A Markov Model for HSDPA TNL Flow Control and Congestion Control Performance Analysis

Abstract

High Speed Downlink Packet Access (HSDPA) is an extension of the Universal Mobile Telecommunication System (UMTS) technology of 3GPP Rel-99, with the objective of increasing the data rate and also reducing the latency in the downlink. The focus of this study is to develop an analytical model for HSDPA flow control (FC) and congestion control (CC) performance analysis. HSDPA flow control adapts transport network data flow to the time varying radio channel capacity of a MAC-d flow [4, 21, 22]. Often the Transport Network Layer (TNL) is congested when there is no sufficient transport capacity to fulfil the radio capacity demand for all connections giving rise to the degradation of the overall HSDPA performance [5, 17, 23, 22]. There are several simulation based investigations carried out so far for the performance analysis of flow control and congestion control algorithms [5, 17, 19, 22, 23]. Further an analytical model which is based on the Embedded Markov Chain with Multiple Departures (EMC- MD) was developed to analyse the effect of CC for the HSDPA performance [18]. However it does not include the flow control functionality which works in cascade with the congestion control algorithm. In order to overcome this issue a joint Markov model is developed which is the focus of this paper. The paper provides detailed modelling aspects of above two algorithms and also the comparison of results with simulation results. There is close agreement with analytical and simulation results.