• DocumentCode
    499152
  • Title

    Understanding directional load balancing using per call measurement data

  • Author

    Vaidyanathan, Anu ; Wong, Wilfred ; Billinghurst, Mark ; Sirisena, Harsha

  • Author_Institution
    HIT Lab., Univ. of Canterbury, Christchurch, New Zealand
  • Volume
    41
  • fYear
    2009
  • fDate
    13-16 July 2009
  • Firstpage
    213
  • Lastpage
    220
  • Abstract
    Several load balancing techniques have been proposed and studied for CDMA systems using a variety of simulation-based data-sets. In this study, we present actual call data captured at Wellington, New Zealand, from their leading telephone services provider, in order to understand how calls in the active-set are shared amongst cell-sites. Two groups of cell-sites are analysed to understand this phenomenon including (a) the busiest cell-sites, which experience the highest call volume and (b) the cell-sites with the poorest coverage or reference pilot signal strength. We obtain a day long trace on the date of an All Blacks Rugby game in Wellington, New Zealand and further delineate the rush-hour data on this day. Rush hour is defined as the hour at which the most calls are placed during the day, which co-incided with the time at which the Rugby game started. We observe that the busiest sites are located in downtown Wellington and calls in the active set migrate in the direction of caller traffic, which was moving towards the Westpac stadium, where the Rugby game was held. Calls are not picked up by the nearest neighbors (to the busiest sites) and are absorbed by the neighbors that are proximal to the stadium, suggesting directional load balancing. In the case of cell-sites with poor coverage, the load is distributed amongst the nearest neighbors, when the coverage of a particular cell-site reduces, over time.
  • Keywords
    cellular radio; code division multiple access; All Blacks Rugby game; CDMA systems; Wellington, New Zealand; cell-sites; directional load balancing; per call measurement data; rush hour; simulation-based data-sets; telephone services provider; Cities and towns; Computational modeling; Electric variables measurement; Load management; Multiaccess communication; Nearest neighbor searches; Pulse measurements; Signal analysis; Telecommunication traffic; Telephony; CDMA Networks; Load balancing; Per Call Measurement Data;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Performance Evaluation of Computer & Telecommunication Systems, 2009. SPECTS 2009. International Symposium on
  • Conference_Location
    Istanbul
  • Print_ISBN
    978-1-4244-4165-5
  • Electronic_ISBN
    978-1-56555-328-6
  • Type

    conf

  • Filename
    5224123