The recent increasing interest in cognitive radio networks has motivated the study and development of new approaches capable of coping with the intrinsic challenges of this kind of network,such as dynamic spectrum availability,distributed and heterogeneous network architectures,and soaring complexity.The bio-inspired approaches,with appealing characteristics such as autonomy,adaptation and collective intelligence of collaborative individuals,have been extensively studied.This paper presents a comprehensive survey of bio-inspired approaches for cognitive radio networks,emphasizing their domains of application.Specifically,ant colony optimization and particle warm optimization are further investigated with examples and numerical simulation.
HE ZhiQiangNIU KaiQIU TaoSONG TaoXU WenJunGUO LiLIN JiaRu
In this study, the robustness of small-world networks to three types of attack is investigated. Global efficiency is introduced as the network coefficient to measure the robustness of a small-world network. The simulation results prove that an increase in rewiring probability or average degree can enhance the robustness of the small-world network under all three types of attack. The effectiveness of simultaneously increasing both rewiring probability and average degree is also studied, and the combined increase is found to significantly improve the robustness of the small-world network. Furthermore, the combined effect of rewiring probability and average degree on network robustness is shown to be several times greater than that of rewiring probability or average degree individually. This means that small-world networks with a relatively high rewiring probability and average degree have advantages both in network communications and in good robustness to attacks. Therefore, simultaneously increasing rewiring probability and average degree is an effective method of constructing realistic networks. Consequently, the proposed method is useful to construct efficient and robust networks in a realistic scenario.
A novel adaptive power control and beam-forming joint optimization algorithm is proposed in cognitive radio （CR） underlay networks, where cognitive network share spectrum with primary network which spectrum is licensed. In this paper, both primary base station （PBS） and cognitive base station （CBS） are all equipped with multi antennas, while each primary user （PU） and cognitive user （CU） has only one antenna. Different from traditional algorithms, an adaptive weight factor generating solution is supplied to different access users （both PUs and CUs） in this paper, and the different priority of users is also considered, because PUs have higher priority, the weight factor of PUs is fixed as constant and signal-to-interference and noise ratio （SINR） threshold is unchanged, while for CUs, it is set adaptively and SINR threshold is also changed accordingly. Using this algorithm, the transmit power is decreased, which relax the strict requirements for power amplifier in communication systems. And moreover, owing to PUS has fixed SINR threshold, the calculated SINR at receiver is nearly unchanged, but for CUs, the S1NR is changing with the adaptive weight factor. Under the assurance of quality of service （QoS） of PUs, the solution in this paper can enable CRs access to the CR network according to adaptive SINR threshold, therefore which supplies higher spectrum utilization efficiency.
This paper proposes a joint nonlinear transceiver design scheme based on minimum mean square error （MMSE） criterion for non-regenerative multiple input multiple output （MIMO） relay system. The proposed scheme decomposes the error covariance matrix, reformulates the original joint design problem as two separate optimization problems, and then provides a closed-form solution with only local channel state information （CSI） available at the source and destination. Performance evaluation shows that the proposed scheme significantly outperforms linear schemes, and has a competitive performance compared with existing global CSI based nonlinear schemes, both iterative and non-iterative.
