Combined authorized user perception and link state estimation method and device

Abstract

The invention provides a combined authorized user perception and link state estimation method and device in order to resolve the problem of spectrum sensing in the cognitive radio technology. The method takes link state information between a main user and a secondary user into account, and takes the link state information as another hidden state in spectrum sensing, and then a universal dynamic state space model is built. A Bernoulli smoothing mechanism is designed based on the Bayesian sequential estimation framework and the Bernoulli random finite set theory, and combined blind estimation of the state information of a main link and a secondary link can be obtained while the work state of an authorized frequency band is perceived and detected (the algorithm process is shown in the picture). The combined authorized user perception and link state estimation method makes full use of the state information of the main link and the secondary link, thereby obviously improving the performance of spectrum sensing. Meanwhile, the obtained state information of the main link and the secondary link can be used for subsequent resource optimization. In addition, the combined authorized user perception and link state estimation method and device have universality and can be expanded to other state information of the main link and the secondary link.

Description

technical field [0001] Aiming at the problem of licensed spectrum sensing in cognitive radios (Cognitive Radios, CRs), the present invention designs and proposes a new joint licensed spectrum sensing and link state estimation method, so it can be called deep sensing (Deep sensing). While detecting authorized users, it also takes into account the link status information between unknown authorized users and secondary users, and establishes a dynamic state-space model (Dynamic State-space Model, DSM); the above-mentioned authorized user perception and joint The link state estimation problem is modeled as a Bernoulli Random Finite State (BRFS), and a Bernoulli filtering mechanism is designed to realize the PS-LSI The joint blind estimate of . The method and device have sufficient scalability and can handle any PS-LSI (such as time-varying channel gain). belongs to the field of communication. Background technique [0002] Cognitive radio technology dynamically accesses the all...

AI Technical Summary

Problems solved by technology

However, in practice, the unknown PS-LSI will increase the received signal uncertainty, which in turn will negatively affect the spectrum sensing performance

Especially considering the actual time-varying fading channel, the dynamic channel gain that changes with time will greatly increase the difficulty of detection, resulting in poor spectrum sensing performance

Although the existing literature considers the statistical probability distribution function (Probability Density Function, PDF) of channel fading, and reduces the negative impact of random fading to a certain extent through the idea of ​​marginalization [29]; however, the statistical probability distribution can only reflect the fading The instantaneous random behavior of the channel fails to describe the relevant time-varying migration properties, so it is impossible to track the time-varying channel in real time, making its detection performance still unacceptable

Images