About this Research Topic
Future wireless communication systems are not used only for communication but also for sensing. The number of devices used in these systems is also increasing daily. This naturally brings up the problem of spectrum insufficiency. For this reason, spectrum should be used more efficiently than before. To improve spectrum utilization efficiency requires the accurate prediction of spectrum occupancy and efficient resource allocation/management. Numerous model-based methods have been developed for this purpose. However, classical methods may be insufficient for complex scenarios (such as when the number of users is large, the environment is constantly changing, and the number of datasets is large).
Recently, machine learning (ML) techniques have been used for complex problems involving wireless communication. This is especially the case where the spectrum occupancy predictions are tried under correlated channels and resource allocation/management has been attempted, by ascertaining the correlations in different dimensions such as time, frequency, and space. Therefore, this Research Topic aims to explore new algorithms/models, comparisons of model and ML methods, measurements and performance tests carried out with real data in different environments, and theoretical analyses with varying sets of data. It is hoped that this Research Topic will thus develop and compare ML-based spectrum occupancy prediction and resource allocation/management, specifically pointing out multiple dimensions.
Themes of interest include (but are not limited to):
1. Designing novel ML models or algorithms for spectrum occupancy prediction and resource allocation/management;
2. Real-world experiments for spectrum occupancy prediction and resource allocation/management;
3. Theoretical comparison of ML models with model-based approaches;
4. Novel feature selection methods for ML models in spectrum occupancy prediction and resource allocation/management problems;
5. Complexity, accuracy, and memory requirements comparisons of ML algorithms for spectrum occupancy prediction and resource allocation/management;
6. ML-based multi-dimensional spectrum occupancy prediction and resource allocation/management algorithms design and comparisons.
Recently, machine learning (ML) techniques have been used for complex problems involving wireless communication. This is especially the case where the spectrum occupancy predictions are tried under correlated channels and resource allocation/management has been attempted, by ascertaining the correlations in different dimensions such as time, frequency, and space. Therefore, this Research Topic aims to explore new algorithms/models, comparisons of model and ML methods, measurements and performance tests carried out with real data in different environments, and theoretical analyses with varying sets of data. It is hoped that this Research Topic will thus develop and compare ML-based spectrum occupancy prediction and resource allocation/management, specifically pointing out multiple dimensions.
Themes of interest include (but are not limited to):
1. Designing novel ML models or algorithms for spectrum occupancy prediction and resource allocation/management;
2. Real-world experiments for spectrum occupancy prediction and resource allocation/management;
3. Theoretical comparison of ML models with model-based approaches;
4. Novel feature selection methods for ML models in spectrum occupancy prediction and resource allocation/management problems;
5. Complexity, accuracy, and memory requirements comparisons of ML algorithms for spectrum occupancy prediction and resource allocation/management;
6. ML-based multi-dimensional spectrum occupancy prediction and resource allocation/management algorithms design and comparisons.
Keywords: Spectrum Opportunities, wireless communications, resource allocation and management, cognitive radio, machine learning, spectrum occupancy prediction
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
