Wideband spectrum sensing using sub–Nyquist sampling

Abstract

Spectrum sensing is the process of identifying the frequencies of a spectrum in which Signals Of Interest (SOI) are present. In case of continuous time signals present in a wideband spectrum, the information rate is seen to be much less than that suggested by its bandwidth and are therefore known as sparse signals. A review of the literature in [1] and [2] indicates that two of the many techniques used in wideband spectrum sensing of sparse signals are the Wideband Compressive Radio Receiver (WCRR) for multitoned signals and the mixed analog digital system for multiband signals. In both of these techniques even though the signals are sampled at sub-Nyquist rates using Compressive Sampling (CS), the recovery algorithms used by them are different from that of CS. In WCRR, a simple correlation function is used for the detection of carrier frequencies and in a mixed analog digital system, a simple digital algorithm is used for the identification of frequency support. Through a literature survey, we could identify that a VHSIC hardware descriptive ModelSim simulation model for wideband spectrum sensing of multitoned and multiband signals using sub Nyquist sampling does not exist. If a ModelSim simulation model can be developed using VHDL codes, it can be easily adapted for FPGA implementation leading to the development of a realistic hardware prototype for use in Cognitive Radio (CR) communication systems. The research work reported through this dissertation deals with the implementation of simulation models of WCRR and mixed analog digital system in ModelSim by making use of VHDL coding. Algorithms corresponding to different blocks contained in the conceptual design of these models have been formulated prior to the coding phase. After the coding phase, analyses of the models are performed using test parameter choices to ensure that they meet the design requirements. Different parametric choices are then assigned for the parametric study and a sufficient number of iterations of these simulations were carried out to verify and validate these models.