Compressed broadband spectrum blind detection method based on high-order-statistic

Abstract

The invention discloses a compressed broadband spectrum blind detection method based on a high-order-statistic. The method comprises the steps that: the spectrum of signals of a primary user PU is divided into a plurality of non-overlapped sub-bands; the node number of secondary users is set as K, then the kth secondary user SUk carries out compressed sampling on signals in any sub-band formed by division in the step 1, compressed sampling data observed by the kth secondary user SUk is obtained, and the compressed sampling data follows the gauss distribution; an estimated value of a three-order matrix of the compressed sampling data is calculated and output to a fusion center FC; the fusion center FC carries out statistical averaging according to the received estimated value of the three-order matrix of the compressed sampling data and obtains a statistical decision parameter d; a decision threshold [lambda] is calculated according to the compressed sampling data observed in the step 2 and a preset false alarm probability; and the fusion center FC judges whether the sub-band is occupied by the PU according to decision rules. According to the invention, the high-order statistic is used as the spectrum detection decision parameter, and the detection performance under a low signal to noise ratio is improved.

Description

technical field [0001] The invention relates to the field of cognitive radio technology, in particular to a blind detection method of compressed wideband spectrum based on high-order statistics. Background technique [0002] With the dramatic growth of new wireless networks and services, the shortcomings of today's fixed spectrum allocation policies are becoming more and more apparent. This is mainly manifested in the scarcity of spectrum for new users, while the utilization rate of spectrum for authorized primary users (Primary User, PU) is not high. In order to change this situation and improve spectrum utilization, Mitola proposes a spectrum reuse policy, which allows cognitive users (or secondary users) (SecondaryUser, SU) to reuse spectrum not occupied by PUs. This triggered a surge of research enthusiasm for spectrum reuse, spectrum sharing, and dynamic spectrum access in the next few years. [0003] Cognitive Radio (CR) technology has been widely considered as a ver...

AI Technical Summary

Problems solved by technology

[0005] In many existing spectrum sensing schemes, it is necessary to reconstruct the original signal before realizing spectrum sensing, which greatly increases the complexity of the algorithm

There are also compressed spectrum sensing methods that directly use compressed observations for non-reconstruction, but these methods still use a spectrum detection scheme similar to energy detection (Energy Detection, ED). When the noise fluctuates and the signal-to-noise ratio is low, it will leading to a sharp deterioration in detection performance

Images