KNN and SVM based Optimal Adaptive Modulation Selection in Spatial Modulation

Orthogonal Frequency Division Multiplexing (OFDM) is a multicarrier transmission scheme. OFDM transmits data by using a large number of narrow bandwidth carriers. In an OFDM transmission system, each subcarrier is attenuated individually under the frequency-selective and fast fading channel if the same fixed transmission scheme is used for all OFDM subcarriers thus it results in highest attenuation and hence poor performance. The purpose of this paper is to introduce the adaptive modulation to get an understanding of the differences between fixed and adaptive modulations schemes. In this work adaptive modulation is implemented by dividing whole subcarriers into blocks of adjacent subcarriers. Based on calculated average instantaneous signal to noise (SNR) same modulation scheme is applied to all subcarriers of same block. Here average bit error rate (BER) performance of OFDM system under fixed modulation and adaptive modulation is observed. Average BER performance of these modulation techniques is observed with various inverse fast Fourier transform (IFFT) size and using simpler adaptive quadrature amplitude modulation (QAM) schemes. The simulation results show that BER performance of OFDM system using adaptive modulation is better than fixed modulation.

In this letter, an adaptive spatial modulation (ASM) transmission scheme is proposed to achieve better system performance under a fixed data rate. The proposed scheme is based on a developed modulation order selection criterion (MOSC), so as to minimize the conditioned pairwise error probability (PEP) for each channel realization. Furthermore, the special information-conveying mode of spatial modulation (SM) is utilized to reduce the complexity of the proposed algorithm. Simulation results show that the proposed adaptive SM scheme provides considerable system performance improvement compared to original SM system.

This paper proposes a novel machine learning (ML) assisted low‐latency low density parity check (LDPC) coded adaptive modulation (AM) system, where short block‐length LDPC codes are used. Conventional adaptive modulation and coding (AMC) system includes fixed look‐up table method, which is also called inner loop link adaptation (ILLA) and outer loop link adaptation (OLLA). For ILLA, the adaptive capability is achieved by switching the modulation and coding modes based on a look‐up table using signal‐to‐noise ratio (SNR) thresholds at the target bit error rate (BER), while OLLA builds upon the ILLA method by dynamically adjusting the SNR thresholds to further optimize the system performance. Although both improve the system overall throughput by switching between different transmission modes, there is still a gap to optimal performance as the BER is comparatively far away from the target BER. Machine learning (ML) is a promising solution in solving various classification problems. In this work, the supervised learning based k‐nearest neighbours (KNN) algorithm is invoked for choosing the optimum transmission mode based on the training data and the instantaneous SNR. This work focuses on the low‐latency communications scenarios, where short block‐length LDPC codes are utilized. On the other hand, given the short block‐length constraint, we propose to artificially generate the training data to train our ML assisted AMC scheme. The simulation results show that the proposed ML‐LDPC‐AMC scheme can achieve a higher throughput than the ILLA system while maintaining the target BER. Compared with OLLA, the proposed scheme can maintain the target BER while the OLLA fails to maintain the target BER when the block length is short. In addition, when considering the channel estimation errors, the performance of the proposed ML‐LDPC‐AMC maintains the target BER, while the ILLA system's BER performance can be higher than the target BER.

An underwater acoustic multimedia communication (UWAMC) system is proposed with 2400 transmission modes according to time-varying multipath underwater acoustic (UWA) channel conditions. The orthogonal variable spreading factor (OVSF) scheme and Gold sequence (GS) scramble code are integrated into multi-input multi-output UWAMC system based on orthogonal frequency-division multiplexing to achieve the quality of service of multimedia transmission in the UWA channel. Binary phase shift keying (BPSK) and quadrature phase shift keying (QPSK) adaptive modulation, direct mapping (DM) or space–time block code (STBC) transmission strategies, convolution channel code with rate 1/2 and 1/3, and a power assignment mechanism were adopted in the proposed system. Simulation results show that the bit error rate (BER) and power saving ratio (PSR) performance of the STBC strategy with transmission diversity is superior to that of the DM strategy without transmission diversity, and the performance of the BERs and PSRs of the transmission scheme with the GS scramble code surpasses that of the scheme without the code. The performance of the BERs and PSRs of BPSK modulation with a channel code rate of 1/3 is better than that of BPSK modulation with a channel code rate of 1/2, and the performances of BERs and PSRs of BPSK modulation with a channel code rate of 1/3 are better than that of QPSK modulation with a channel code rate of 1/3. As the length of the OVSF codes increases, the UWAMC system’s BERs decrease, and its PSRs increase. The UWAMC system can achieve either maximum transmission speed or maximum transmission power efficiency.

We can state undoubtedly that Bangla language is rich enough to work with and implement various Natural Language Processing (NLP) tasks. Though it needs proper attention, hardly NLP field has been explored with it. In this age of digitalization, large amount of Bangla news contents are generated in online platforms. Some of the contents are inappropriate for the children or aged people. With the motivation to filter out news contents easily, the aim of this work is to perform document level sentiment analysis (SA) on Bangla online news. In this respect, the dataset is created by collecting news from online Bangla newspaper archive. Further, the documents are manually annotated into positive and negative classes. Composite process technique of “Pipeline” class including Count Vectorizer, transformer (TF-IDF) and machine learning (ML) classifiers are employed to extract features and to train the dataset. Six supervised ML classifiers (i.e. Multinomial Naive Bayes (MNB), K-Nearest Neighbor (K-NN), Random Forest (RF), (C4.5) Decision Tree (DT), Logistic Regression (LR) and Linear Support Vector Machine (LSVM)) are used to analyze the best classifier for the proposed model. There has been very few works on SA of Bangla news. So, this work is a small attempt to contribute in this field. This model showed remarkable efficiency through better results in both the validation process of percentage split method and 10-fold cross validation. Among all six classifiers, RF has outperformed others by 99% accuracy. Even though LSVM has shown lowest accuracy of 80%, it is also considered as good output. However, this work has also exhibited surpassing outcome for recent and critical Bangla news indicating proper feature extraction to build up the model.

Gesture recognition is a topic in computer science and language technology that aims to interpret human gestures with computer programs and many different algorithms. It can be seen as the way computers can understand human body language. Today, the main interaction tools between computers and humans are still the keyboard and mouse. Gesture recognition can be used as a tool for communication with the machine and interaction without any mechanical device such as a keyboard or mouse. In this paper, we present the results of a comparison of eight different machine learning (ML) classifiers in the task of human hand gesture recognition and classification to explore how to efficiently implement one or more tested ML algorithms on an 8-bit AVR microcontroller for on-line human gesture recognition with the intention to gesturally control the mobile robot. The 8-bit AVR microcontrollers are still widely used in the industry, but due to their lack of computational power and limited memory, it is a challenging task to efficiently implement ML algorithms on them for on-line classification. Gestures were recorded by using inertial sensors, gyroscopes, and accelerometers placed at the wrist and index finger. One thousand and eight hundred (1800) hand gestures were recorded and labelled. Six important features were defined for the identification of nine different hand gestures using eight different machine learning classifiers: Decision Tree (DT), Random Forests (RF), Logistic Regression (LR), Linear Discriminant Analysis (LDA), Support Vector Machine (SVM) with linear kernel, Naïve Bayes classifier (NB), K-Nearest Neighbours (KNN), and Stochastic Gradient Descent (SGD). All tested algorithms were ranged according to Precision, Recall, and F1-score (abb.: P-R-F1). The best algorithms were SVM (P-R-F1: 0.9865, 0.9861, and 0.0863), and RF (P-R-F1: 0.9863, 0.9861, and 0.0862), but their main disadvantage is their unusability for on-line implementations in 8-bit AVR microcontrollers, as proven in the paper. The next best algorithms have had only slightly poorer performance than SVM and RF: KNN (P-R-F1: 0.9835, 0.9833, and 0.9834) and LR (P-R-F1: 0.9810, 0.9810, and 0.9810). Regarding the implementation on 8-bit microcontrollers, KNN has proven to be inadequate, like SVM and RF. However, the analysis for LR has proved that this classifier could be efficiently implemented on targeted microcontrollers. Having in mind its high F1-score (comparable to SVM, RF, and KNN), this leads to the conclusion that the LR is the most suitable classifier among tested for on-line applications in resource-constrained environments, such as embedded devices based on 8-bit AVR microcontrollers, due to its lower computational complexity in comparison with other tested algorithms.

Multicarrier transmission techniques like multicarrier code division multiple access (MC-CDMA) is the strongest contender for the future generation wireless networks which is supposed to cater to the need of very high data rate with mobility and at the same time combat multipath and intersymbol interference in a spectrally efficient manner. To enhance the spectral efficiency further, some form of adaptive modulation is being proposed (Wasantha et al. (2002)). In this paper, first of all, we simulate the performance of various single-h continuous phase modulation (CPM) schemes in an MC-CDMA system by varying the two CPM parameters - size of symbol alphabet (M) and modulation index (h). The parameter sets considered are {M=2, h=0.5}, {M=2, h=0.875}, {M=4, h=0.5} and {M=4, h=0.875}. Simulation results show that the higher order modulation schemes (M=4) need better channel conditions than the lower order schemes (M=2) for a given bit error rate (BER). Moreover, as expected (Xiong (2000)), h=0.875 (approximate optimum value) gives a better BER performance than h=0.5 for a given BER. Based on these simulation results, we then design two adaptive modulation based systems, one considering h=0.5 and the other with h=0.875, each with M=2 and 4. Results of the adaptive systems clearly indicate that the adaptive switching of the modulation formats in each of these two systems increases the system capacity per given bandwidth (spectrum efficiency) effectively without sacrificing the expected BER performance. Also, it is seen that the adaptive system with h=0.875 is superior to the h=0. 5 system, both in terms of BER performance and the number of simultaneous users, but obviously at the cost of reduced bandwidth efficiency as higher the value of h, the higher bandwidth it requires.

This paper offers a novel method for modeling and assessing network resilience against cyber-attacks by utilizing discrete structures and algebraic techniques. Utilizing edge and node representations taken from actual intrusion detection datasets build a network graph and examine its structural aspects to identify attack trends and resilience traits. The selection of key characteristics via correlation analysis, supervised machine learning classifiers— Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbors (KNN)—are used to predict and categorize attack types. Confusion matrices and common classification metrics, accuracy, precision-, recall, and F1-score-0.99, are used to assess the models. This research supports SDG 9 by advancing secure and resilient infrastructure through innovative use of algebra and discrete structures. It also aligns with SDG 16 Peace Justice and strong institutions by contributing to the development of secure institutions.

Freshwater ecosystems play a crucial role in maintaining ecological balance and supporting biodiversity. Assessing water quality and identifying pollution factors affected by seasonal changes is of utmost importance. The traditional approach for identifying pollution sources and seasonal water quality assessment is often labor-intensive, time-consuming, and complex. In response, our study introduces a hybrid receptor, multivariate statistical and machine learning (ML) based framework as an alternative. Positive matrix factorization (PMF) and receptor model were used to identify key pollutant sources, while Principal Component Analysis (PCA) was used to corroborate pollutant sources. Additionally, various supervised ML classification algorithms are used to predict and identify pollution sources on full physicochemical datasets and reduced in situ parameter scenarios. ML model stability and generalization capacity were evaluated using learning curves, cross-validation method. The findings indicate significant seasonal variability in water quality index, ranging from 38.5 (poor) to 78.6 (good). PMF and PCA revealed three dominant pollution factors, such as biogeochemical processes, sediment resuspension/nutrient influx, and anthropogenic pollution, affecting the seasonal water quality by 15-30%, 25-45%, and 40-70%, respectively. Model performance indicators using confusion matrices, ROC-AUC, and other evaluation metrics, for full dataset identified XGBoost as the best model, achieving the highest accuracy of 96.3%, followed by CatBoosting (93.6%), Random Forest (93.6%), Support Vector Classifier (91.8%), K-Nearest Neighbors (87.2%) and Decision Tree (84.3%). However, with the in situ data the ML models show stable and consistent classification accuracy in the range of approximately 70-81%. Overall, the modelling framework provides an effective approach for capturing seasonal variability and pollution source attribution, contributing to improved ecological understanding and predictive management of freshwater system.

By combing Spatial modulation (SM) with the Amplify-and-forward (AF) cooperative communication, a cooperative SM scheme with AF (AF-SM) protocol is presented, and the corresponding Bit error rate (BER) performance is investigated. Based on the performance analysis, the error probability of antenna index estimation ( P a ) and the error probability of symbol estimation ( P d ), which constitutes the overall average BER, are derived, respectively. As a result, tightly approximate closedform expressions of P a and P d are attained, respectively. Using these expressions, the closed-form overall average BER is achieved. During the analysis, the computational complexity of AF-SM detector is also provided in terms of the complex addition and multiplication numbers. Besides, the asymptotic BER at high SNR and the corresponding diversity gain are further derived, and the resultant diversity gain of N r + 1 is obtained for the system with N r receive antennas. Simulation results show that our theoretical analysis is valid, and provides good performance evaluation method for cooperative SM system. Moreover, the BER performance can be effectively improved as N r increases due to higher diversity gain.

Generalized Frequency Division Multiplexing (GFDM) is a recently proposed multicarrier waveform which is being considered as one of the possible candidates for 5G wireless access technology. Key advantages of GFDM are flexibility and low spectral side-lobes which render it the ability of carrier aggregation and reduced adjacent channel interference. However due to its non-orthogonal nature GFDM suffers from the problem of self-inter-carrier interference (self-ICI) which occurs between time-slots of adjacent subcarriers in the signal. Due to self-ICI, the bit-error rate (BER) performance of GFDM is severely affected under low SNR regime. Moreover application of MIMO technique to GFDM further adversely affects the BER due to inter-antenna interference (IAI) between the transmit antennas. To mitigate this problem, spatial modulation has been applied to GFDM and its BER performance with respect to V-BLAST based GFDM is studied under flat Rayleigh fading channel condition through computer simulations.

Few-mode fibers (FMFs) have garnered significant attention in mode-division multiplexing (MDM) systems due to their parallel mode transmission capability. To achieve efficient mode control and characterization of FMFs, we present an adaptive spatial light modulation (SLM) system based on the genetic algorithm (GA) by combing the digital holography (DH). Single exposure DH technology can achieve precise decomposition of fiber modes, laying the foundation for subsequent precise mode control. The implementation of GA optimization can effectively select the phase distribution loaded onto SLM corresponds to the expected mode field in FMFs. Through closed-loop optimization of output mode fields via a SLM, this methodology achieves high-purity linearly polarized (LP) mode generation by solely monitoring the output intensity distribution. The experimental results show that in FMFs with 5.3 µm and 8.2 µm core diameters, LP mode excitation with purity approaching 90% is demonstrated. Simultaneous excitation of composite LP mode superposition fields generates customized intensity profiles showing similarity indices nearing 0.9. This self-adaptive mode control approach eliminates complex coupling requirements while overcoming static encoding limitations, providing critical technical support for advancing MDM systems.

Optimizing machine learning methods for groundwater quality prediction: Case study in District Bagh, Azad Kashmir, Pakistan.

Spatial modulation (SM) is a promising multiple- input multiple-output system used to increase spectral efficiency. The maximum likelihood (ML) decoder jointly detects the transmitted SM symbol, which is of high complexity. In this paper, a novel reliable sphere decoder (RSD) algorithm based on tree-search is proposed for the SM system. The basic idea of the proposed RSD algorithm is to reduce the size of the tree- search, and then, a smart searching method inside the reduced tree-search is performed to find the solution. The proposed RSD algorithm provides a significant reduction in decoding complexity compared to the ML decoder and existent decoders as well. Moreover, the RSD algorithm provides a flexible tradeoff between the bit error rate (BER) performance and decoding complexity, so as to be reliable for a wide range of practical hardware implementations. The BER performance and decoding complexity analysis for the RSD algorithm are studied, and Monte Carlo simulations are then provided to demonstrate the findings.

Throughput improvement of STS-based MC DS-CDMA system with adaptive modulation