Search

Due to stochastic wireless environment, the process of modulation classification has become a challenging task. Because of its powerful feature extraction ability and promising performance over the conventional schemes, deep learning (DL) models are employed to automatic modulation classification (AMC) problems. Most of the conventional models proposed are tested for the limited set of modulation schemes transmitted over additive white Gaussian noise (AWGN) channels without considering the effect of multipath fading and Doppler shift. The next-generation networks use adaptive and higher-order quadrature amplitude modulation (QAM) schemes for higher spectral efficiency. The classification accuracy of conventional DL-based AMC schemes drastically reduces, when different order QAM modulation schemes are accommodated. In this work, different scaling factors are selected for the generation of M -QAM frames. The combination of scaling factors, which maximize the classification accuracy is chosen. A convolutional neural network (CNN) with six stages is employed for AMC. The simulation results show that the classification accuracy of proposed scheme is higher than the conventional DL-based schemes under various signal-to-noise ratio (SNR) conditions. The proposed scheme shows at least 4% improvement in classification accuracy over the other DL-based schemes. [ABSTRACT FROM AUTHOR]

Summary: Fifth‐generation (5G) networks are expected to offer unprecedented amounts of spectral and energy efficiency. Spatial modulation (SM) techniques have the capability to achieve an appealing trade‐off between these conflicting design parameters. Fully generalized SM (FGSM) and fully quadrature SM (FQSM) are the recent high‐rate SM variants, where the spectral efficiency is linearly proportional to the number of transmit antennas. The transmit antenna selection schemes can efficiently improve the average bit error rate (ABER) performance of SM techniques. The main objective of this work is to investigate the employment of transmit antenna selection schemes to FGSM and FQSM. Initially, Euclidean distance optimized antenna selection (EDAS) scheme is employed to FGSM and FQSM. It offers superior performance than conventional FGSM/FQSM without transmit antenna selection at the cost of higher computational complexity. In order to reduce the complexity cost, four suboptimal schemes based on channel capacity, correlation and combination of them are proposed and employed to FGSM and FQSM. The suboptimal transmit antenna selection based on capacity and correlation (TAS‐A‐C) offers a minimum gain of ∼6 dB over conventional FGSM/FQSM with significantly lower complexity. In addition, the performance of all schemes is investigated for increased number of available transmit antennas. As the proposed schemes address the spectral and energy efficiency trade‐off effectively, while corroborating better ABER performance, they have the potential to become a competing candidate for next‐generation networks. [ABSTRACT FROM AUTHOR]

The network performance is an imperative factor for the customers to select a mobile network operator (MNO). The customers demand seamless mobility and services with minimal packet loss and ultra-low latency from the subscribed MNO. Device-to-Device (D2D) communication is one of the key enabling solutions of fifth generation (5G), which has the potential to enhance throughput, latency, packet loss rate (PLR) performances of the network. 5G is expected to support high mobility and smaller range heterogeneous cells. This leads to frequent handovers. The unessential handovers may cause wastage of network resources. The improper network selection may prompt extreme quality degradation. In this work, a three-stage fuzzy-logic-based handover necessity estimation and target selection scheme is proposed for general heterogeneous networks. The simulation results prove that PLR, number of handovers executed and throughput performances of the proposed scheme are superior than the conventional and fuzzy-based multi-attribute decision-making (MADM) schemes. Even though this scheme is demonstrated for D2D application, it can be extended for any heterogeneous network scenarios. [ABSTRACT FROM AUTHOR]

Spatial Modulation (SM) is an innovative digital modulation scheme, which is expected to be a competitive candidate for next generation networks. All the variants of SM exhibit poor performance under spatial correlation and line of sight (LOS) Rician channel conditions. To combat the adverse effects of spatial correlation, a new variant of SM designated as modified spatial modulation (MSM) is proposed for 8 × 8 multiple input multiple output configuration. MSM uses a unique and dynamic mapping, which activates either one antenna or two antennas at the transmitter. Optimum maximum likelihood (ML) detection at the receiver though give accurate results but leads to high computational complexity as it performs extensive search of all possible antennas and symbols. In order to reduce the computational complexity, signal vector based minimum mean square error (SVMMSE) detection scheme is employed for MSM scheme. Estimation of antenna indices and transmitted symbols is done for both single antenna active and double antenna active scenarios. Performance of SVMMSE detection scheme is also analyzed under spatial correlation and LOS Rician channel conditions. The promising performance of SVMMSE under spatially uncorrelated Rayleigh, Spatial correlation and LOS Rician channel conditions with low computational complexity justifies the suitability of the proposed scheme for 5G based compact battery driven wireless devices. [ABSTRACT FROM AUTHOR]

A major challenge for fifth generation (5G) wireless communication, which is expected to be deployed around 2020 is to provide 10 times higher spectral efficiency and energy efficiency than fourth generation (4G). Spatial modulation (SM) is a novel multiple input and multiple output (MIMO) technique, which provides an exceptional trade-off between spectral efficiency and energy efficiency. All the conventional SM variants do not perform well under spatially correlated channel conditions. Hence, they cannot be employed for indoor wireless local area network (WLAN) and compact wireless mobile applications, where the effect of spatial correlation is imminent. In this paper, a new SM variant, dynamic spatial modulation (DSM) is introduced for 4 × 4 and 8 × 8 MIMO configurations, which activates either one or two transmit antennas based on input bit pattern. DSM employs dynamic mapping, which maximizes the Euclidean distance between active spatial bits by choosing antenna subsets with maximum spatial separation to minimize the effect of spatial correlation. The average bit error rate (ABER) performance of DSM under uncorrelated and spatially correlated channel conditions is quantified in comparison with conventional redesigned spatial modulation (ReSM). Later, a low complexity, suboptimal, ordered weight minimum mean square error based conditional maximum likelihood search (OWMMSE-CML) detection scheme is proposed for the DSM system. Through extensive simulations under various channel conditions and complexity analysis, it is shown that proposed OWMMSE-CML detection scheme offers performance nearer to optimal ML and also outperforms suboptimal signal vector based minimum mean square error (SVMMSE) detection scheme with a significant reduction in computational complexity. [ABSTRACT FROM AUTHOR]

The next-generation of networks should have better spectrum and energy efficiency. These competing design parameters can be dealt with spatial modulation (SM). Transmit antenna selection (TAS) strategies can enhance the average bit error rate (ABER) performance of SM. The purpose of this research is to examine the employment of machine learning (ML)-based TAS schemes for SM. The conventional Euclidean distance-based antenna selection (EDAS) scheme offers optimal performance at the expense of computational complexity. This paper discusses the application of four different ML algorithms to SM-TAS: Support vector machines (SVM), K -nearest neighbour (KNN), decision tree (DT), and naïve Bayes (NB). With reduced complexity, an SVM-based TAS scheme for SM can offer a minimum gain of ∼ 3.4 dB over SM with no transmit antenna selection (SM-NTAS). Furthermore, the ABER performance of all schemes is tested with a higher modulation order and a number of transmit antennas. In every scenario, ML-based TAS schemes outperform SM-NTAS in terms of ABER performance. [ABSTRACT FROM AUTHOR]

Spatial Modulation techniques (SMTs) offer an exceptional trade-off between spectral and energy efficiencies. In this work, we propose a new SMT termed as permutation index-quadrature spatial modulation (PI-QSM), aiming for enhanced spectral efficiency and better average bit error rate (ABER) performance compared to rival SMTs. To enhance spectral efficiency, PI-QSM exploits permutations of active transmit antennas by conveying additional information bits innovatively through indices of active antenna permutations. PI-QSM also alleviates the major constraint in some of the SMTs, where the total number of transmit antennas need to be a power of two. At any time instance, PI-QSM transmits real and imaginary parts of two complex modulated symbols separately using four different active antennas. To avoid any ambiguities at the receiver and to improve the error performance of the overall system, a rotation angle is introduced between the two modulated symbols before transmission. The rotation angle is optimized for different M -quadrature amplitude modulation (QAM) constellations using extensive Monte Carlo simulations. Performance of PI-QSM scheme is analyzed by deriving closed-form expressions for upper bound and asymptotic average bit error probability (ABEP) over Rayleigh fading channels. The Simulation results substantiate the accuracy of analytical analysis and also indicate that the proposed PI-QSM scheme offers better ABER performance than rival SMTs such as QSM and generalized QSM (GQSM). [ABSTRACT FROM AUTHOR]