A method of blind recognition of the coding parameters for binary Bose-Chaudhuri-Hocquenghem (BCH) codes is proposed in this paper. We consider an intelligent communication receiver which can blindly recognize the coding parameters of the received data stream. The only knowledge is that the stream is encoded using binary BCH codes, while the coding parameters are unknown. The problem can be addressed on the context of the non-cooperative communications or adaptive coding and modulations (ACM) for cognitive radio networks. The recognition processing includes two major procedures: code length estimation and generator polynomial reconstruction. A hard decision method has been proposed in a previous literature. In this paper we propose the recognition approach in soft decision situations with Binary-Phase-Shift-Key modulations and Additive-White-Gaussian-Noise (AWGN) channels. The code length is estimated by maximizing the root information dispersion entropy function. And then we search for the code roots to reconstruct the primitive and generator polynomials. By utilizing the soft output of the channel, the recognition performance is improved and the simulations show the efficiency of the proposed algorithm.

Deep learning is used for automatic modulation recognition in neural networks, and because of the need for high classification accuracy, deeper and deeper networks are used. However, these are computationally very expensive for neural network training and inference, so its utility in the case of a mobile with memory limitations or weak computational power is questionable. As a result, a trade-off between network depth and network classification accuracy must be considered. To address this issue, we used a knowledge distillation method in this study to improve the classification accuracy of a small network model. First, we trained Inception–Resnet as a teacher network, which has a size of 311.77 MB and a final peak classification accuracy of 93.09%. We used the method to train convolutional neural network 3 (CNN3) and increase its peak classification accuracy from 79.81 to 89.36%, with a network size of 0.37 MB. It was also used similarly to train mini Inception–Resnet and increase its peak accuracy from 84.18 to 93.59%, with a network size of 39.69 MB. When we compared all classification accuracy peaks, we discover that knowledge distillation improved small networks and that the student network had the potential to outperform the teacher network. Using knowledge distillation, a small network model can achieve the classification accuracy of a large network model. In practice, choosing the appropriate student network based on the constraints of the usage conditions while using knowledge distillation (KD) would be a way to meet practical needs.