Showing papers by "Celimuge Wu published in 2014"

A MAC protocol for delay-sensitive VANET applications with self-learning contention scheme.

Abstract: Packet delivery ratio and end-to-end delay are the two most important metrics for vehicular ad hoc network applications. In this paper, we propose a MAC layer protocol which can provide a high packet delivery ratio, low end-to-end delay, and high fairness for various scenarios. The proposed protocol uses a Q-Learning algorithm to adjust the contention window size in order to provide an efficient channel access scheme for various network situations. The simulation results demonstrate the advantage of the proposed protocol over other alternatives.

Multi-Hop Broadcasting in VANETs Integrating Intra-Flow and Inter-Flow Network Coding

Abstract: Multi-hop data dissemination in vehicular ad hoc networks (VANETs) is very important for the realization of collision avoidance systems and other many interesting applications. However, designing an efficient data dissemination protocol in VANETs has been a challenging issue due to vehicle movements, limited wireless resources and lossy characteristics of wireless communication. In this paper, we propose a protocol which employs intra-flow and inter-flow network coding to reduce the protocol overhead as compared to traditional protocols. The protocol also can improve the packet reception probability at the receiver nodes by using the network coding. Therefore, the protocol can provide a lightweight and reliable solution for data dissemination in VANETs. We use theoretical analysis and computer simulations to show the advantage of the proposed protocol over other existing alternatives.

Joint MAC and Network Layer Control for VANET Broadcast Communications Considering End-to-End Latency

Abstract: In vehicular ad hoc networks (VANETs), multi-hop broadcast communications are required for many applications including driver assistance systems. However, providing a low end-to-end latency has been very challenging. In this paper, we propose a joint MAC network layer multi-hop broadcast protocol. For the network layer, the proposed protocol reduces the number of sender nodes by using common forwarder nodes for different traffic flows. At the MAC layer, the proposed protocol uses a Q-Learning algorithm to adjust the contention window size. By interacting with the environment, the protocol can find the best contention window size to transmit data packets and therefore can provide a high packet dissemination ratio and low end-to-end delay for various scenarios. The simulation results demonstrate the advantage of the proposed protocol over other existing alternatives.

A QoS supporting ad hoc network protocol combing admission based TDMA and 802.11 DCF

Abstract: We propose a QoS supporting ad hoc network protocol which employs a hybrid approach utilizing both TDMA and 802.11 DCF. TDMA period provides contention free transmissions for QoS flows, and DCF period is used to provide contention-based access for best effort or low priority flows. TDMA time slots are assigned on admission basis in a distributed manner by utilizing network routing information and considering current assignment situation. In DCF period, frames are transmitted not to disturb the TDMA period. By combing TDMA and 802.11 DCF, the proposed ad hoc network is able to provide an efficient solution for QoS control in ad hoc networks. We use QualNet simulator to evaluate the proposed scheme.

Toward a Totally Distributed Flat Location Service for Vehicular Ad Hoc Networks

Abstract: Many routing protocols in vehicular ad hoc networks (VANETs) utilize location information to find a route to the destination. However, tracking the location information of other nodes is very challenging in highly dynamic VANETs. We propose a location service which can provide location information with low overhead and low delay. The location service periodically disseminates the location of each vehicle to 3-hop distance for every second with very low overhead. By eliminating location errors by taking account of the velocity of a vehicle, the proposed protocol can provide accurate position information. The location service also provides a lightweight location query mechanism for longer distance destination nodes. We show the effectiveness of the protocol by using theoretical analysis and computer simulations.

Making inter-flow network coding possible for unicast routing in VANETs

Abstract: Designing an efficient unicast routing protocol for vehicular ad hoc networks has been a challenging issue due to the vehicle movement, limited wireless resources and lossy feature of wireless channel. We propose a protocol which utilizes inter-flow network coding to route data messages. The proposed protocol uses common backbone vehicles for different traffic flows. A reliability connected backbone is selected by considering vehicle movement dynamics and link quality between vehicles. By using inter-flow network coding at the backbone vehicles, the protocol can reduce the number of generated packets by 25% in most cases as compared to the traditional routing approach. As a result, the proposed protocol can provide a high packet delivery ratio, low overhead and low delay. We show the effectiveness of the protocol using theoretical analysis and computer simulations.

Analyzing the drawbacks of node-based delays in Tor

Abstract: Tor Network is a low-latency anonymity and censorship resistance network, which relaying traffics through onion routers before being forwarded to the destination. The distribution of trust offered by Tor creates resilience in the face of compromise and censorship. However, there are performance and usability issues in the Tor network. In this paper, we analyze the delay contributions from the host TCP stack of the Tor relay mode. We measured and analyzed the transfer data passing through the Tor nodes to evaluate the main causes of delays. The experimental analysis show that increasing delays on nodes are affected by the heavy TCP packets queuing, kernel memory exhaustion and TCP advertised window not able to accept more data. We present our experimental analysis and measurement results as a prerequisite for addressing the Tor's end-to-end increasing latency and poor usability.

Greenly offloading traffic in stochastic heterogeneous cellular networks

Abstract: This paper puts forwards an on-line reinforcement learning framework for the problem of traffic offloading in a stochastic Markovian heterogeneous cellular network (HCN), where the time-varying traffic demand of mobile terminals (MTs) can be offloaded from macrocells to small-cells. Our aim is to minimize the average energy consumption of the HCN while maintaining the Quality-of-Service (QoS) experienced by MTs. For each cell (i.e., a macrocell or a small-cell), the energy consumption is determined by its system load which is coupled with the system loads served in other cells due to the sharing over a common frequency band. We model the energy-aware traffic offloading in such HCNs as a constrained Markov decision process (C-MDP). The statistics of the C-MDP depends on a selected traffic offloading strategy and thus, the actions performed by a network controller have a long-term impact on the network state evolution. Based on the traffic demand observations and the traffic offloading operations, the controller gradually optimizes the strategy with no prior knowledge of the process statistics. Numerical experiments are conducted to show the effectiveness of the proposed learning framework in balancing the tradeoff between energy saving and QoS satisfaction.

Providing fast broadcasting by reserving time slots for multi-hop distance in VANETs

Abstract: In vehicular ad hoc networks (VANETs), multi-hop broadcast protocols are required for many applications including driver assistance systems. However, providing a low end-to-end latency has been very challenging. In this paper, we propose a protocol which can provide a fast channel access for broadcast traffic flows in VANETs. The protocol introduces new control messages to reserve time slots for the forwarder nodes and eliminate the hidden terminal problem. By providing a contention-free multi-hop forwarding scheme based on reliably generated forwarding backbone, the protocol can provide low end-to-end delay and high packet dissemination ratio. The theoretical analysis and simulation results demonstrate the advantage of the proposed protocol over other existing alternatives.

Experimental Analysis of TCP Behaviors against Bursty Packet Losses Caused by Transmission Interruption

Abstract: Although TCP was originally designed to provide the reliable data transfer over the Internet, packet losses detected in TCP are considered as an indication of network congestion due to the high quality of data transmission provided by recent transmission technologies and media access control technologies. However, packet losses can be caused by transmission interruptions such as handoffs in the mobile networks and protection switching in the transport networks. These packet losses are bursty because the transmission interruptions continue for tens of miliseconds through several seconds. In this paper, we describe the experimental analysis of TCP behaviors by inserting errors such that all packets are lost during transmission interruptions. We have tested various TCP versions including those in Linux, that in Windows and that in Mac OS. This paper suggests (1) that the tested TCPs in Linux follow the similar procedure and retransmit lost packets quickly, (2) that TCP in Windows also behaves well but the increase of congestion window seems to be limited, and (3) that TCP in Mac OS has shown some problems in retransmitting contiguously lost packets. Keywords-TCP; Transmission Interruption; Bursty Packet Losses; Retransmission; SACK Based Loss Recovery.

A Circuit Switching Method for Improving Congestion of Tor Network

Abstract: The Tor Network is a low-latency anonymity and censorship resistance network, which relaying traffics through series of onion routers (OR) before being forwarded to the destination. Tor is vulnerable to network congestion and performance problems, due to bulk and light traffics are competing to transfer data on a single TCP connection. A large fraction of the available network capacity is consumed by bulk users' traffic, resulting in higher buffer occupancy and delays for interactive light users. To address the problems, we propose a method of switching the bulk traffic that contributes to increasing delays to another path, which built preemptively at the beginning. This approach allows the light interactive circuits to utilize the bandwidth resources well and, easily writes their cells from the circuit queue to the output buffer and outgoing TCP socket. Our approach of switching bulk circuits in the overloaded ORs comes to effect if the Tor output queues increases beyond the threshold value, and when sending TCP socket is congested and un-writable. Our goal is to reduce the congestion in the ORs and improve the throughput from the incoming traffic (upstream), and outgoing traffic (downstream). Through measurements and evaluation conducted in the live Tor network, we observed that the throughput of the light interactive traffic is improved. The light circuit traffics are no longer forced to wait behind the bulk traffic when transfer their data to the outgoing TCP socket.

RESOLVING BUFFERBLOAT PROBLEM IN 802.11n WLAN BY WEAKENING MAC LOSS RECOVERY FOR TCP STREAM

Abstract: IEEE 802.11n wireless LANs provide high-speed data transfer. If the radio condition degrades, however, the transfer rate will be reduced significantly and there may be some problems in the data transfer. In order to clarify this problem, we have evaluated the TCP performance over IEEE 802.11n LAN by changing the distance between the access point and a terminal. As a result, a severe increase of round trip time has been measured in the case of low data rate. This is a sort of bufferbloat problem, which is being actively studied in recent years. The detailed analysis of this experiment is our first contribution. We have inferred that one of the reasons for this increased delay is the powerful retransmission function of 802.11n. So, we propose, as the second contribution, a method to improve TCP delay performance by reducing this retransmission capability and by generating TCP segment loss intentionally. This paper describes the performance evaluations for our proposal in comparison with CoDel, which is an active queue management approach, and with the native 802.11n.

Multi-hop broadcasting in WAVE multi-channel vehicular networks: Single transceiver case

Abstract: Multi-hop data dissemination in vehicular ad hoc networks (VANETs) has been very challenging due to the vehicle movement, various vehicle densities and the lossy characteristics of a wireless channel. The problem of how to provide an efficient broadcasting of time-critical broadcast messages under multi-channel operation of IEEE WAVE standard has not been discussed sufficiently. In this paper, the authors discuss the multi-hop broadcast problem in multi-channel environment defined by IEEE WAVE, and propose a multi-hop broadcast protocol which conducts efficient channel utilization with single transceiver. They use computer simulations to show the advantage of the proposed protocol over other existing alternatives.