Paal E. Engelstad

Non-saturation and saturation analysis of IEEE 802.11e EDCA with starvation prediction

An analytical model is proposed to describe the priority schemes of the EDCA mechanism of the IEEE 802.11e standard. EDCA provides class-based differentiated QoS to IEEE 802.11 WLANs. The main contribution of this paper opposed to other works is that the model predicts the throughput, delay and frame dropping probabilities of the different traffic classes in the whole range from a lightly loaded, non-saturated channel to a heavily congested, saturated medium. Furthermore, the model describes differentiation based on different AIFS-values, in addition to the other adjustable parameters (i.e. window-sizes, retransmission limits, TXOP lengths etc.) also encompassed by previous models. AIFS differentiation is described by a simple equation that enables access points to predict at which traffic loads starvation of a traffic class will occur. The model is calculated numerically and validated against simulation results. We observed a good match between the analytical model and simulations.

Analysis of the Total Delay of IEEE 802.11e EDCA and 802.11 DCF

A packet sent from an upper-layer protocol or application over IEEE 802.11 [1] will first be placed in a transmission queue. The packet delay caused by waiting here is referred to as the queueing delay. When the packet reaches the head of the queue it will start contending for channel access until it is successfully transmitted over the medium (or finally dropped). The delay associated with the medium access is referred to as the MAC delay. The majority of analytical work on the delay performance of IEEE 802.11 focuses on predicting only the mean MAC delay, although higher layer applications and protocols are interested in the total performance of the MAC layer. The main contribution of this paper opposed to other works is that it provides analytical predictions of the total delay, which also includes the queueing delay. The analyses presented apply to the priority schemes of the Enhanced Distributed Channel Access (EDCA) mechanism of the IEEE 802.11e standard [2]. However, by using an appropriate parameter setting, the results presented are also applicable to the legacy 802.11 Distributed Coordination Function (DCF) [1]. The model predictions are calculated numerically and validated against simulation results.

Improving the performance of quality-adaptive video streaming over multiple heterogeneous access networks

Devices capable of connecting to multiple, overlapping networks simultaneously are becoming increasingly common. For example, most laptops are equipped with LAN- and WLAN-interfaces, and smart phones can typically connect to both WLANs and 3G mobile networks. At the same time, streaming high-quality video is becoming increasingly popular. However, due to bandwidth limitations or the unreliable and unpredictable nature of some types of networks, streaming video can be subject to frequent periods of rebuffering and characterised by a low picture quality. In this paper, we present a client-side request scheduler that distributes requests for the video over multiple heterogeneous interfaces simultaneously. Each video is divided into independent segments with constant duration, enabling segments to be requested over separate links, utilizing all the available bandwidth. To increase performance even further, the segments are divided into smaller subsegments, and the sizes are dynamically calculated on the fly, based on the throughput of the different links. This is an improvement over our earlier subsegment approach, which divided segments into fixed size subsegments. Both subsegment approaches were evaluated with on-demand streaming and quasi-live streaming. The new subsegment approach reduces the number of playback interruptions and improves video quality significantly for all cases where the earlier approach struggled. Otherwise, they show similar performance.

The availability and reliability of wireless multi-hop networks with stochastic link failures

The network reliability and availability in wireless multi-hop networks can be inadequate due to radio induced interference. It is therefore common to introduce redundant nodes. This paper provides a method to forecast how the introduction of redundant nodes increases the reliability and availability of such networks. For simplicity, it is assumed that link failures are stochastic and independent, and the network can be modelled as a random graph. First, the network reliability and availability of a static network with a planned topology is explored. This analysis is relevant to mesh networks for public access, but also provides insight into the reliability and availability behaviour of other categories of wireless multi-hop networks. Then, by extending the analysis to also consider random geometric graphs, networks with nodes that are randomly distributed in a metric space are also investigated. Unlike many other random graph analyses, our approach allows for advanced link models where the link failure probability is continuously decreasing with an increasing distance between the two nodes of the link. In addition to analysing the steady-state availability, the transient reliability behaviour of wireless multi-hop networks is also found. These results are supported by simulations.

Using HTTP Pipelining to Improve Progressive Download over Multiple Heterogeneous Interfaces

Today, mobile devices like laptops and cell phones often come equipped with multiple network interfaces, enabling clients to simultaneously connect to independent access networks. Even though applications, such as multimedia streaming and video-on-demand delivery systems, could potentially benefit greatly from the aggregated bandwidth, implementation and standardization challenges have so far hindered the deployment of multilink solutions. Previously, we have explored the benefits of collaboratively using multiple Internet connections to progressively download and play back large multimedia files. In this paper, we present an improved version of our approach that utilizes HTTPs capability of request pipelining in combination with range retrieval requests. While, in our earlier work, the optimal choice of file segmentation size presented a tradeoff between throughput and startup latency, the enhanced solution is able to overcome this tradeoff. The use of very small segments no longer impairs the efficiency of throughput aggregation, which additionally makes our solution robust against link variances and agnostic to network heterogeneity.

A network-layer proxy for bandwidth aggregation and reduction of IP packet reordering

With todays widespread deployment of wireless technologies, it is often the case that a single communication device can select from a variety of access networks. At the same time, there is an ongoing trend towards integration of multiple network interfaces into end-hosts, such as cell phones with HSDPA, Bluetooth and WLAN. By using multiple Internet connections concurrently, network applications can benefit from aggregated bandwidth and increased fault tolerance. However, the heterogeneity of wireless environments introduce challenges with respect to implementation, deployment, and protocol compatibility. Variable link characteristics cause reordering when sending IP packets of the same flow over multiple paths. This paper introduces a multilink proxy that is able to transparently stripe traffic destined for multihomed clients. Operating on the network layer, the proxy uses path monitoring statistics to adapt to changes in throughput and latency. Experimental results obtained from a proof-of-concept implementation verify that our approach is able to fully aggregate the throughput of heterogeneous downlink streams, even if the path characteristics change over time. In addition, our novel method of equalizing delays by buffering packets on the proxy significantly reduces IP packet reordering and the buffer requirements of clients.

Internet connectivity for multi-homed proactive ad hoc networks

A prerequisite for a widespread and successful deployment of proactive ad-hoc networking technology is its ability to provide easy access to the Internet. Normally, proactive routing protocols provide routing messages that establish default routes to ensure connectivity for outgoing IPv4 packets destined for the Internet. However, mechanisms to ensure connectivity for incoming IPv4 packets from the Internet are yet poorly documented in published material. Possible solutions include implementing a modified mobile IPv4 foreign agent (MIP-FA) or network address translation (NAT) on each Internet gateway node in the ad hoc network. In this paper we discuss different strategies for providing Internet access to proactive ad hoc networks. We also describe problems experienced in our lab test-bed with default routes under the condition of site multi-homing. Based on this experience, we propose working solutions for Internet access from proactive ad hoc networks.

Fixed WiMAX Field Trial Measurements and Analyses

Much hype is spread about the performance of WiMAX, and many contrary statements are put forward. A reason for the confusion is that little or no published material exists about WiMAX performance based on measurements in real life field trials. We therefore set up a test-bed using WiMAX equipment now on the marked, and did extensive, real life field trial measurements. This paper presents measured throughput and received signal strength under conditions typical for fixed WiMAX systems. Based on the results, throughput is analyzed and analytical expressions representing the measurements are derived for easy use in models.

Evaluation of Service Discovery Architectures for Mobile Ad Hoc Networks

Discovery of services and other named resources is expected to be a crucial feature for the usability of mobile ad-hoc networks (MANETs). Different types of service discovery architectures are distinguished by the extent that service coordinators (SCs) are implemented in the network. SCs are nodes that hold a central repository for the caching of service attributes and bindings of servers located in its neighborhood. This paper describes and investigates the performance of different architectures in terms of service availability, messaging overhead and latency. The architectures are analyzed and evaluated based on a large range of parameters. The paper also shows that in an on-demand MANET, where the service discovery mechanism will have a direct impact on the reactive routing protocol, the routing effects influence the evaluation result. It also demonstrates the benefits of combining the service discovery with route discovery, especially in on-demand MANETs where reactive routing protocols are being used.

Energy minimization approach for optimal cooperative spectrum sensing in sensor-aided cognitive radio networks

In a sensor-aided cognitive radio network, collaborating battery-powered sensors are deployed to aid the network in cooperative spectrum sensing. These sensors consume energy for spectrum sensing and therefore deplete their life-time, thus we study the key issue in minimizing the sensing energy consumed by such group of collaborating sensors. The IEEE P802.22 standard specifies spectrum sensing accuracy by the detection and false alarm probabilities, hence we address the energy minimization problem under this detection accuracy constraint. Firstly, we derive the bounds for the number of sensors to simultaneously guarantee the thresholds for high detection probability and low false alarm probability. With these bounds, we then formulate the optimization problem to find the optimal sensing interval and the optimal number of sensor that minimize the energy consumption. Thirdly, the approximated analytical solutions are derived to solve the optimization accurately and efficiently in polynomial time. Finally, numerical results show that the minimized energy is significantly lower than the energy consumed by a group of randomly selected sensors. The mean absolute error of the approximated optimal sensing interval compared with the exact value is less than 4% and 8% under good and bad SNR conditions, respectively. The approximated optimal number of sensors is shown to be very close to the exact number.