GlobalSIP 2016 - 2016 IEEE Symposium on Machine Learning for Characterization of Cognitive Communications and Radar

Over the next 3-5 years wireless communication is projected to grow exponentially, while resources in terms of power and bandwidth will remain limited. The widening gap between demand and available resources is emerging as one of the major challenges for all entities sharing the electromagnetic spectrum. Cognitive Radio (CR), with its capability to sense its environment and flexibly adjust its transceiver parameters, has established itself as an enabling methodology for dynamic time-frequency-space resource allocation and management, offering significant improvement of spectral utilization. However, existing cognitive radio models would no longer be adequate, given the massive demands for capacity, connectivity, high reliability and low latency, so novel models and algorithms are needed to help improve spectrum utilization.
A natural approach to handling these challenges is the development of a broad range of efficient machine learning algorithms, as well as new frameworks for cooperative learning and sharing, based on complex signal patterns in space, frequency and time. Proliferation of software defined radio technology, as well as applications in Self-organized Networks, Machine-to-Machine Communications, Internet of Things etc, will necessarily create even more complex environments in which CR networks of secondary users will compete for spectrum access not only with primary users, but also with other CR networks. Many of these dense multi-user cognitive radio systems would be difficult to capture using conventional machine learning models.
We recognize that characterization of cognitive communication and radar is emerging as a topic area with rich potential, high relevance and broad applicability for machine learning research and development. The goal of this Symposium is to bring together researchers from the cognitive communication and machine learning communities, to showcase state-of-the-art machine learning approaches to CR network problems, and to provide a forum for discussing ideas to address present and future challenges using novel synergistic approaches.