- Radio Signal Propogation and Processing
- Radio Modem Innovations
Develop a novel low-cost intelligent spectrum sensing architecture so that CR can adapt its transmission more flexibly and intelligently.
Radio spectrum over-crowding resulting in congestion is a problem concerning both military and commercial users alike. Recently, cognitive radio technology has attracted a lot of interests recently to deal with this problem. For example, in IEEE 802.22 standard, secondary user is allowed to detect unused spectrum bands called spectrum holes) in TV band to transmit its own data without interfering with primary users. There are three major components in cognitive radio: Sense, Learn and Adapt. The cognitive radio senses the spectrum to find existence of primary users and spectrum holes, then learns the environment and adapts its transmission to fit in the current environment.
Evidently, the effectiveness of a cognitive radio is directly dependent on its ability to accurately detect its environment and the transmission opportunities in time and frequency domain. Spectrum sensing for cognitive radio has been an active research topic in recent years and many spectrum sensing methods have been proposed. However, existent spectrum sensing methods are not without their drawbacks. The most commonly used spectrum sensing method is energy based sensing. Energy based sensing has very low computation complexity and is easy to implement. However, its performance degrades significantly in low SNR range, especially in multi-path fading channels. Matched filter based spectrum sensing offers the best detection accuracy, but it requires full knowledge of the target signal and is very complicated. Waveform based sensing provides almost the same performance of the matched filter based sensing at lower complexity, but it still requires some a priori knowledge of the target signals. Cyclostationary sensing takes advantage of the cyclostationary features of the target signal and offers excellent detection accuracy without a priori knowledge of the target signals. However, current cyclostationary detection still requires significant computational power. It is highly desired to develop a new spectrum sensing architecture which provides superior detection accuracy at lower computational complexity and reasonable processing power.
More importantly, we believe there is another fatal weakness in current spectrum sensing research. Current spectrum sensing only offers a binary decision on the existence of the target signal (and spectrum hole), nothing more. Simply detecting the existence of some RF transmissions is not enough to maintain reliable and secure communication. It is highly desired, sometimes required, to obtain important RF characteristics of the target signals and identify them, and to act accordingly.
Hence, the proposed innovation is closely related to two emphasis areas of this challenge: (1) Radio signal propagation and processing; and (2) Radio modem innovations.
Applicant chose not to share this document publicly.
Our team is focusing on software defined radio based cognitive radio, wireless sensor network, physical layer design, and signal processing. We had the privilege of being the first in the world to demonstrate an autonomous overlay cognitive radio in mobile environment via software defined radio, which is able to stitch multiple spectrum holes and operate over multiple non-contiguous spectrum fragments. Meanwhile, real-time seamless video transmission without interference from/to primary uses and without interference on each and every subcarrier in mobile environment is maintained, and the co-existence between primary users and secondary users is provided. This implementation and demonstration won the prestigious IEEE Globecom Best Demo Award in 2010. We also have close collaboration with AFRL and industry companies. Therefore, we possess strong background and experience to solve the current/future problems in wireless communications with our innovative design.