AHSWN Home • Issue Contents • Forthcoming Papers
Spectrum Sensing by Correlation Modulation Format Identifier and Improved Energy Detection Using Noise Estimation Based Dynamic Threshold
Appala Raju Uppala, C Venkata Narasimhulu and K Satya Prasad
Cognitive Radio (CR) systems are being implemented recently to tackle the spectrum underutilization problems and aids efficient data traffic. Spectrum sensing is the crucial step in cognitive applications in which cognitive user detects the presence of PU in a particular channel, thereby switching to another channel for continuing transmission. Though a lot of techniques are available in the literature, their application in Industrial Wireless Sensor and Actuator Networks (IWSANs) is limited due to its complexity and accuracy constraints. IWSANs demand less complex detection techniques without compromising efficiency and determinism. Hence, in this paper, we propose a spectrum sensing algorithm for Industrial, Scientific, and Medical (ISM) band where autocorrelation characteristics as well as the noise of received signal are used to identify the presence of Primary User (PU) in the channel. Our hypothesis is based on the fact that the presence of modulation format itself specifies the PU presence in the channel. Four Modulation techniques viz Orthogonal Frequency Division Multiplexing (OFDM), Amplitude Shift Keying (ASK), Quadrature Phase Shift Keying (QPSK), and Gaussian Frequency Shift Keying (GFSK) were considered for the experiments. For signals with other modulation formats and low Signal to Noise Ratio (SNR), an Improved Energy Detection (IED) method is proposed where the dynamic threshold is selected based on noise power estimation. Simulations were carried out in an Integrated Development Environment (IDE) under IQ imbalance, DC and frequency offsets, and Gaussian noise. The results indicate that the proposed algorithm has a high probability of detection under varying SNRs. This also proves the suitability of the approach for implementation in IWSANs.
Keywords: Cognitive Radio, Industrial Networks, Spectrum Sensing, Cyclo-stationary Characteristics, Improved Energy Detection, Noise Power Estimation, Probability of Detection