Successive cancellation list decoding parameter optimization method based on convolutional neural network
A technology of convolutional neural network and optimization method, which is applied in the direction of using convolutional code error correction/error detection, digital transmission system, data representation error detection/correction, etc., which can solve the problem of increasing complexity and the probability of failure of AD-SCL algorithm advanced questions
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0020] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.
[0021] The invention provides a serial offset list decoding parameter optimization method based on a convolutional neural network, which mainly includes three parts: preparing sample data, building and training a convolutional neural network, and decoding. In the sample data preparation stage, the adaptive serial cancellation list decoding algorithm is firstly executed 100,000 times under different signal-to-noise ratios, and the likelihood ratio calculated from the received signal when decoding is successful and the corresponding likelihood ratio when decoding is successful L records it, then randomly selects 60,000 sets of sample data, and finally randomly selects 75% of the 60,000 sets of data as training samples, and uses the remaining 25% of data as test samples; in the stage of building and training convolutional neural networks , first determine ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 

