الفهرس | Only 14 pages are availabe for public view |
Abstract The rapid growth of communication systems and the high demand in frequency bands have caused a shortage in the available RF spectrum. Cognitive radio is the persuasive solution to the spectral crowding problem. It allows wireless networks to sense the surroundings, adjust, and learn from the previous knowledge to obtain better communication. The efficiency of cognitive radio relies on the speed and accuracy of spectrum sensing. Spectrum sensing requires a continuous listening to the spectrum to provide an opportunistic spectrum allocation for the secondary user according to the state of the primary user. In the first part, a new spectrum sensing technique is proposed to efficiently detect only one primary user. The main idea is based on using an antenna arra at the cognitive receiver instead of using a single antenna element. Versions of the received signal are delayed and summed by using a conventional beamformer. As a result, the signal-to noise ratio of the combined signal is improved due to the array gain. The detection process is performed under the framework of GLRT approach. Direction of arrival of the received signal is assumed to be known or estimated.Closed-form expressions of the test statistic, probability of detection, probability of false alarm, and their associated threshold are derived. Different scenarios of comparison between the proposed GLRT/DOA detector,conventional energy detector (ED), minimum-tomaximu eigenvalue (MME) detector,arithmetic to-geometric mean (AGM)detector, signal-to-noise eigenvalue(SNE)detector, and volume-based(VD)detector are performed. Simulations revealed that the proposed technique outperforms the other well-known techniques. Moreover, simulation results were found to be in agreement with theoretical analysis. |