Homo/Hetero-geneous Cooperation for Cognitive Mobile Communications

Abstract

As the required bandwidth of the communication system gets larger, the transmit signal experiences more severe frequency selective channels. Due to the scarcity of spectrum in the lower frequency bands, next-generation wireless communication systems will be deployed in the higher frequency bands having a large pathloss. To mitigate these problems, modern wireless transceiver adopts complex algorithms, which were considered computationally infeasible in the past. Further, it tries to mimic human behaviors, namely Cooperation and Cognition. Block transmission schemes such as orthogonal frequency division multiplexing (OFDM) and single-carrier frequency domain equlization (SC-FDE) have been combined with multiple input multiple output (MIMO) techniques to attain higher spectral efficiency. Cooperative diversity overcomes the limitation of size and complexity of the mobile equipment without the additional complexity of multiple antennas. The cooperation with the other nodes, so called relay, benefits from pathloss savings, additional power from relays, and diversity in the presence of fading. In this thesis, the distributed space-frequency block coded (D-SFBC) SC-FDE is firstly presented for uplink fast fading channels, where the operations of the relay are processed in the time domain. The proposed D-SFBC SC-FDE achieves the diversity gain and it does not introduce any increase of peak to average power ratio (PAPR) and computational complexity at the mobile equipment, while only the transmit sequences of the relay have PAPR increase. This is desirable especially for the infrastructure-based relay scenario. The performance of the distributed space-time block coded (D-STBC) and D-SFBC SC systems are then analytically compared over fast fading channels.The cognitive radio is defined as ``an intelligent wireless communication system that is aware of its surrounding environment, and uses the methodology of understanding-by-building to learn from the environment and adapt its internal states ...." In this thesis, the terms -- awareness, learning, and adaptation -- are more detailed and discussed in an engineering point of view, namely spectrum awareness (sensing), threshold adaptation for local sensors, and unsupervised learning for data fusion. It will be shown that, as in the classical communications, the cooperation enhances the performance of the cognitive radio. For example, cooperative sensing mitigates the fading effects by making a global decision with the local sensing results from a group of sensors, providing a diversity gain. After investigating the performance of spectrum sensing over frequency-selective fading channels, the mode selection between spectrum underlay and interweaving will be addressed in a unified framework. Threshold adaptation enables smooth migration between the two spectrum sharing modes. In general, the global decision, which is based on the likelihood ratio test, requires the knowledge of operating point and probability of primary transmission. To estimate these parameters, unsupervised learning technique will be also presented.