An improved blind spectrum sensing scheme is established by the covariance matrix Cholesky decomposition and radial basis function (RBF)-support vector machine (SVM) decision classification at low signal-to-noise ratios (SNRs). Under strong background noises, the proposed scheme improves the recognition rate of primary users (PUs) than that of the current blind spectrum sensing. First, the ratio of the maximum-to-minimum eigenvalue of a covariance matrix obtained by the Cholesky decomposition is used to construct the statistics. Second, the statistics are labeled with “+1” or “-1,” namely, the energy characteristics of the training samples are extracted and marked with “+1” for PUs and “-1” for noises. Finally, an RBF-SVM classification model, with an intelligent RBF as the SVM kernel function, is obtained by training the above-mentioned statistics and the labels. Thus, the received signals are classified as PUs or not be trained in the SVM model. The threshold possesses self-learning ability, and it distinguishes PU signals from noises effectively. The classification among PU signals and noises is implemented by the optimal SVM decision boundary, derived from maximizing the margin of the decision boundary of trained samples for efficient detection. In addition, the complexity of the statistic construction is lower than that of the conventional maximum minimum eigenvalue (MME). The simulation results show that the RBF in our scheme has 77.5% accuracy at -10 dB, and it outperforms linear kernel function significantly by about 27.5% in accuracy at -10 dB. In addition, the average error probability of the proposed scheme is reduced by about 26% when compared with those of original SVM schemes at -20 dB. The proposed scheme also outperforms the current MME detection in detection probability over 10% at -20 dB. Therefore, the proposed blind spectrum sensing scheme can be efficiently used to detect the PUs by the covariance matrix Cholesky decomposition and the RBF-SVM decision classification in the fifth-generation (5G) communications, especially at low SNRs.

In view of professional learning and practical training in traditional electronic information education of application-oriented universities, this paper constructs electronic information–innovation practice teaching reform (EI-IPTR).In this scheme, by an integrating artificial intelligence (AI)-enabled curriculum with a multi-mode integrated platform and open-style module, big data-based comprehensive education resources are optimally configured. We jointly perform the multi-mode construction of innovative practice teaching, professional education stage design, and teaching management improvement, respectively. Subsequently, new practice teaching mechanisms with information technology and its implementation and management methods are established to achieve better teaching effects. It first strengthens learning and intra-group competition to promote students’ innovation in competitions. Then, the AI technique, i.e., attention mechanism-aided long short-term memory (LSTM), is used to model individual students’ abilities. Thus, it accurately evaluates them for teachers to efficiently manage their teaching process in accordance with their aptitude. The teaching reform practice verifies that the AI-enabled big data optimization of teaching reform has a better effect by the above multi-mode innovation. It exhibits an obvious improvement in the quantity and quality of students’ professional knowledge, personal ability, teamwork, and innovative practice. It is also in accordance with the independent completion of practical course teaching in the analysis of big education data. In addition, it realizes high-quality practical teaching by combining multi-mode, multi-level, and open discipline foundations together with efficient, professional skills.

A novel piecewise normalized bistable stochastic resonance (PNBSR) strengthened cooperative spectrum sensing is established by the PNBSR, residual covariance matrices, credibility weighted matrix fusion and a convolutional neural network (CNN) classification at low signal-to-noise ratios (SNRs). First, the PNBSR based on the traditional bistable stochastic resonance (TBSR) is proposed to improve the SNRs of received signals. Second, the output of the PNBSR is demodulated to obtain in-phase (I) and quadrature-phase (Q) covariance matrices. Third, the I/Q covariance matrices from different secondary users (SUs) are Cholesky decomposed to construct residual covariance matrices in a fusion center (FC). Subsequently, a new credibility weighted coefficient is proposed to fuse residual covariance matrices of the SUs. Finally, both training and test samples of the fusion detection statistics are fed into the CNN to train a high-performance classification model of cooperative spectrum sensing. The main innovations include the PNBSR, optimization index, Cholesky decomposition-based matrix cancellation to construct residual covariance matrices and credibility weighted matrix fusion. Simulation results show that the SNR of received signals strengthened by the PNBSR is improved by 3.36 dB other than received signals on average, which is 0.22dB larger than that of the TBSR. The detection probability of the proposed scheme also outperforms those of the support vector machine (SVM) and CNN schemes by 77% and 75% at -15 dB, respectively.

A cooperative blind spectrum detection with the Doolittle decomposition and support vector machine (SVM) classification is proposed to improve the performance and reduce the complexity and latency of spectrum detection in hybrid Geostationary Earth Orbit and Low Earth Orbit (GEO-LEO) satellite constellation networks. First, a principal component analysis (PCA) algorithm is adopted to process perceive signals. Second, the maximum-to-minimum eigenvalue ratio of a covariance matrix is utilized to construct statistics with self-learning ability by using the Doolittle decomposition. Third, primary user and noise signals are labeled with “+1” and “$-$1” as training labels, respectively. Finally, a high-performance PCA-SVM classification is generated by training the labels and the corresponding statistics. Innovations mainly include the Doolittle decomposition of the covariance matrix in the SVM and the PCA to preprocess the perceived signals. Simulation results show that the detection probability of the proposed scheme outperforms the energy detection and the SVM algorithm by 55 and 15% at a signal-to-noise ratio (SNR) of $-$15 dB, respectively. The average error probability of the proposed scheme is 20% lower than that of the SVM at an SNR of $-$20 dB. Therefore, in hybrid GEO-LEO satellite constellation networks, the cooperative blind spectrum detection with the Doolittle decomposition and SVM classification can detect the spectrum with low complexity and high performance effectively.

An incremental selection hybrid decode-amplify forward (ISHDAF) scheme for the two-hop single relay systems and a relay selection strategy based on the hybrid decode-amplify-and-forward (HDAF) scheme for the multirelay systems are proposed along with an optimized power allocation for the Internet of Thing (IoT). Given total power as the constraint and outage probability as an objective function, the proposed scheme possesses good power efficiency better than the equal power allocation. By the ISHDAF scheme and HDAF relay selection strategy, an optimized power allocation for both the source and relay nodes is obtained, as well as an effective reduction of outage probability. In addition, the optimal relay location for maximizing the gain of the proposed algorithm is also investigated and designed. Simulation results show that, in both single relay and multirelay selection systems, some outage probability gains by the proposed scheme can be obtained. In the comparison of the optimized power allocation scheme with the equal power allocation one, nearly 0.1695 gains are obtained in the ISHDAF single relay network at a total power of 2 dB, and about 0.083 gains are obtained in the HDAF relay selection system with 2 relays at a total power of 2 dB.

To solve low-resource utilization, complex storage, and busy traffic in space–air–ground integrated networks (SAGINs), a new multilayered decentralized coded caching (ML-DCC) scheme is proposed in hierarchical networks with nonuniform file popularity and multilevel cache capacity. First, a file popularity prediction is performed by a random forest model with grid search. Second, a multipopularity coded caching (MPCC) strategy by grouping files is executed in a two-hop network with fixed cache size. Finally, a new ML-DCC scheme is proposed to adopt nonuniform file popularity and multilevel cache capacity with both block-coded caching and zero-bit padding to obtain high-resource utilization and efficient file exchange in file transmissions. The innovations are random forest classifier with bagging integration and grid search to improve network traffic and link load, decentralized coded caching to reduce average transferred files, and coded caching and grouping files by popularity to improve network load. Simulation results show that the data payload of the proposed scheme is significantly improved by about 1.88, 1.18, 1.38, 3.53, and 4.39 times, when compared with those of the shared cache, multilevel popularity, hierarchical coded caching, MPCC, and uncoded caching schemes, respectively. Under the expected load of $R=211.57F$ bits of the shared link, the cache size used by the proposed ML-DCC is only 1/42.85 and 1/24.39 of those of the highest popularity first (HPF) and nonuniform cache schemes, respectively.