Handbook of Intelligent Computing and Optimization for Sustainable Development. Группа авторов

Чтение книги онлайн.

Читать онлайн книгу Handbook of Intelligent Computing and Optimization for Sustainable Development - Группа авторов страница 57

Handbook of Intelligent Computing and Optimization for Sustainable Development - Группа авторов

Скачать книгу

Data-driven DL approaches use a massive amount of data, which in communication application can be generated in real-time across an enormous number of users. The data sets generated by users and BSs in various environments can be helpful for the 5G networks to learn [50].

Schematic illustration of the auto encoder model for CSI feedback.

       5.3.2.1 Proposed Network Model

Schematic illustration of the inception block. Schematic illustration of the encoder and decoder blocks of InceptNet.

       5.3.2.2 Results and Discussion

      The training and testing of the models are set up in Keras built on top of TensorFlow using Google Colaboratory. COST 2100 channel model is used to generate the data set. The data set provided by Wen et al. [20] is used for simulation here. The training data consists of 100,000 samples. The validation and test set contain 30,000 and 20,000 samples, respectively. All the test samples are independent of the training and validation samples. The network is trained for 100 epochs with a batch size of 200. The learning rate is set to 0.001. Adam optimizer is used to update the parameters, and the mean squared error (MSE) function is used as the loss function. The NMSE quantitatively provides the difference between the original channel matrix image and the recovered channel matrix Hr.

      (5.1)image

      The CSI feedback serves as a beamforming vector. Consider hrn as the reconstructed channel vector of the nth sub-carrier and image as the original channel vector of the nth subcarrier. Cosine similarity (ρ), which measures the quality of the beamforming vector can be given as

      (5.2)image

      where Nc is the number of sub-carriers.

Graphs depict the pseudo gray plots of (a) original image (b) image recovered by CsiNet for CR= 1 over 4 (c) image recovered by InceptNet for CR = 1 over 4.
CR CsiNet InceptNet
NMSE (dB) Cosine Similarity NMSE (dB) Cosine Similarity
1/4 −14.43 0.979 −18.68 0.989
1/8 −11.35 0.963 −13.014 0.973
1/16 −8.08 0.932 −9.357 0.941
1/32 −5.32 0.874 −7.987 0.918

Скачать книгу