グエン キエン

The specifications of OpenFlow switch define the fast failover(1) group table, which allows the switch quickly adapt to failure based on its local states. In a failure scenario, the traffic on a faulty path is fast switched to a predetermined path in a so-called active/standby manner. That is likely the standard and only way of achieving fast switchover. In this paper, we introduce a novel method for fast switchover/failover on the OpenFlow switch. The conceptual idea is that the new switchover follows an active/active manner rather than the active/standby one in the standard. Specifically, instead of using the fast failover group table we leverage the another one named select. By doing so, the traffic of an end-to-end connection is striped over multiple active paths on a weighted round robin (WRR) fashion. Whenever a failure happens across the paths, the data flow on a failed path is rapidly switched over to the remaining active paths. We modify an OpenFlow 1.3 switch to include the proposed switchover and evaluate the new switch under failure scenarios. The new OpenFlow switch has a better resource allocation as well as less traffic lost during the recovery process than the standard one.

Energy harvesting technology potentially solves the problem of energy efficiency, which is the biggest challenge in wireless sensor networks. The capability of harvesting energy from surrounding environment enables an achievement of infinitive lifetime at a sensor node. The technology promisingly changes the fundamental principle of communication protocols in wireless sensor networks. Instead of saving energy as much as possible, the protocols should keep the efficient operation and maximum performance of networks while guaranteeing the harvested energy is equal or bigger than the consumed energy. In this paper, we propose ERI-MAC a new receiver-initiated MAC protocol for energy harvesting sensor networks. ERI-MAC leverages the benefit of receiver-initiated and packet concatenation to achieve good performance both in latency and energy efficiency. Moreover, ERI-MAC employs a queuing mechanism to adjust the operation of a sensor node following the energy harvesting rate from the surrounding environment. The extensive simulation results in ns-2 show that ERI-MAC achieves good network performance, as well as, enables infinitive lifetime of sensor networks. © Springer-Verlag Berlin Heidelberg 2014.

Linux bonding is a feature allowing to group multiple physical network interfaces into a logical one on Linux machines. Known as a low-cost method to improve fault tolerance and network throughput, the Linux bonding with seven supported modes is increasingly deployed in various scenarios such as datacenters, home networks, etc. However, the strengths and weaknesses of different modes have not been well investigated. While previous works mostly pay attention on the performance of the popular round-robin mode, this work extensively and additionally evaluates other modes based on three major criteria: throughput improvement, load balancing, and fault tolerance. To the best of our knowledge, this is the first work investigating the capabilities of fault tolerance using Linux bonding. The evaluation results show that the activebackup mode achieves the flow switch-over time, which is the duration of traffic flow discontinuation due to a network failure, as small as 10 milliseconds. Moreover, in the round-robin mode with two bonded network interfaces, Linux machines can achieve the maximum throughput close to double of that in case of non-bonding. However, the out-of-order and switch compatibility issues may limit the utilisation of the round-robin mode in certain scenarios. In the 802.3ad mode, the out-of-order issue can be avoided, although load balancing is not always optimal.

In this paper, we extensively explore the operation of Link Aggregation (LA) on OpenFlow switches in comparison to the LA in conventional switches. The comparison of two LA implementations has been conducted in a real testbed under the UDP and TCP traffic loads. The testbed includes Pica8 P-3925 switches, which support two modes: an OpenFlow switch (i.e., using Open vSwitch) and a conventional switch (i.e., using the operating system called XorPlus). The evaluation results show that two LA implementations achieve similar performance in improving throughput. However, the XorPlus implementation provides a better resilience than the other. Specifically, the LA implementation on Xor- Plus spends less than 1.49538 seconds to switch the TCP traffic on the faulty link to the other links of a Link Aggregation Group (LAG) while the switchover time is four times longer on the Open vSwitch. In the case of UDP traffic, the maximum switchover time on the Open vSwitch is twice the one on XorPlus.

This paper introduces a cross-layer implementation, named WiPoMu, that aims to improve efficiency and resilience of the traditional WiFi model. WiPoMu is developed based on three key technologies: Wireless virtualization, Policy routing, and Multipath TCP (MPTCP). Specifically, WiPoMu adopts the wireless virtualization to create multiple virtual WiFi interfaces on a single physical one, hence WiPoMu enables concurrent connections with different access points (APs). In addition, WiPoMu leverages the policy routing and MPTCP in order to efficiently direct and schedule traffic flows over multiple virtual interfaces. On the other hand, WiPoMu is transparent to users since it requires no modification in the physical and application layers. We have conducted evaluations to validate the efficiency of WiPoMu on an indoor testbed and a real home network. The evaluation results show that a WiFi client equipped WiPoMu is able to establish multiple active paths to Internet across different APs. Besides that, WiPoMus is resilient to path failure by achieving seamless hand over between the active paths. Furthermore, WiPoMu improves up to 300% aggregated throughput comparing to the traditional WiFi model using TCP in the testbed.

In wireless sensor networks (WSNs), the multi-hop MAC protocol combines duty cycling radio and forwarding packet via multiple hops (i.e., multi-hop forwarding) to achieve a good balance between energy and latency efficiency. The multi-hop MAC protocol has been proven to outperform other traditional duty cycling protocols, which allow forwarding a packet at most one hop in an operational cycle. Among state-of-the-art multi-hop protocols, MAC 2, which additionally leverages packet concatenation technique, achieves the best performance in normal environments (i.e., without considering noisy). The concatenation technique, which lets several small packets be concatenated into a bigger one before sending out at a node, effectively saves control overhead and shortens delivery latency in WSNs. However, the packet concatenation may downgrade the performance of the network in the noisy environment since the cost for retransmission is high. In this paper, we investigate the negative effect of the noisy level of environment on the efficiency of MAC 2. We observe that in the noisy environment, MAC 2 still keeps the reasonable performances, which are better than the one without concatenation (i.e., the demand-wake up MAC DW-MAC). The simulation results using ns-2 confirm our observations.

In recent years, there has been an increasing interest in Software Defined Networking (SDN)/OpenFlow, which is a novel network architecture splitting control and data planes. The SDN-enabled products have also been widely deployed in many production networks aiming to bypass the limitations of current Internet architecture. There are several efforts on building Software Defined Routing Platform (SDRP), which integrates SDN/OpenFlow within IP routing. However, to the best of our knowledge, there are a few successful works on realizing SDRP despite of the flexibility and rich features of SDN. In this paper, we investigate state-of-the-art works towards SDRP. We have evaluated the performance of a mainstream SDRP named RouteFlow. Our evaluations have been conducted using both the emulator Mininet and a real testbed with physical switches. Different to other works, we focus on the resilience aspect of the routing platform. The evaluation results show that RouteFlow is resilient against both single and multiple link failures. Additionally, OSPF provided by RouteFlow achieves a shorter failover time than the legacy OSPF. In the case of RIPv2, RouteFlow's protocol has a comparable performance value with the distributed one.

Energy harvesting technology potentially solves the problem of energy efficiency, which is the biggest challenge in wireless sensor networks. The sensor node, which has a capability of harvesting energy from the surrounding environment, is able to achieve infinitive lifetime. The technology promisingly changes the fundamental principle of communication protocols in wireless sensor networks. Instead of saving energy as much as possible, the protocols should guarantee that the harvested energy is equal to or bigger than the consumed energy. At the same time, the protocols are designed to have the efficient operation and maximum network performance. In this paper, we propose ERI-MAC, a new receiver-initiated MAC protocol for energy harvesting sensor networks. ERI-MAC leverages the benefit of receiver-initiated and packet concatenation to achieve good performance both in latency and in energy efficiency. Moreover, ERI-MAC employs a queueing mechanism to adjust the operation of a sensor node following the energy harvesting rate from the surrounding environment. We have extensively evaluated ERI-MAC in a large scale network with a realistic traffic model using the network simulator ns-2. The simulation results show that ERI-MAC achieves good network performance, as well as enabling infinitive lifetime of sensor networks.

The Minimum Latency Problem (MLP) is a class of combinational optimization problems that has many practical applications. In the general case, the MLP is proved to be NPhard. One of the approaches to solve the problem is using exact algorithms. However, the algorithms which were recently proposed are applied only to the problems with small size, i.e., 26 vertices. In this paper, we present a new exact algorithm to solve the MLPs with a larger size. Our algorithm is based on the branch and bound method and it has two new rules that improve the pruning technique. We have evaluated the algorithm on several data sets. The results show that the problems up to 40 vertices can be solved exactly. © 2013 National Institute of Informatics.

Person Re-Identification problem aims at matching people across a network of non-overlapping cameras. When multiple probe people appear concurrently, human could compare them together to give a more accurate matching. However, existing approaches treat each probe person independently, skipping the concurrent information. In this paper, we propose a re-ranking method which utilize that kind of information to refine ranked lists produced by any person re-identification method to create more precise ranked lists. The experimental results on VIPeR dataset show the improved performance when our method is applied.

Software-Defined Networking (SDN)/OpenFlow is recently recognized as an emerging technology for network virtualization and programmability. In an OpenFlow network, forwarding (data) planes (i.e., OpenFlow switches) are distributed and remotely controlled by a so-called network controller, which centralizes the network's control functions. The centralized control model of SDN has been proven to be applicable in various deployments ranging from datacenter, campus networks to wide area networks (WANs). In the OpenFlow-based networks, the communication channel between the network controller and switches, which is defined in the scope of OpenFlow specification, plays the most important role. In this paper, we argue that in the state-of-the-art specification establishing one path between a switch and controller potentially decreases the resilience of OpenFlow WANs. Therefore, we propose to use multiple paths for each switch-controller communication and discuss applicable multipath technologies.

Wireless access network is an appropriate approach providing Internet connection in disasters where communication infrastructures might completely be damaged. This paper clarifies essential requirements for a resilient disaster recovery multihop access network (DRAN). Based on those requirements, which could not be satisfied by existing access network technologies, a novel approach, thereby multihop access networks can be established quickly using commodity mobile devices (laptops, tablets, smart phones), has been proposed. Users can connect to the proposed wireless access network as easily as connecting to conventional access points (APs) and unconsciously contribute to the network extension. As a result, the network is extended incrementally providing more Internet access opportunities to far apart nodes. Experimental evaluations reveal the feasibility as well as the scalability of the proposed approach.

It is increasingly popular that there are multiple accessible access points (APs) surrounding a WiFi client. In an ideal case, the client would simultaneously connects to all the APs and maximize the connections' utilization, for example, aggregating the bandwidth of APs' backhauls, load balancing among the connections, etc. In this paper, we present WM (Wireless virtualization with Multipath TCP) a cross-layer approach that aims to improve performance of mobile WiFi users. The demonstration shows that a WiFi client equipped WM can keep multiple concurrent connections to APs by using wireless virtualization. Moreover, WM enhances the aggregated bandwidth and achieves seamless handover by adopting Multipath TCP.

Disasters may destroy everything including communication infrastructures isolating people in the disaster-stricken areas. Recovery of these infrastructures is often prolonged which is not suitable for disastrous fast-responses. This paper proposes a wireless multihop access network virtualization (WMANV) approach thereby users/victims are provided Internet access transparently. Concretely, users are provided Internet connection via multihop wireless access networks as if they are connecting to conventional access points (APs) using their commodity mobile devices. A tree-based architecture was proposed to realize the concept of WMANV. This approach allows participating mobile nodes (MNs) to contribute on growing up access networks, thus provide the Internet connection means to further disconnected MNs. The feasibility and the scalability of the proposed approach have been verified by real deployments and experimental evaluations. © 2013 IEEE.

Person Re-Identification is the problem of matching people across a network of non-overlapping cameras. One of the challenges is how to match body parts to body parts for comparison between images of two people in the context of different viewpoints as well as deformable human bodies. Existing approaches usually use fixed models to localize body parts or detect human shapes to extract body parts from the shapes. Therefore, it is difficult to change to a new model or structure of body parts. Moreover, those approaches could not deal with multiple human poses simultaneously. We propose a machine learning-based method to extract body parts that is based on Deformable Part Models (DPM). DPM is easy to train and has robust performance. In addition, with DPM, we could use multiple models for multiple human poses concurrently. Experiments on standard dataset ETHZ1 show that the proposed method outperforms state of the art methods. © Springer-Verlag 2013.

The current technology of Internet provides an acceptable level of resilience in normal operation. However, the Internet may get catastrophic impacts when unexpected disasters such as earthquakes, tsunami, etc. happen. Specifically, the Internet infrastructure including Wide Area Networks (WANs) face many challenges to normal operation including power outage, link, device failures, rerouting packets, traffic engineering, etc. In this paper, we follow a systematic approach for realizing disaster-resilient WANs using Software-Defined Networking (SDN) technology. SDN enables the network control plane to be decoupled from the network forwarding hardware, and moves the control plane to a programmable component, i.e., the network controller. The network management and operation therefore increase flexibility. To confirm the feasibility of SDN-based resilient network towards fast disaster recovery, we have constructed two evaluations under the real large scale network topology. One is to investigate the latencies between controllers and switches in order to find the appropriate number and locations of controllers. Another one is to simulate a reactive switch-over from a faulty link to an alternative link, assuming a realistic scenario. The results show that the switch-over time depends on the latencies between networking devices and a controller. Additionally, even those latency values equal the worst-case latency, the fast rerouting of TCP traffic is achievable. © 2013 IEEE.

Fast recovering backbone networks from failures caused by unexpected disasters is critical to enhance the Internet resilience. The major challenge is fast switchover traffic from a faulty path to an alternative one. In this paper, we present an approach toward achieving a zero switchover time (i.e., optimal) by using two emerging technologies: Multipath TCP (MPTCP) and Software-Defined Networking (SDN). The conceptual idea is keeping multiple concurrent paths in an end-to-end communication, which is continuously alive unless all the paths get failure. Besides that, when a disaster occurs, traffic engineering is adopted to quickly balance the traffic on available paths. In order to establishes the concurrent paths, MPTCP divides an application's byte stream into multiple TCP subflows, each of which is forwarded in an end-to-end route. The routes are normally determined by IP routing protocols, such as OSPF or BGP, which have two disadvantages. The first one is the protocols always take long time to reestablish new routes in disasters (i.e., long convergence time). The second one is they are originally designed without traffic engineering. On the other hand, SDN enables the use of central controllers with global view of dynamic network state. In addition, the controllers can remotely controls both direction and capacity of traffic flows in many flexible ways (i.e., using OpenFlow). Therefore, SDN potentially solves the disadvantages of IP routing protocols.

Wireless access network is an appropriate solution to provide Internet connection to users in disasters where communication infrastructures might completely be damaged. However, existing researches commonly focus on separate issues, given strong requisite assumptions. Basically, those assumptions could not be satisfied in harsh environments like in disasters. In addition, a combination of those separate research aspects into an integrated system has not been discussed. This paper thoroughly analyzes current states of existing wireless access network technologies by which suitable solutions for resilient disaster recovery access networks (DRANs) are proposed. Moreover, a novel wireless multihop access network virtualization (WMANV) approach to resilient DRANs is proposed. The feasibility of the proposed approach is verified using experimental evaluations.

Information and communication technology (ICT) is indispensable for today's society and business environment. The continuous usage of ICT is an essential requirement of the society, especially during a disaster. Besides that, resilient ICT is desirable not only for almost any business environment but also for business continuity planning of such businesses in order to make their internal ICT resilient. We first discuss the requirements of resilient ICT based on surveys of the latest technologies that aim to improving resiliency, such as virtualization of servers and networks. We then argue that it is necessary to consider two levels of ICT resiliency. Additionally, we propose how to select evaluation means of the technologies for facilitating resiliency, and operational frameworks for achieving resiliency. © 2013 IEEE.

We propose the low overhead media access control protocol (LO-MAC), a new low latency, energy efficient MAC protocol for low data rate wireless sensor networks. LO-MAC uses both duty cycling and multihop forwarding from the routing-enhanced MAC protocol (RMAC) to reduce idle listening and sleep latency, respectively. Besides that, LO-MAC introduces a traffic-adaptive mechanism, which is based on the fact that a node can sense a busy channel within its carrier sensing range. This mechanism uses carrier sensing as a binary signal, and effectively notifies nodes of the existence of a data packet. The nodes then either keep their radios on to take part in multihop data forwarding or turn them off to save energy. Moreover, LO-MAC takes full advantage of the broadcast nature of wireless communication and lets a packet have different meanings when it is in transmission range of different nodes. In LO-MAC, not only is a request-to-send/clear-to-send pair replaced with a Pioneer (PION) packet, as in RMAC, but also a data packet can play both data and acknowledgement functions. Therefore, control overhead and overhearing energy are significantly reduced. Our simulation results show that LO-MAC outperforms RMAC in terms of energy efficiency while achieving comparable end-to-end latency. © 2013 Kien Nguyen et al.

Wireless sensor network MAC protocols switch radios,off periodically, employing the so-called duty cycle mechanism, in order to conserve battery power that would otherwise be wasted by energy-costly idle listening. In order to minimize the various negative side-effects of the original scheme, especially on latency and throughput, various improvements have been proposed. In this paper, we introduce a new MAC protocol called MAC(2) (Multi-hop Adaptive with packet Concatenation-MAC) which combines three promising techniques into one protocol. Firstly, the idea to forward packets over multiple hops within one operational cycle as initially introduced in RMAC. Secondly, an adaptive method that adjusts the listening period according to traffic load minimizing idle listening. Thirdly, a packet concatenation scheme that not only increases throughput but also reduces power consumption that would otherwise be incurred by additional control packets. Furthermore, MAC2 incorporates the idea of scheduling data transmissions with minimum latency, thereby performing packet concatenation together with the multi-hop transmission mechanism in a most efficient way. We evaluated MAC2 using the prominent network simulator ns-2 and the results show that our protocol can outperform DW-MAC-a state of the art protocol both in terms of energy efficiency and throughput.

In wireless sensor networks (WSNs), MAC protocols utilize duty cycling to extend the network lifetime. Receiver-Initiated MAC (RI-MAC) is an asynchronous duty cycling protocol, which schedules transmissions based on receivers. By letting a receiver initiate the rendezvous for the transmissions between senders and the receiver, the protocol achieves good latency performance at the cost of high energy consumption at the senders. On the other hand, different types of traffic are very popular in WSNs. The MAC protocols are necessarily to be equipped QoS (quality of service) mechanisms to handle with the variation of traffic. In this paper, we propose AQ-MAC a new asynchronous MAC protocol with QoS awareness. AQ-MAC adopts the receiver-initiated manner and provides QoS service per packet following the priority imposed. When a node has an incoming packet with a high priority, it immediately turns on the radio in order to wait for a transmission initiated by a receiver. Otherwise, the node keeps a low priority packet in a queue and sends out in a burst until a high priority packet comes or after a timeout value. In addition, AQ-MAC utilizes a packet concatenation scheme to improve the energy efficiency by reducing control overhead. We have evaluated AQ-MAC in multiple scenarios using ns-2. The results show that AQ-MAC adapts well with different types of traffic as well as achieves good performance in terms of energy efficiency and latency. © 2012 IEEE.

Achieving low latency and high energy efficiency are critical requirements in many of sensor network applications. Combining the duty cycling mechanism and multi-hop transmission is one of the solutions to this problem. Multi-hop protocols such as demand wakeup MAC (DW-MAC) schedule data transmissions during the sleep period. In so doing, a data packet can traverse multiple hops within a single operational cycle. By introducing a scheduling parameter, called proportional mapping function, DW-MAC can provide better performance in terms of both latency and energy efficiency, while avoiding data collision. In this paper, we derive the minimum value Rmin of the scheduling function by numerically analyzing the delivery latency. Using that value, a DW-MAC-like protocol not only guarantees no collisions at the intended receivers but also achieves the minimum latency in the multi-hop scenario. In addition, we also show the upper and lower bounds of the delay. We evaluated the accuracy of our findings by running ns-2 simulations in multiple scenarios. © 2011 IEEE.

The MAC protocol in wireless sensor networks (WSNs) plays an important role in conserving energy and it generally adopts a duty cycling mechanism to eliminate idle listing energy at the cost of high delivery latency. The most effective method of bypassing the latency disadvantage is forwarding a packet over multiple hops in an operational cycle. The combination of duty cycling and multi-hop forwarding creates a new class of MAC protocol called multi-hop MAC, which has been proven to outperform other single-hop duty cycling protocols both in term of energy efficiency and latency. A packet in such a protocol is always relayed via multiple hops toward a sink in a cycle. The flow from a source node to the sink is affected by contention because of the sharing characteristic of wireless channels. We investigated what effect contention had on the flow capacity of Demand Wakeup MAC (DW-MAC), which is a state-of-the-art multi-hop MAC protocol. We also found that instead of using the original large contention window (CW), a DW-MAC's flow using a smaller CW could avoid unnecessary contention and use channels more efficiently. The analysis and resulting ns-2 simulation revealed that DW-MAC with the new values for CW achieves higher throughput, lower end-to-end latency, and greater energy efficiency. © 2011 ACM.

In wireless sensor networks, most MAC protocols switch radios off periodically, employing the so-called duty cycle mechanism, in order to conserve battery power that would otherwise be wasted by energy-costly idle listening. In order to minimize the various negative side-effects of the original scheme, especially on latency and throughput various improvements have been proposed. The novel Aggregation MAC or AG-MAC protocol presented in this paper combines two of the most promising improvements into one protocol. Firstly, the idea to forward packets over multiple hops within one duty cycle as initially introduced in RMAC. Secondly, a packet aggregation scheme that not only increases throughput but also reduces power consumption that would otherwise be incurred by additional control packets. Since data packets in sensor networks are comparatively small significant gains are to be expected. Furthermore, AG-MAC incorporates our previously presented idea of scheduling data transmissions with minimum latency, thereby performing packet aggregation together with the multi-hop transmission mechanism in a most efficient way. We evaluated AG-MAC using the prominent network simulator ns-2 and the results show that our protocol can outperform state of the art protocols both in terms of energy efficiency and throughput.

Energy efficiency is a requirement when designing a MAC protocol for wireless sensor networks. To reduce needless energy use, most MAC protocols exploit the duty cycling technique, in which the radio frequently turns on and off in each operational cycle. Among those protocols, the multi-hop MAC protocol routing enhanced MAC (RMAC) enables multi-hop transmission to minimize the latency burden, which is the main disadvantage of duty cycling. In each cycle, when nodes are awake, RMAC can exploit the cross-layer information to initialize the multi-hop flow and the data packet transmission is scheduled in the subsequent sleep period. This technique, however, introduces the long listening period problem in which nodes have to keep the radio on even when no flow data is scheduled. We propose a solution to solve that problem by adding a short period after the synchronized process. In this period, we use carrier sensing as a binary signal, which lets the nodes know the traffic status of the network. After the period, nodes go to the sleep state when no data exists in the network otherwise nodes operate similarly to the RMAC's basic scheme. Simulations showed that our solution can improve energy efficiency while only slightly increasing latency. ©2010 IEEE.

Sensor network MAC protocols use a duty cycling mechanism and a multi-hop transmission to create a good trade-off between their energy efficiency and latency. The nodes in duty cycling multi-hop MAC protocols such as RMAC periodically sleep/listen during the operational cycle to reduce their energy consumption by idle listening and the listening period is usually long in order to support packets in reaching a multi-hop destination in a single cycle (multi-hop transmission). The traffic network in the wireless sensor network is otherwise very scarce, and keeping the radio on during such long listening periods when there is no data transmission in the network wastes energy. In addition, these protocols incur a large amount of control overhead, which is one of the main energy wasted sources. This paper proposes an adaptive low overhead MAC protocol we call AM- MAC (Adaptive Multi-hop MAC). It can reduce the wastage caused by long listening period and minimizes the control overhead. The nodes in AM-MAC use an adaptive method, and thus they can adjust the duration of the listening period according to the traffic load. To reduce the control overhead, a single packet has more than one role. During a multi-hop transmission, the new control packet replaces the RTS/CTS pair, and one DATA packet can play both DATA/ACK roles. An extended evaluation of AM-MAC has been conducted through simulation, against RMAC. The results illustrate that AM-MAC significantly reduces the energy consumption and notably lessens the end-to-end latency Copyright © 2010 ACM.

By making the best use of limited bandwidth, quality of service (QoS) provisioning over internet is essential for satisfying various types of internet-application requirements. The traffic classification and scheduling are the key functions to provide various kinds of class of service (CoS) under an overload condition. This paper investigates QoS performance in a network equipment testbed when implementing these main functions. We examine the major CoS functions provided by the juniper T320 router, and measure their performance. In addition to fundamental analysis of the QoS behavior, we show the impact of QoS operations on a parallel system distributed in multi-domain networks as a practical case study of grid environments. (C) 2008 Elsevier B.V. All rights reserved.

In wireless sensor networks (WSNs), a duty-cycling scheme is typically applied to the medium access control (MAC) protocol to reduce energy consumption due to idle listening. However, this scheme introduces huge end-to-end latency and still suffers from a large control packet overhead. We propose a new MAC protocol with low latency and low control overhead for WSNs use (the LCO-MAC). In our protocol, a DATA packet can be transmitted through multiple hops in a single duty cycle to shorten end-to-end latency. To reduce energy consumption caused by control packet overhead, we force one packet to play more than one role. In the initial transmission period, a control packet acts as an RTS (request to send) for a downstream node and a CTS (clear to send) for an upstream node. In the actual data transmission period, a DATA packet keeps its original DATA role for the downstream node, but also plays an ACK (acknowledgment) role for the upstream node. Our simulation using ns-2 has shown that LCO-MAC enables a notable improvement in energy efficiency and decreases end-to-end latency compared to those of RMAC. © 2008 IEEE.

Quality of Service (QoS) in Layer 3 is essential for satisfying various types of Internet-application requirements by making the best use of limited bandwidth. A large number of such applications still use TCP/IP in order to connect various types of computer nodes. This paper investigates QoS performance in a network equipment testbed. We examine the major Class of Service (CoS) functions provided by the Juniper T320 router, and measure their performance. In addition to fundamental analysis of the QoS behavior, we show the impact of QoS operations on a parallel system distributed in multi-domain networks as a practical case study of grid environments.