Research On Wireless Resources Scheduling Algorithm In B3G System

Future mobile communication system develops towards the trend of mobile broadband and broadband mobile. It will implement B3G mobile communication system which supports high-speed mobile, satisfies various applications'requirement and realizes the all IP framework and the peak rate up to 100Mb/s even 1Gb/s combined with cellular mobile communication, fixed wireless broadband access and mobile wireless broadband access system. It really should satisfy anyone to communicate at any time and any where. This article is based upon "Research of Wireless Transmission Link Technology for Next Cellular Mobile Communication System" funded by the National 863 High-Tech Program. Wireless resources scheduling algorithm in B3G system is researched deeply in order to maximize the limited resources and support QoS of various high-bandwidth multimedia applications.Fair scheduling algorithms which are used in wireline networks such as WFQ and WF2Q can't be addressed effectively to lower the wireless transmission delay and packet loss probability because of two unique characteristics of wireless media: bursty channel errors and location-dependent channel capacity and errors. Resources utilization determined by wireless channel fading is taken into account and dynamic adjustment of the users'priority according to channel state is proposed in current B3G system. But it is not a fundamental solution to the waste of resources caused by channel errors. In this paper, we propose a scheduling algorithm which could allocate wireless resources for data flow according to the channel state information at present time via the research on the ideal wireless fair queuing algorithm, combined with one-step prediction.Simulation result shows that when channel correlation is great, improved wireless resources scheduling algorithm can reduce end-to-end delay by 30% with approaching zero loss probability. Especially when system is heavily loaded, applications'performance can be improved more significantly than the previous algorithm. It's important to the future B3G system in the reference.