Research On Spectrum Management Technologies For Multiple Sensing Users

The deepening of users’ business needs for wireless communication technologies is being driven by the convergence of new business models such as digital twin and interconnection of all things.As the carrier for information transmission tasks,spectrum resources have been the key cornerstone for the sustainability of emerging communication technologies.However,the task of more efficient information transmission can hardly be met by the current wireless spectrum technology.The traditional spectrum management strategy has been unable to break through the bottleneck of communication data congestion and intense business competition.And the upper limit of spectrum resources exploitation is implied by the newly proposed spread spectrum technologies such as terahertz and millimeter wave.Therefore,finding an intelligent spectrum management technology that essentially improves the utilization of spectrum resources is a key way to solve the above problems.The ultimate goal of cognitive radio is to achieve reasonable dynamic spectrum management,and through real-time monitoring of the environmental state information of spectrum resources,to deeply explore and fully utilize the unused spectrum space,and to design spectrum sensing,allocation and sharing strategies and methods with high accuracy and high benefit indexes for communication users,so as to significantly improve spectrum utilization and further solve the problem of scarcity of spectrum resources that exceeds demand.However,there are still many challenges in this technology,i.e.,how to balance the computational complexity and discriminating accuracy to improve the sensing performance of the system while reducing the noise and interference for communication,and how to set reasonable spectrum allocation guidelines and reliable spectrum access timing to realize spectrum sharing among multiple users according to the communication needs of different users.Based on the above problems,this thesis conducts research on the energy detection methods and dynamic spectrum management techniques in cognitive radio.The main works and contributions are as follows.(1)In order to solve the problem that traditional energy detection methods in the field of spectrum sensing are highly susceptible to interference from low signal-to-noise ratio environment,and neglecting the localization of available spectrum also affects the discriminating results of the spectrum state.The adaptive energy detection method with joint wavelet-frequency domain transform is proposed to improve the noise sensitivity and discriminating accuracy of energy detection.The discrete wavelet packet transform is used to decompose the signal and calculate the sub-band energy;the energy norm is combined to reduce the computational complexity of the adaptive threshold for comparison with the sub-band energy;the fast Fourier transform is used to locate the available spectral range.The method is also simulated to explore the variation relationship between the adaptive threshold and different performance parameters.The simulation results show that the method has good environmental adaptability and system stability,while the detection error is smaller in different signal-to-noise environments.In addition,the frequency domain analysis of sub-band signals to achieve the reordering of normalized frequency range further improves the accuracy of spectrum sensing.(2)Aiming at the spectrum allocation method with energy detection as the premise,which is highly sensitive to the noise environment and the incompletely occupied primary user spectrum is not fully utilized.A spectrum management scheme driven by wavelet packet energy is proposed,which aims to set reasonable spectrum allocation criteria while weighing the computational complexity.The signal energy and adaptive threshold are calculated by discrete wavelet packet transform and the energy norm to discriminate the spectrum state of primary users;a user energy matching mechanism is established to screen the available spectrum space and complete the spectrum sharing between primary and secondary users.In addition,the performance of the spectrum management scheme is solved and analyzed from the energy domain,which is found to have relatively higher user matching and scene compatibility.And the method is simulated to explore the variation between channel occupancy,secondary users allocation rate and channel access rate and different performance parameters.The simulation results show that the scheme can adaptively achieve efficient utilization of the normalized spectrum under different conditions such as SNR and the number of wavelet packet decomposition layers.