A low-overhead real-time spectrogram construction method based on learning from historical decision data

Abstract

The invention relates to a low-overhead real-time spectrum map construction method based on historical decision data learning. The method comprises the following steps: step S1, loading the history decision data of target spectrum availability obtained by all sensing nodes in the collected large-scale cognitive radio network through a single-point spectrum sensing algorithm; step S2, loading the history decision data of target spectrum availability obtained by all sensing nodes in the large-scale cognitive radio network; Step S2, using the least square support vector machine to analyze the loaded historical decision data to determine the spectrum state boundary to construct the spectrum map; Step S3: determining the support vector in the historical decision data by the boundary user searchalgorithm to optimize the spectrum map. Compared with the prior art, the method screens secondary users which are farther from the spectrum state boundary through the boundary user search algorithm,and these secondary users do not need to participate in detecting the target frequency band, thereby reducing the energy and time overhead of the spectrum map construction.

Description

technical field [0001] The invention relates to a method for constructing a spectrogram, in particular to a method for constructing a real-time spectrogram with low overhead based on historical decision data learning. Background technique [0002] Interference occurs frequently in today's wireless communication environment, and there are a large number of public network communication interruption problems caused by radio interference. Cognitive radio technology (CR) can effectively solve these problems. Cognitive radio is defined as an intelligent wireless communication technology based on software defined radio (SDR). It can continuously perceive the surrounding radio environment, analyze it, make decisions, dynamically select idle frequencies for communication, and then use Radio Knowledge Representation Language (RKRL) to adaptively modify the communication parameters inside the device. [0003] Cognitive radio network consists of two types of users, primary users (PU) ...

AI Technical Summary

Problems solved by technology

[0004] The current spectrum sensing does not make full use of historical data, and there is still a lot of room for research. Introducing machine learning algorithms into the field of cognitive radio is also an effective way to improve the performance of cognitive radio networks.

In a large-scale cognitive radio network, due to the large amount of historical data and the large amount of redundant information, the time complexity of building a spectrum map using existing technologies is relatively high

Images