No abstract

A novel goodness-of-fit-based non-parametric spectrum sensing scheme in a non-Gaussian noise environment, modelled by Middleton class A distribution, is proposed. The sampling distribution of the proposed test statistic is derived and the detection performance is shown using Monte Carlo simulations. Results are presented and it is concluded that the performance is degraded if the Gaussian component in the Middleton noise is higher than the non-Gaussian component.Introduction: Cognitive radio is a key enabling technology to improve the spectrum efficiency for next-generation wireless networks [1]. To achieve this, many spectrum-sensing algorithms are proposed [2], in which energy detection is the simplest one and a more realistic scheme for spectrum sensing in cognitive radio. However, it has several demerits such as high sensing time and poor performance at low signal-to-noise ratios (SNRs). Therefore, detection of a primary user (PU) signal for low false alarm probability with a lower number of received observations at low SNR is a challenging problem.Recently, some goodness-of-fit (GoF)-based sensing schemes have been proposed in the category of non-parametric sensing such as Anderson-Darling (AD) sensing [3], the Kolmogorov-Smirnov (KS) GoF test [4], order statistics-based sensing [5] and so on. All these schemes outperform energy detection in a Gaussian environment assuming thermal noise. However, in a real scenario, the environment is of non-Gaussian noise (NGN) due to narrowband interference, and a mixture of manmade and natural electromagnetic sources [6]. In this case, the resulting non-Gaussian environment is modelled by the Middleton class A interference model. This model has been derived by combining appropriate physical and statistical descriptions of actual noise environments [7]. Furthermore, the co-channel interference in cellular and wireless fidelity networks is also modelled using the Middleton class A model [8]. However, sensing schemes designed for additive Gaussian noise perform poorly in this NGN environment [9]. Therefore, in [10], an adaptive L p -norm-based non-parametric sensing scheme has been proposed under a NGN environment modelled by a Gaussian mixture model.In this Letter, we propose a GoF-based non-parametric spectrum sensing scheme in a NGN environment modelled by Middleton class A noise. We present the results and conclude that the performance is degraded if the Gaussian component in the Middleton noise is higher than the non-Gaussian component.

The authors consider a multiple input single output (MISO) system in which both transmit beamforming and antenna selection (AS) are implemented using delayed channel state information (CSI) at the transmitter (CSIT). The performance of the system has been analysed for three different AS schemes wherein two out of N antennas are selected at the transmitter. The authors have derived closed-form expressions for the probability density function of the received signal-to-noise, bit error rate and outage probability for each AS scheme considered. The expressions have been obtained as a function of the correlation between perfect CSI at the receiver (CSIR) and delayed CSIT. The authors also discuss some special cases and compare them with the results available in the literature.

No abstract

Spectral unmixing is an important problem for remotely sensed hyperspectral data exploitation. Automatic spectral unmixing can be viewed as a three-stage problem, where the first stage is subspace identification, the next one is endmember extraction, and the final one is abundance estimation. In this sequence, endmember extraction is the most challenging problem. Many researchers have attempted to extract endmembers from hyperspectral images using spectral information only. However, it is well known that the inclusion of spatial information can improve the endmember extraction task. In this paper, we introduce a new endmember extraction algorithm that exploits both spectral and spatial information. A main innovation of the proposed algorithm is that spatial information is exploited using entropy, while spectral information is exploited using convex set optimization. In the literature, none of the spatial-spectral algorithms has used entropy as spatial information. The inclusion of this entropy-based spatial information improves the accuracy of the endmember extraction process. The results obtained by the proposed algorithm are compared (using a variety of metrics) with those obtained by other state-of-the-art methods, using both synthetic and real datasets. Our experimental results demonstrate that the proposed algorithm outperforms many available algorithms.