A Spectrum Detection Method in Time-varying Fading Channel

Abstract

The present invention proposes a spectrum sensing algorithm under a time-varying fading channel, designs a dynamic state space model to describe the time-varying characteristics of the authorized user state and the time-varying channel gain, and regards the authorized user working state and the fading channel gain as two The state of the system is hidden, and the first-order FSMC model is introduced to describe the time-varying migration characteristics of the time-slow fading channel gain, and the energy accumulation sum of the received signal is used as the system observation value; on this basis, a new spectrum sensing method is proposed for the time-varying fading channel , fully explore the prior probability of the authorized user state and the state transition characteristics of the fading channel, and jointly estimate the authorized user state and time-varying channel gain, which greatly improves the performance of spectrum sensing in the fading channel, and does not need to implement complex multi-node cooperative sensing. Under the condition, good detection performance can also be obtained (see attached figure). The scheme uses accumulated energy as the observation quantity, and the new method also retains the advantages of not requiring authorized user signal characteristics and short detection time.

Description

technical field [0001] Aiming at the spectrum detection problem under time-varying fading channel, the present invention designs and proposes a dynamic state-space system model (Dynamic State-space Model, DSM), which is based on two-state Markov state probability transition model and first-order finite state Markov respectively The channel (Finite-state Markov Channel, FSMC) model uses the authorized user (also known as the primary user) state and the time-varying channel state as the hidden state (Hidden States); on this basis, a joint estimation of channel gain and authorized A new spectrum sensing method for user state, and use particle filter (ParticleFiltering, PF) technology to estimate the posterior probability involved in the non-stationary system, so as to realize the joint estimation of time-varying fading channel gain and primary user state. belongs to the field of communication. Background technique [0002] With the rapid development of modern wireless communic...

AI Technical Summary

Problems solved by technology

Unfortunately, energy detection is easily affected by the channel state. Specifically, the decision threshold in ED is closely related to the energy of the received signal. Therefore, in a time-varying fading channel, the characteristics of channel gain changing with time will undoubtedly greatly affect Increases detection difficulty and significantly degrades spectrum detection performance in practical applications

Although existing studies have considered the statistical probability distribution density characteristics of time-varying fading channels, the statistical probability distribution can only reflect the instantaneous random characteristics of fading channels, and cannot describe its properties that change with time migration, so it is difficult to analyze the time-varying channel Descriptive and traceable, leading to suboptimal ED detection performance

Images