Polarized Antenna Aided Spectrum Sensing Based on Stochastic Resonance
- PDF / 1,140,429 Bytes
- 12 Pages / 439.37 x 666.142 pts Page_size
- 61 Downloads / 179 Views
Polarized Antenna Aided Spectrum Sensing Based on Stochastic Resonance Jin Lu1 · Ming Huang1 · Jingjing Yang1 · Peng Li1
© Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Wireless communications is one of the most rapidly developing segments of the telecommunications industry. A large amount of intelligent terminal occupying the spectrum results in the reduction of radio spectrum resources. Cognitive radio is considered to be the most effective approach to solve this problem, which required rapid and exact spectrum sensing. This paper proposes a novel polarized antenna method based on likelihood ratio test and stochastic resonance. In the condition of adiabatic approximation, the stochastic resonance can increase signal to noise ratio, and adequately transfer the energy of noise to original signal. The proposed method applies the stochastic resonance to each polarized component. The experiments show that the proposed spectrum sensing method is suitable for generalized likelihood ratio test in additional white Gussion noise and lower signal to noise ratio, rather than polarized channel matrix and idealistic likelihood ratio test. Keywords Spectrum sensing · Stochastic resonance · Polarized antenna
1 Introduction Cognitive radio (CR) allows secondary users (SUs) to access the primary users’ (PUs) authorization spectrum hole, which is the effective method to solve the problem of spectrum shortage [1]. In which, spectrum sensing is an important technology. It can be used not only for the detection of spectrum holes but also for the detection of abnormal radio signals in spectrum management [2]. However, detecting weak signal in the low signal to noise ratio (SNR) is still a challenge. IEEE 802.22 requires that the detection probability should achieve 0.9 when the SNR is − 20 dB and false alarm probability is 0.1 [3]. The current spectrum sensing algorithm mainly include the classical energy detection (ED), matched filter detection (MFD), covariance detection and cyclostationary detection [4]. These algorithms can hardly extract signal character in lower SNR. The spectrum sensing method based on stochastic resonance (SR) achieved good effects in the weak signal condition. SR is a nonlinear phenomenon, which can enhance the output signal by some nonlinear systems under certain conditions. In which, The * Jin Lu [email protected] 1
School of Information Science and Engineering of Yunnan University, Kunming, Yunnan, China
13
Vol.:(0123456789)
J. Lu et al.
bi-stable SR (BSR) system has been extensively studied. BSR is applied to ED spectrum sensing with fixed system parameters in [5]. Further, BSR spectrum sensing method based on adaptive parameter adjustment is proposed in [6]. Cooperative sequence spectrum sensing based on suprathreshold BSR is proposed, but can only be applied to non Gaussion noise environment [7]. Recently, particle swarm algorithm is applied to optimize parameters of spectrum sensing based on tri-stable SR in [8]. Other SR system, e.g., optimal dynamic o
Data Loading...
