Spectrum sensing method and system based on gating circulation unit

Abstract

The invention discloses a novel spectrum sensing method and a novel spectrum sensing system, which can realize high-performance and low-consumption spectrum sensing on a frequency band of a higher frequency point at a lower sampling rate. The sampling sequence is subjected to absolute value averaging through window sliding, so that the input dimension is reduced, and the calculation amount is reduced; the network model adopts a gating cycle unit (GRU), so that the relatively high performance can be achieved, the parameters are fewer, and the complexity is lower; a spectrum sensing problem is seen as a regression problem, a multi-threshold thought is applied, distance threshold values participating in judgment are set according to different accuracy and sensing quantity requirements, and whether a frequency band is occupied or not is judged according to the distance output to 0 and 1; an adaptive thought is quoted, and different sensing speeds are adjusted and selected according to the fuzzy degree of the output probability so as to ensure the sensing accuracy and sufficient preparation number and reduce unnecessary resource overhead. Therefore, the resource overhead can be reduced while the sensing accuracy is ensured, and the method has certain adaptability in different scenes.

Description

technical field [0001] The present invention relates to the technical field of spectrum sensing, in particular to a spectrum sensing method and system based on a gated cycle unit. Background technique [0002] Due to the increasing number of wireless communication devices, limited wireless spectrum resources are becoming increasingly scarce. Cognitive radio technology enables different users to use the same spectrum resources, that is, spectrum sharing. Spectrum sensing is to find out idle spectrum resources called "spectrum holes" in multiple dimensions such as time domain, frequency domain and air domain as accurately as possible in a complex and changeable electromagnetic environment for allocation and utilization. The performance of spectrum sensing detection based on machine learning is good, and it has been widely researched at present. The existing spectrum sensing method based on (Long Short-Term Memory, LSTM) benefits from the natural advantages of the recurrent n...

AI Technical Summary

Problems solved by technology

The existing (Long Short-Term Memory, LSTM) based spectrum sensing method benefits from the natural advantages of the cyclic neural network model in processing time series, and has achieved good performance, but the complexity is high

Images