A Study on Validity of Random Frame Error Assumption from the Viewpoint of TCP Performance in DS-CDMA Cellular System

Jumpei TAKETSUGU  Shinsuke HARA  

IEICE TRANSACTIONS on Communications   Vol.E88-B    No.7    pp.2994-3004
Publication Date: 2005/07/01
Online ISSN: 
DOI: 10.1093/ietcom/e88-b.7.2994
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Terrestrial Radio Communications
TCP,  transmission power control,  download,  DS-CDMA,  cellular system,  

Full Text: PDF>>
Buy this Article

Many reports have investigated TCP performance over wireless links, where a high and time-invariant frame error rate is assumed for cellular systems. However, the frame error rate is temporally and geographically changeable by fading and interference in cellular systems. On the other hand, SINR-based transmission power control, which is employed for the randomization of frame errors in DS-CDMA cellular systems, can not always work properly depending on the control parameters or the channel characteristics. In this paper, we investigate the TCP performance over the wireless links in a DS-CDMA cellular system by computer simulation. From the simulation results, it has been found that the assumption of random frame error is valid only for a part of the TCP performance even in the system with an SINR-based transmission power control scheme.