Degree distribution analysis of LT codes

In this paper we mainly analyze two kinds of degree distribution of LT (Luby Transform) codes, the Ideal Soliton distribution and the Robust Soliton distribution, and find that the one-degree symbol that needed during decoding process is not well distribute. Then we give an improved adaptive encoding method. Simulation results show that, with respect to the mean value, the max value, the standard deviation and the CDF (Cumulative Distribution Function), the adaptive encoding method has better performance. Last we compare our adaptive algorithm with the Robust Soliton distribution in AWGN (Additive White Gaussian Noise) channel of different SNR (Signal Noise Rate) and get the same result.