A study on simulation probability density function design for fast simulation of turbo codes over slow Rayleigh fading channels

This paper proposes a fast simulation method of turbo codes over slow Rayleigh fading channels. The proposed method reduces the simulation runtime by applying importance sampling (IS). The performance of IS is affected by the design of the simulation probability density function (PDF), which is different from the true PDF of the noise. While the conventional IS method for turbo codes over Rayleigh fading channels focuses only on the modification of the noise sequence of the additive white Gaussian noise channel, the proposed simulation PDF biases channel gains of the Rayleigh fading channel in addition to the channel noises. Moreover, the effectiveness of the proposed method is confirmed by some numerical examples. On the evaluation of a frame error rate of 10 ⁶, the computer runtime of the proposed method is about 1/5 of that of the conventional IS method. When we compare with the Monte Carlo simulation method, the proposed method needs only 1/100 simulation runtime under the condition of the same accuracy of the estimator.