Hint for AI/ML/Deep learning

What is Machine Learning: definitions will be updated soon

AI allows machines to develop capabilities that are equal to or surpass human intelligence (e.g., intelligent robots). Another relevant factor is machine learning (ML) as a subcategory of AI. It is, for instance, used to build systems that learn from data sets rather than from programmed instructions, thus leading to a learning process based on artificial multi-layer neural networks or deep learning. Future wireless network that comes with an AI-native air interface, making radios capable of learning from the environment and from each other based on trained neural networks. Neural networks are a subcategory of machine learning and relevant in wireless communication. The following neural networks will be considered on this page:

Recurrent neural network (RNN)

Convolutional neural network (CNN)

Concept of an autoencoder

Observation: Deep learning is an art. Every dataset is unique; It is recommended to evaluate different strategies empirically. There is currently no theory that will tell in advance precisely what to do to optimally solve a problem. It is useful to establish a common-sense baseline before trying machine learning approaches. It is useful to try simple, cheap machine learning models before looking into complicated and computationally expensive models. It is good to make sure any further complexity throw at the problem is legitimate and delivers real benefits.

In addition to AI/ML books or lecture notes (plenty FREE docs can be found online) you might have, Scikit-learn, Pytorch, TensorFlow, and Keras are good tools across the machine learning and data science industry. The best Youtube tutorial I came across was from Aladdin Persson.

Dataset used for AI/ML/Deep learning or training

NIST RF Dataset of Incumbent Radar Systems in the 3.5 GHz CBRS Band

RF Data Factory Dataset

DEEPSIG Dataset

RF Challenge at MIT

Kaggle RF Signal

IEEE DRONE REMOTE CONTROLLER RF SIGNAL DATASET

IEEE Radio-frequency dataset

NextG Channel Model Alliance

Fig.1 - Deep Learning network architecture model

Challenges of Machines Learning in Wireless Communications:

Quality of data

Time-Consuming task

Issue of overfitting & underfitting

Difficulty in deployment

Algorithms to develop on this page

Deep learning channel estimation

Modulation classification via machine learning

Deep learning for indoor localization based on CSI

Classification logistic regression

Classification support vector machine

Classification decision tree

Classification Random Forest

Classification Bayes Naive

Regression Linear Regression

Clustering K-Means

Clustering Mean Shift

KNN

This work is in progress ...