Advanced Markov Model Based Spectrum Sensing In UAV Networks

With the advantage of high flexibility and low deployment cost,unmanned aerial vehicles(UAVs)have been widely used in military communications,weather monitoring,cargo transportations,and emergency rescue.The large-scale deployment of UAVs has exacerbated the shortage of spectrum resources.However,the existing spectrum allocation strategies cannot effectively use the scarce spectrum resources,which becomes the bottleneck for enhancing the communications performance of UAVs.Cognitive UAV networks is proposed to solve the problem,which improves the spectrum efficiency by perceiving spectrum holes and providing secondary UAVs with opportunities to reuse idle spectrum.Most of the existing research are based on multi-user information exchange or spatial information exchange,and they cannot attach good sensing performance.Exploring the temporal correlation of spectrum states by focusing on the Markov property is an effective and novel idea to enhance the detection probability,and thus this thesis studies the advanced Markov model-based spectrum sensing methods in UAV networks.Firstly,considering the temporal dimension information of the spectrum,a continuous hidden Markov model(CHMM)based sensing scheme for cognitive UAV networks with prefect SNR estimation is proposed.Note that the true states of the unauthorized spectrum are not observable,but the sensing results originated from the unobservable spectrum states can be easily obtained.Therefore,hidden Markov chain model fits well with the spectrum state.To fully exploit the temporal information and avoid loss of the information,we combine spectrum sensing scheme with CHMM,and then get a preliminary prediction.This prediction can further guide the sensing decision procedure to further enhance the detection performance.Secondly,in view of the challenge of obtaining the signal-to-noise ratio in advance caused by the dynamics and the sudden tasks of UAV networks,a CHMM-based sensing scheme with a novel space smoothing second-and fourth-order moments(SS-M2M4)SNR estimator is proposed.The knowledge of SNR is required before typical commonly-used spectrum sensing.But large flying area,unstable links,and dynamic network topology lead to the variable SNRs in CUAVNs,which makes the assumption of constant SNRs no longer applicable and brings new challenges for sensing in UAV networks.Considering imperfect SNR estimation in practical applications,based on a universal SNR estimation algorithm and the clustering results,we design a novel SNR estimation scheme which is inspired by the reconstruction of signal on the graph to enhance the proposed CHMM-based sensing scheme with practical SNR estimation.Thirdly,considering the homogeneity and observation independence assumption of the hidden Markov model,a linear chain conditional random field(Linear-Chain CRF)based spectrum sensing method is proposed.Linear-Chain CRF is used to model the cooperative spectrum sensing problem of UAV networks.The sequence of energy detection value and cyclic spectrum statistics value of each UAV is modeled as input,and the real state of spectrum is modeled as output,which can make the modeling results of spectrum sequence and observation sequence closer to reality.Further,more accurate prediction results of spectrum state can be obtained,so as to improve the detection probability of spectrum.Simulation results demonstrate the proposed CHMM-based spectrum sensing scheme outperforms the one without CHMM,and the CHMMbased sensing scheme with the proposed SNR estimator can outperform the existing algorithm considerably.Besides,the proposed liner-chain conditional random filed based spectrum sensing scheme can further enhance the detection performance.Therefore,this research can effectively improve the spectrum sensing performance of UAV networks and help the vigorous development of UAV industry.