Conference papers

Experimental Tuning of AIFSN and CWmin Parameters to Prioritize Voice over Data Transmission in 802.11e WLAN Networks

Miroslaw Narbutt et al. — Mon, 24 Jan 2011 05:55:52 PST

In this paper we experimentally study the impact of two EDCA parameters, namely AIFSN and CWmin, on a mixed voice/data wireless transmission. In particular we investigate how the tuning of these parameters affects both the voice transmission quality and background data throughput. We predict end-to-end voice transmission quality from time varying transmission impairments using the latest Appendix to the ITU-T E-model. Our experimental results show that the tuning of the EDCA parameters can be used to successfully prioritize voice transmission over data in real 802.11e networks. We also demonstrate that the AIFSN parameter more effectively protects voice calls against background data traffic than CWmin. To the best of our knowledge, this is the first experimental investigation on tuning of MAC layer parameters in a real 802.11e WLAN network from the perspective of end-to-end voice transmission quality and end user satisfaction.

Assessing the Quality of VoIP Transmission Affected by Playout Buffer Scheme

Miroslaw Narbutt et al. — Sat, 22 Jan 2011 04:45:00 PST

Delay, echo, encoding scheme, and packet loss all influence perceived quality of conversational speech transmitted over packet networks. Therefore, the choice of a buffer algorithm cannot be solely based on statistical loss/delay trade-off metrics. Also subjective “listening tests” or the newer ITU-T PESQ method, which don’t consider the effect of mouth-to-ear delay are inappropriate. We proposed a method for assessing VoIP call quality by extending the ITU-T E-model concept. This method provides a direct link to perceived conversational speech quality by estimating user satisfaction from the combined effect of information loss, delay and echo.

An Experimental Study of the Impact of Using Multi-Radio in WLAN Mesh Networks

Chenzhe Zhang et al. — Sat, 22 Jan 2011 04:44:58 PST

Next-generation wireless mobile communications will be driven by converged networks that integrate disparate technologies and services. Wireless Mesh Networks (WMNs) are expected to be one of the next generation of wireless interconnection technologies, providing flexible high bandwidth wireless backhaul over large geographical areas. While single radio mesh nodes operating on a single channel suffer from capacity constraints, equipping mesh routers with multiple radios using multiple non-overlapping channels can significantly alleviate the capacity problem and increase the aggregate bandwidth available to the network. However, it is well known that multi-radio Mesh platforms face many limitations such as interference, radiation leakage, crosstalk and limited computing resource etc. A misperception among researchers is that the so called “crosstalk” is one of the limiting factors to the performance decrease in multi-radio Mesh networks, but through a series of experiments the results shows otherwise. In this paper, we present a unique experimental approach that utilizes antenna cables, splitters, couplers and attenuators etc. to create a controlled wireless environment, and results indicates, despite negligible effect, crosstalk is not a limiting factor, but rather the finite computing resource constrains the aggregated performance of multi-radio Mesh networks, and several other factors have been investigated as well.

User Quality of Experience-aware Multimedia Streaming over Wireless Home Area Network

Gabriel-Miro Muntean et al. — Sat, 22 Jan 2011 04:44:57 PST

For multimedia streaming over wireless networks, there is a trade-off between the capacity of the wireless links and the end-user perceived-quality, which can be affected by the compression scheme used, content characteristics and adaptation algorithm (if any). In this paper, this trade-off is investigated for streaming various motion content multimedia over an IEEE 802.11b-based Wireless-Home Area Network using the Quality-Oriented Adaptation Scheme (QOAS). QOAS performance is compared to that of a non-adaptive scheme when using MPEG-2 and MPEG-4 encoding in terms of average end-user perceived quality, number of streaming sessions concurrently supported, loss rate, delay, jitter and total throughput. Simulation results show that by using QOAS and MPEG-4 encoded streams a much higher number of concurrent streams are supported at an average quality above “good” level on the ITU-T five-point quality scale in comparison with other situations. In this case all the other streaming performance parameters were also significantly better.

High Performance Modified DSR Routing Protocol for WLAN Mesh Networks

Mustafa Ramadhan et al. — Sat, 22 Jan 2011 04:44:55 PST

This work presents a cross-layer modification to the DSR protocol which discovers high throughput paths on multi-hop wireless mesh networks. The modified DSR incorporates a metric based upon the bandwidth availability at a node into the routing discovery mechanism. We introduce the Access Efficiency Factor (AEF) as an alternative metric to the hop-count for the routing selection mechanism. In this modification, the selected path is identified by finding a path with the highest minimum AEF value. The OPNET modeler has been employed to investigate the performance of the modified DSR protocol on randomly generated network opologies of different node densities. Our results suggest that employing the AEF as a routing discovery metric can significantly increase the average global throughput of wireless mesh networks.

The Effects of Background Traffic on the End-to-End Delay for Video Streaming Applications over IEEE 802.11b WLAN Networks

Nicola Cranley et al. — Thu, 20 Jan 2011 09:58:28 PST

The bursty nature of video streaming applications is due to the frame-based structure of video and this has an important impact on the resource requirements of the WLAN, affecting its ability to provide Quality of Service (QoS) particularly under heavily loaded conditions. In this paper we analyse this bursty behaviour in depth. We show how each video frame is queued at the AP causing the packet delay to vary in a sawtooth manner that is related to the frame rate, the number of packets per video frame, and the packet size. We infer the maximum background traffic load that can be supported so that it does not negatively impact on the video streaming application. We demonstrate that there is a critical threshold load value above which the AP can no longer reliably support the video stream and compare it to the threshold load values calculated through analysis. Using this knowledge, the AP can employ resource allocation mechanisms to regulate the incoming traffic to the AP transmission queue so that QoS can be provided for streaming applications.

A Bandwidth Aware Modification to the DSR Routing Protocol for Wireless Mesh Networks

Mustafa Ramadhan et al. — Mon, 10 Jan 2011 07:05:39 PST

This work proposes a bandwidth aware cross-layer modification to the DSR routing protocol. We include the Access Efficiency Factor (AEF) parameter in addition to the hop-count in the routing discovery mechanism. AEF is a measure to the local availability of bandwidth at a node. Employing the AEF as a metric in the routing discovery mechanism attempts to avoid routing through a congested area in the network. In this modification, we impose a limit on the hop-count in order to control the delay time in the network. The path selection procedure operates by finding a path with the highest minimum AEF value. We have utilized the OPNET modeler simulator to investigate the performance of the modified DSR protocol on a series of randomly generated network topologies of different hopcount limits. The simulator was run twice for each network topology. The first run implemented the standard DSR algorithm while the second implemented the modified DSR protocol. The average global throughput and the average global delay time were recorded for each run. We have calculated the percentage throughput improvement and the percentage delay time increment for each topology. Our results show that using AEF as a routing metric, it is possible to significantly enhance the average global throughput of the network. Furthermore, assigning different values to the hop-count limit allows us to control the network delay time.

MeshScan: Performance of Passive Handoff and Active Handoff

Yin Chen et al. — Mon, 10 Jan 2011 07:02:39 PST

A core problem of fast handoff is when handoff should perform and which Mesh Node (MN) should associated with. We have developed a fast handoff management scheme called MeshScan to provide a novel use of channel scanning latency, by employing open system authentication in both Passive Handoff and Active Handoff. This scheme comprises three steps: firstly a client device takes advantage of the Wireless Mesh Network (WMN) architecture to maintain a list of active MNs. Secondly MeshScan Handoff Sensor performs handoff when it receives a disassociation management frame from the serving MN or when the measured signal strength from the serving MN exceeds a given threshold. Thirdly when handoff is required, a client transmits Authentication Request frames to all MNs from the list instead of broadcasting Probe Request frames, as in an active scan to discover the available MNs. The handoff delay is used as criteria for system performance. Numerical results are presented to demonstrate the feasibility of MeshScan with Active Handoff algorithm. This fast handoff scheme is feasible by upgrading the software only on the client side. This paper compares the theoretical handoff latency of MeshScan with other approaches and we demonstrate the effectiveness of our scheme through experiment.

MeshScan: Fast and Efficient Handoff in IEEE802.11 Mesh Networks

Yin Chen et al. — Mon, 10 Jan 2011 07:02:37 PST

Handoff delay is one of the major problems in Wireless Mesh Network (WMN) that needs to be solved in order to allow time-critical and real-time applications run continuously during handoff. We have developed a fast handoff scheme called MeshScan to provide a novel use of channel scanning latency by employing open system authentication. This scheme comprises two steps: firstly a client device takes advantage of the WMN architecture to maintain a list of active mesh nodes. Secondly when handoff is required, a client transmits Authentication Request frames to all mesh nodes (MNs) from the list instead of broadcasting Probe Request frames as in an active scan to discover the available MNs. This fast handoff scheme is feasible by upgrading the software only on the client side. This paper compares the theoretical handoff latency of MeshScan with other approaches and we demonstrate the effectiveness of our scheme through experiment.

Performance Study of a Cross-layer Optimization to the DSR Routing Protocol in Wireless Mesh Networks

Mustafa Ramadhan et al. — Mon, 10 Jan 2011 07:02:35 PST

A cross-layer modification to the DSR protocol is presented in this work which is intended to enhance the global throughput performance of wireless mesh networks. This modification involves the use of an access efficiency factor (which is a measure of the local availability of bandwidth at a node) as an alternative metric to the hop-count metric in order to allow for improved routing selection. The key feature of this modified route selection procedure is that it attempts to discover paths that have a large availability of bandwidth which can support high throughputs. In this work we have employed the OPNET modeler simulator to investigate the performance of the modified DSR protocol on a series of randomly generated network topologies of different node densities. The simulator was run twice for each network topology. The first run implemented the standard DSR algorithm while the second implemented the modified DSR protocol. The average throughput was recorded for each run and the percentage improvement for the particular topology was calculated. Our results demonstrate that a significant enhancement in the global throughput of the networks can be achieved by implementing our modified DSR protocol.

An Assessment of the Audio Codec Performance in Voice over WLAN (VoWLAN) Systems

Miroslaw Narbutt et al. — Wed, 28 Jan 2009 08:28:29 PST

In this paper we present results of experimental investigation into the performance of three audio codecs (ITU-T G.711, G.723.1 and G.729A) under varying load conditions on a Voice over WLAN system utilizing the IEEE 802.11b wireless LAN standard. The analysis is based upon a new technique for estimating user satisfaction of speech quality calculated from packet delay and packet loss/late measurements. We also demonstrate the importance of the de-jitter buffer playout scheme for insuring speech quality. From our results we conclude that the use of the G.711 audio codec in conjunction with the new adaptive playout scheme gives the highest user satisfaction of the Voice over WLAN schemes considered.

Performance Evaluation of Resource Usage for Unicast Video Streaming over IEEE 802.11 WLAN Networks

Nicola Cranley et al. — Wed, 28 Jan 2009 08:28:28 PST

Multimedia streaming applications are a demanding and challenging service to deliver over wireless networks. Such services have a large impact on the resource requirements of the WLAN. However, there are many variables involved in video streaming, such as the video content being streamed, how the video is encoded and how it is sent. This makes the role of radio resource management and the provision of QoS guarantees extremely difficult. In this paper we investigate the network resource requirements for three scenarios: a single unicast video streaming session; multiple unicast video streaming sessions and finally multiple unicast streaming sessions in the presence of background traffic. We present several key findings: we show the effect that the hint track MTU values has on the access and load requirements of the WLAN. We show that the WLAN becomes saturated when the offered load reaches a certain threshold that is related to the hint tract MTU setting. Finally, we present some preliminary results that show how the access and load requirements of the WLAN are affected when thee is background traffic contending for access to the medium.

Performance Analysis of Network-level QoS with Encoding Configuration for Unicast Video Streaming over IEEE 802.11 WLAN Networks

Nicola Cranley et al. — Wed, 28 Jan 2009 08:28:27 PST

Video streaming has a large impact on the resource requirements of the WLAN. However, there are many variables involved in video streaming, such as the video content being streamed, how the video is encoded and how it is sent. This makes the role of radio resource management extremely difficult. In this paper we investigate the effect that video encoding configurations has on the network resource requirements for unicast video streaming in a WLAN environment. We compare the network resource requirements of several content types encoded at various encoding configurations with varying I-frame frequencies, target encoding bit rates and hint track settings. We present two key findings: We show that by halving the hint track MTU values, the access requirements of the WLAN are increased by 20%. Furthermore, we show how the I-frame frequency of the encoded file relates to the resource requirement of the WLAN.

A Novel Framework for Radio Resource Management in IEEE 802.11 Wireless LANs

Mark Davis et al. — Wed, 28 Jan 2009 08:28:26 PST

In this paper we address the need to characterize and quantify resource usage in IEEE 802.11 WLANs in order to support radio resource management. We present a compact and intuitive framework for performance characterization and resource utilization that is based upon the concept of MAC bandwidth components. These MAC bandwidth components are directly related to the transmission rate and serve to quantify the resource requirements associated with accessing the wireless medium. We also introduce a graphical technique for presenting these MAC bandwidth components that illustrates how WLAN stations interact in contending for access to the wireless medium. We demonstrate the usefulness of this framework for radio resource management using a number of computer simulations based upon the emerging IEEE 802.11e QoS standard.

A Wireless Traffic Probe for Radio Resource Management and QoS Provisioning in IEEE 802.11 WLANs

Mark Davis — Wed, 28 Jan 2009 08:28:24 PST

The emergence of real-time services such as voice over IP (VoIP) and video streaming imposes stringent requirements on the performance of a network if quality of service (QoS) targets are to be achieved. In the case of wireless networks, some form of radio resource management (RRM) is typically required to allocate the available resources among the contending stations in accordance with their needs and respective priorities. A critical aspect of any RRM scheme is the ability to monitor resource usage and to determine the resource requirements on a per-station basis. In this paper we describe a wireless traffic probe for IEEE 802.11 WLANs capable of obtaining this information and presenting it in a compact and intuitive format. The probe also shows how the wireless stations interact with one another in competing for the resources of the WLAN in a clear and quantifiable way. The results from the WLAN traffic probe obtained in a series of video streaming test scenarios are also presented that clearly demonstrate its usefulness and importance as a network tool in RRM and QoS provisioning schemes.

Rare: Resource Aware Routing for mEsh

Karol Kowalik et al. — Wed, 28 Jan 2009 07:09:22 PST

An important element of any routing protocol used for Wireless Mesh Networks (WMNs) is the link cost function used to represent the radio link characteristic. The majority of the routing protocols for WMNs attempt to accurately characterise the radio link quality by constructing the link cost function from the measurements obtained using active probing techniques, which introduces overhead. In this paper we propose a new approach called Resource Aware Routing for mEsh (RARE) which instead employs passive monitoring to gather radio link information. This results in a smaller overhead than the other methods that require active network probing, and is load independent since it does not require an access to the medium. Moreover, se show that our RARE approach performs well in a real radio environment through a number of experiments performed on a static 17 node WLAN mesh testbed.

Experimental Investigation on VoIP Performance and the Resource Utilization in 802.11b WLANs

Miroslaw Narbutt et al. — Wed, 28 Jan 2009 07:09:20 PST

In a shared medium network like the 802.11b WLAN, predicting the quality of VoIP calls from the resource usage of the wireless medium is highly desirable. Analyzing the bandwidth usage at the L2/MAC layer may be especially useful for potential QoS provisioning and call admission schemes. This paper experimentally investigates the relationship between resource utilization in WLANS and the quality of VoIP calls transmitted over wireless medium. Specifically we evaluate how the amount of free bandwidth influences transmission impairments (i.e. delay, loss and jitter) and thus call quality. Resource utilization (under the MAC bandwidth components framework) is calculated by a WLAN resource monitoring application that passively “sniffs” packets at the L2/MAC layer and analyses their headers and temporal characteristics. The quality of VoIP calls is predicted using an extended version of the ITU-T E-model, which estimates user satisfaction from time varying transmission impairments.

Experimental Analysis of an 802.11b Wireless Mesh Network

Brian Keegan et al. — Wed, 28 Jan 2009 07:09:17 PST

Wireless Mesh Networks (WMNs) are a type of radio-based network which require minimal configuration and infrastructure. They can be built using relatively low cost radio and computing platforms. In this paper we attempt to analyze the performance of a simple WNM using various packet sizes as well as a multi-hop WMN.

Experimental Investigation of the Effects of Background Traffic Loads on Streamed Video over 802.11b WLANs

Tanmoy Debnath et al. — Wed, 28 Jan 2009 07:09:16 PST

Video is a frame based media and video streaming is often described as “bursty” which has a large impact on the Quality of Service of the video streaming application over WLAN networks. In this paper the impact of background load on the performance of streaming MPEG-4 video with a video server located on the wired network streaming to wireless client is analysed. We experimentally investigate the performance for both uplink and downlink loads. The performance is measured in terms of the key parameters of bit rate, loss rate and mean delay since these are the primary factors that affect the perceived video quality at the receiver.

The Impact of TCP Sliding Window on the Performance of IEEE 802.11 WLANs

Enrique Roques Gomez et al. — Wed, 28 Jan 2009 07:09:14 PST

In this paper the TCP sliding window mechanism is experimentally investigated as one of the possible causes of the unfairness often observed as IEEE802.11 wireless LANs. We show how by appropriately sizing the sliding window it is possible to re-introduce fairness into the operation of the WLAN.