The decoding algorithm of LDPC codes is often implemented by integer quantization in FPGA

which leads to the degradation of decoding performance. The NMS decoding algorithm is introduced to reduce the error by multiplying an improved factor with a value range of (0

1) on the basis of quantization of check point information data. By studying the reasonable value of the improvement factor

a new quantization method is proposed to improve the quantization results by taking different improvement factors with the iteration times. The research object is the (8176

7154) LDPC code of near earth space application in the CCSDS standard. The encoding and decoding algorithm program is designed and simulated on MATLAB. Simulation results show that the improved quantization method needs less iterations to complete decoding and improves the decoding performance. By analyzing the change of iteration times under different signal-to-noise ratio

the advantage is more obvious under higher noise interference.