Novel Variable-Rate Convolutional Coding Scheme for Flat Fading Channels

In wireless communications, the channel gain usually varies according to the Rayleigh distribution. Since the channel condition is time-varying, the fixed-rate convolutional encoder cannot facilitate the best error protection that can be achieved in practice. To mitigate this drawback, we propose a new variable-rate convolutional encoder which can adapt to the dynamical channel conditions subject to the channel state information (CSI) available at the transmitter. Appropriate coding rates are selected based on the instantaneous channel state information and thus robust error protection can be undertaken. Through numerous Monte Carlo simulations, we compare the performances in terms of the average error rate between the conventional fixed-rate convolutional encoder and the variable-rate convolutional encoder for different channels. At the same average transmission rate, our proposed new method outperforms the conventional fixed-rate convolutional encoder by a margin of 3-dB signal-to-noise ratio.