Optimal demodulator for satellite-based wireless ATM networks

In this paper, we propose a Bayesian (known as optimal) demodulation scheme for satellite-based time-division-multiple-access (TDMA) wireless asynchronous transfer mode (ATM) networks. Rician flat fading is assumed for the satellite channel (no multipath). Because it is asynchronous between the transmission and receiver in wireless ATM networks, it is most important to focus on the demodulator design. The parameters for the Bayesian demodulator are determined using an unsupervised (blind) clustering method - fuzzy c-means (FCM). Our simulation results show that our demodulator performs better than the block phase estimation demodulator in terms of uncoded bit error rate (BER) with a gain of 0.2 to 0.3 dB.