Target BER selection scheme in LMS networks using AL-FEC systems

摘要

A land mobile satellite (LMS) system is helpful to various fields such as commercial and military applications thanks to its coverage, flexibility, broadcast capability, costs independent on the distance and easy deployment. Besides, since the low data and voice services can be provided via the hand-held terminal, the availability of the LMS system has been increasing. For deep shadowing and non-line-of-sight conditions in the LMS environment, hybrid satellite-terrestrial relay networks (HSTRNs) are widely used. However, the robust link adaptation level would be selected in the LMS environment to prevent the performance degradation of transmission control protocol (TCP), resulting in reducing the spectral efficiency. In this paper, we propose a reliable file transfer framework with application layer forward error correction (AL-FEC) in the LMS network. In particular, we propose a target bit error rate (BER) selection algorithm in the proposed framework to select the permitted target BER much loosely for the AL-FEC system because it is a rate-based system and provides the full reliability of the file transfer. In this paper, the cross-layer design is also considered for the proposed file transfer framework with AL-FEC and the BER selection algorithm. We also theoretically derive the goodput and the spectral efficiency by the given BER of the LMS channel to demonstrate the effectiveness of the proposed framework. The results of performance evaluation show that the required signal to noise ratio of the proposed framework can be reduced with satisfying the goodput requirement of the system as compared with the TCP solution. Furthermore, the link adaptation with the proposed target BER selection algorithm in the proposed framework can reduce the bandwidth needed to send the data in the satellite link, the number of terrestrial relays in the HSTRN, and the transmission power in the ground station while guaranteeing the goodput requirement. (C) 2017 Elsevier B.V. All rights reserved.