SATNAC 2008 Conference Papers

Network Planning & General Topics

Title:                An Improved Capacity Modeling Approach for 802.11 Protocols

 

Authors:            Neville Greyling (University of Stellenbosch), Riaan Wolhuter (University of Stellenbosch)

 

Abstract:           Despite the widespread deployment of IEEE 802.11 Wireless Local Area Network (WLAN) protocol, not much analysis has been done on modelling its performance, to determine what amount of users such a network can accommodate. In this paper one approach to such a model is introduced, in order to assist with capacity determination on WLAN networks. We introduce a model to compute the number of users that such a WLAN can accommodate. The model is applicable to applies to unsaturated traffic conditions, and is suited for all traffic arrival patterns. Further more, the model tries to emulate model such a network as accurately as possible with moderate simplifications.

 

 

Title:                The Recursive Evaluation of the M/M/r/K/K Queue

 

Authors:            Ian Kennedy (University of the Witwatersrand), Anthony Krzesinski (University of Stellenbosch)

 

Abstract:           We present a recursive form of the equation for the probability that the queueing delay in an M/M/r/K/K queue is larger than a given bound. The recursive form is computationally efficient and is not subject to numerical overflow. The practical implication of the recursion is that the teletraffic engineer has an accurate, practical formula for calculating the delay probability distribution for switches where the calls are offered from a finite population.

 

 

Title:                Characteristics of Seasonal Attenuation and Fading for Line-of-Sight Links in South Africa

 

Authors:            Modupe Odedina (University of KwaZulu-Natal), Thomas Afullo (University of KwaZulu-Natal)

 

Abstract:           The characteristics of seasonal rain attenuation are studied for five different geographical locations in South Africa. This is to investigate particular features and behaviours associated with different seasons in the various locations. These five locations are chosen for study because of their unique climatic features which contributes significantly to their propagation behaviours and fading predictions. And these locations are Bloemfontein, Brandvlei, Cape Town, Durban, and Pretoria. The seasonal cumulative distributions of rain rate and rain attenuation are estimated for each location from a 5-year (2000-2004) point rainfall data collected directly by the South African Weather Services from each location (SAWS).  From these seasonal distributions, appropriate figures for fade margins were derived, which gives the percentage of time for which a certain attenuation level is exceeded for different seasons in each location. And this fade margins gives the necessary allowance to accommodate expected fading that may want to occur on the LOS links in such locations. From the actual signal attenuation measurement recorded for one year in one of these five geographical locations for 6.73 km horizontally polarized 19.5 GHz terrestrial LOS link, the average attenuation for the year was computed. And these experimental results show a good agreement with the ITU-R model.

 

 

Title:                Estimation of Secondary Radioclimatic Variables and Its Application to Terrestrial LOS Link Design in South Africa

 

Authors:            Peter Odedina (University of KwaZulu-Natal), Thomas Afullo (University of KwaZulu-Natal)

 

Abstract:           In order to propose a reasonable prediction models for radioclimatic study, a reliable radio propagation data is required. This radio propagation data can be either primary radio propagation data or secondary. The secondary data can be estimated from a relevant primary data. Primary radioclimatic data include temperature, pressure and humidity or water vapour pressure, while secondary radioclimatic data includes, refractivity data such as refractivity gradients, ducting data, geoclimatic factor and also data that incorporate the effect of earth curvature on radiowave propagation such as effective earth radius factor (k-factor). In this presentation, our focus is on such secondary data. We have used 10 years data (1985-1994) in three different regions of South Africa (Durban, Bloemfontein and Cape Town) for terrestrial LOS link design application.

 

 

Title:                Route Selection Impacts on Achieving Enhanced IMS QoS

 

Authors:            Vitalis Ozianyi (University of Cape Town), Gabriel Andrews (University of Cape Town), Neco Ventura (University of Cape Town)

 

Abstract:           The different planes in the IMS interact via specific reference points to deliver multimedia services to the user. QoS provisioning for IMS communications has been standardized for access networks only, with the assumption of an over provisioned IP core. Effective provisioning of multimedia services requires performance guarantee along the complete path of the sessions. End-to-end QoS in IP networks is affected by the route traversed by the user traffic. Moreover QoS guarantees in one ISP domain are not effective for transit traffic exiting the domain. QoS extensions to exterior gateway routing protocols have been proposed to transfer route QoS information beyond one autonomous system (domain). This paper explores options for mapping inter-domain QoS information learnt on the media plane into control plane session information for IMS QoS control. Through testbed evaluations we show the effect of routing on delays experienced in IMS communications.

 

 

Title:                Solving the Extended Tree Knapsack Problem (ETKP) with Fixed Cost Flow Expansion Functions

 

Authors:            David van der Merwe (North-West University), Giel Hattingh (North-West University)

 

Abstract:           Parts of the LATN (Local Access Telecommunication Network) planning problem can be modelled with the Extended Tree Knapsack problem.  The LATN can constitute up to 60% of total network costs.  In this paper a partitioning algorithms using standard of the shelf software coupled with enhanced modelling is presented.  Enhancements to the algorithms is presented along with empirical results.

 

 

Title:                Hamming Error Correction Techniques for the Improvement of Robustness in Networks using Random Network Coding

 

Authors:            Suné von Solms (North-West University), Albert Helberg (North-West University)

 

Abstract:           In this paper, we introduce an algorithm for error detection and correction in random network coding. The introduced technique exploits the encoding characteristics of random network coding and uses the well known Hamming Code as a decoding algorithm. For a network where random network coding is applied, this technique can be a useful error detecting and correcting method that will improve the network’s robustness.

 

 

Title:                Mapping the Location of 2.4 GHz Transmitters to Achieve Optimal Usage of an IEEE 802.11 Network

 

Authors:            Daniel Wells (Rhodes University), Ingrid Siebörger (Rhodes University), Barry Irwin (Rhodes University)

 

Abstract:           This paper describes the use of a low cost 2.4 GHz spectrum analyser, the MetaGeek WiSpy device, in conjunction with custom developed client-server software for the accurate identification of 2.4 GHz transmitters within a given area. The WiSpy dongle together with the custom developed software allow for determination of the positions of Wi-Fi transmitters to within a few meters, which can be helpful in reducing the work load for physical searches in the process of surveying the Wi-Fi network and geographical area. This paper describes the tool and methodology for a site survey as a component that can be used in organisations wishing to audit their environments for Wi-Fi networks.  The tool produced from this project, the WiSpy Signal Source Mapping Tool, is a three part application based on a client-server architecture. One part interfaces with a low cost 2.4 GHz spectrum analyser, another stores the data collected from all the spectrum analysers and the third part interprets the data to provide a graphical overview of the Wi-Fi network being analysed. The location of the spectrum analysers are entered as GPS points, and the tool can interface with a GPS device to automatically update its geographical location. The graphical representation of the 2.4 GHz spectrum populated with Wi-Fi devices (Wi-Fi network) provided a fairly accurate method in locating and tracking 2.4 GHz devices. Accuracy of the WiSpy Signal Source Mapping Tool is hindered by obstructions, interferences within the area or non line of sight.

 
< Back to Conference Paper Topics