optical metropolitan network, the position of access node on the networkpsilas bus have a large impact on the network performance. We refer to this problem in data transmission as the unfairness property. In fact, the first nodes on the bus have priority over downstream nodes. Another factor which has a significant impact on the network performance is the choice of Timer value, which is used to create the fixed-size optical packet. An unsuitable choice of Timer values may lead to wasted network bandwidth. Indeed, small Timer values generate optical packets with very low filling ratios; while a very big Timer value leads to excessive packet creation delays (and therefore high access delays) when the arrival rate of electronic client packets becomes low. In this paper, we investigate the impact of Timer values on the network performance by simulating the slotted bus-based network that supports multi-service. In this context, we are interested in two study cases: with and without quality of service (QoS) Upgrade mechanism. Additionally, we will show that the slotted bus-based network with QoS Upgrade can improve the network performance, in terms of mean access delay and loss ratio of electronics packets clients, compared with unslotted bus-based network that supports variable-size optical packets.

Dynamic Intelligent MAC Protocol for Metropolitan Optical Packet Switching Ring Networks

This paper presents a new access protocol addressing the problem of inefficient utilization of bandwidth and the fairness issue in metropolitan optical packet switching ring (OPSR) networks. This protocol, called Dynamic Intelligent - Medium Access Control (DI-MAC) protocol, greatly improves the performance of the OPSR network that usually employs the optical Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) protocol. The principle of DI-MAC is to intelligently space out the transmission of local packets at upstream nodes to avoid inefficient bandwidth fragmentation and, as a result, to release more bandwidth for downstream nodes. The key feature of DI-MAC is that it automatically and dynamically adjusts the gap between successive transmission of local packets, according to the change of network state (i.e., bandwidth availability, traffic variation, etc.), in order to favor the transmission of downstream nodes, and at the same time, to satisfy requirements of local clients in terms of quality of service. Our experiments shown that DI-MAC effectively simplifies the network management and optimizes the network performance; thus it promises to be a new cost-efficient access protocol for future metropolitan optical ring networks. In this paper, we explain in detail this protocol and analyze its impact on the network performance.