Low Complexity Machine Learning Research Articles

Dynamically stabilized recurrent neural network optimized with intensified sand cat swarm optimization for intrusion detection in wireless sensor network

Wireless Sensor Networks (WSNs) are susceptible to various security threats owing to its deployment in hostile environments. Intrusion detection system (IDS) contributes a critical role on securing WSNs by identifying malevolent activities and ensuring data integrity. Traditional IDS techniques often struggle with the dynamic and resource-constrained nature of WSNs. In this paper, Dynamically Stabilized Recurrent Neural Network Optimized with Intensified Sand Cat Swarm Optimization for Wireless Sensor Network Intrusion identification (DSRNN-ISCOA-ID-WSN) is proposed. Initially, the input data is amassed from WSN-DS dataset. After that, the pre-processing segment receives the data. In pre-processing stage, redundant and biased records are removed from input data with the help of Adaptive multi-scale improved differential filter (AMSIDF). Then the optimal are selected by utilizing Wolf-Bird Optimization Algorithm (WBOA). DSRNN is used to classify the data as Normal, Grey hole, Black hole, Time division multiple access (TDMA), and Flooding attacks. Then Intensified Sand Cat Swarm Optimization (ISCOA) is employed to optimize the weight parameters of DSRNN for accuracte classification. The proposed DSRNN-ISCOA-ID-WSN technique is implemented Python. The performance of the proposed DSRNN-ISCOA-ID-WSN approach attains 29.24 %, 33.45 %, and 28.73 % high accuracy; 30.53 %, 27.64 %, and 26.25 % higher precision when compared with existing method such as Machine Learning-Powered Stochastic Gradient Descent Intrusions Detection System for WSN Attacks (SGDA-ID-WSN), An updated dataset to identify threats in WSN (CNN-ID-WSN) and Denial-of-Service attack detection in WSN: a Low-Complexity Machine Learning Model (DTA-ID-WSN) respectively.

Machine Learning Techniques for Blind Beam Alignment in mmWave Massive MIMO.

This paper proposes methods for Machine Learning (ML)-based Beam Alignment (BA), using low-complexity ML models, and achieves a small pilot overhead. We assume a single-user massive mmWave MIMO, Uplink, using a fully analog architecture. Assuming large-dimension codebooks of possible beam patterns at UE and BS, this data-driven and model-based approach aims to partially and blindly sound a small subset of beams from these codebooks. The proposed BA is blind (no CSI), based on Received Signal Energies (RSEs), and circumvents the need for exhaustively sounding all possible beams. A sub-sampled subset of beams is then used to train several ML models such as low-rank Matrix Factorization (MF), non-negative MF (NMF), and shallow Multi-Layer Perceptron (MLP). We provide an extensive mathematical description of these models and the algorithms for each of them. Our extensive numerical results show that, by sounding only 10% of the beams from the UE and BS codebooks, the proposed ML tools are able to accurately predict the non-sounded beams through multiple transmitted power regimes. This observation holds as the codebook sizes at UE and BS vary from 128×128 to 1024×1024.

A Low-Complexity Machine Learning Design for mmWave Beam Prediction

A Low-Complexity Machine Learning Design for mmWave Beam Prediction

Low Complexity Non-Linear Spectral Features and Wear State Models for Remaining Useful Life Estimation of Bearings

Improving the reliability and performance of electric and rotating machines is crucial to many industrial applications. This will lead to improved robustness, efficiency, and eco-sustainability, as well as mitigate significant health and safety concerns regarding sudden catastrophic failure modes. Bearing degradation is the most significant cause of machine failure and has been reported to cause up to 75% of low-voltage machine failures. This paper introduces a low complexity machine learning (ML) approach to estimate the remaining useful life (RUL) of rolling element bearings using real vibration signals. This work explores different ML recipes using novel feature engineering coupled with various k-Nearest Neighbour (k-NN), and Support Vector Machines (SVM) kernel and weighting functions in order to optimise this RUL approach. Original non-linear wear state models and feature sets are investigated, the latter are derived from Short-time Fourier Transform (STFT) and Hilbert Marginal Spectrum (HMS). These feature sets incorporate one-third octave band filtering for low complexity multivariate feature subspace compression. Our proposed ML algorithm stage has employed two robust supervised ML approaches: weighted k-NN and SVM. Real vibration data were drawn from the Pronostia platform to test and validate this prognostic monitoring approach. The results clearly demonstrate the effectiveness of this approach, with classification accuracy results of up to 82.8% achieved. This work contributes to the field by introducing a robust and computationally inexpensive method for accurate monitoring of machine health using low-cost vibration-based sensing.

An end-to-end real-time pollutants spilling recognition in wastewater based on the IoT-ready SENSIPLUS platform

The problem of detecting illegal pollutants in wastewater is of fundamental importance for public health and security. The availability of distributed, low–cost and low–power monitoring systems, particularly enforced by IoT communication mechanisms and low-complexity machine learning algorithms, would make it feasible and easy to manage in a widespread manner. Accordingly, an End-to-End IoT-ready node for the sensing, local processing, and transmission of the data collected on the pollutants in the wastewater is presented here. The proposed system, organized in sensing and data processing modules, can recognize and distinguish contaminants from unknown substances typically present in wastewater. This is particularly important in the classification stage since distinguishing between background (not of interest) and foreground (of interest) substances drastically improves the classification performance, especially in terms of false positive rates. The measurement system, i.e., the sensing part, is represented by the so-called Smart Cable Water based on the SENSIPLUS chip, which integrates an array of sensors detecting various water-soluble substances through impedance spectroscopy. The data processing is based on a commercial Micro Control Unit (MCU), including an anomaly detection module, a classification module, and a false positive reduction module, all based on machine learning algorithms that have a computational complexity suitable for low-cost hardware implementation.An extensive experimental campaign on different contaminants has been carried out to train machine-learning algorithms suitable for low-cost and low-power MCU. The corresponding dataset has been made publicly available for download. The obtained results demonstrate an excellent classification ability, achieving an accuracy of more than 95% on average, and are a reliable “proof of concept” of a pervasive IoT system for distributed monitoring.

Data reduction through optimized scalar quantization for more compact neural networks

Raw data generation for several existing and planned large physics experiments now exceeds TB/s rates, generating untenable data sets in very little time. Those data often demonstrate high dimensionality while containing limited information. Meanwhile, Machine Learning algorithms are now becoming an essential part of data processing and data analysis. Those algorithms can be used offline for post processing and post data analysis, or they can be used online for real time processing providing ultra low latency experiment monitoring. Both use cases would benefit from data throughput reduction while preserving relevant information: one by reducing the offline storage requirements by several orders of magnitude and the other by allowing ultra fast online inferencing with low complexity Machine Learning models. Moreover, reducing the data source throughput also reduces material cost, power and data management requirements. In this work we demonstrate optimized nonuniform scalar quantization for data source reduction. This data reduction allows lower dimensional representations while preserving the relevant information of the data, thus enabling high accuracy Tiny Machine Learning classifier models for online fast inferences. We demonstrate this approach with an initial proof of concept targeting the CookieBox, an array of electron spectrometers used for angular streaking, that was developed for LCLS-II as an online beam diagnostic tool. We used the Lloyd-Max algorithm with the CookieBox dataset to design an optimized nonuniform scalar quantizer. Optimized quantization lets us reduce input data volume by 69% with no significant impact on inference accuracy. When we tolerate a 2% loss on inference accuracy, we achieved 81% of input data reduction. Finally, the change from a 7-bit to a 3-bit input data quantization reduces our neural network size by 38%.

MACHINE LEARNING TECHNIQUES FOR ROUTING, LOCALIZATION AND QOS FOR FUTURE WIRELESS SENSOR NETWORKS

WSNs plays a crucial role in adopting new generation techniques and their use in creating future-ready technologies. The key difficulties with Wireless Sensor Networks (WSNs) are energy-efficient routing, localization strategies and QoS, as these tiny sensor nodes rely solely on battery power to operate in hazardous situations. So there is a need to research and develop efficient, resilient communication techniques and localization mechanisms to address the issues of WSNs and keep the network operating for an extended period. As a result, low complexity machine learning models manage several difficult tasks such as routing, data aggregation, localization, and motion tracking to define system behavior. Machine learning approaches are thought to be useful for developing energy-efficient routing and localization strategies. Furthermore, machine learning techniques inspire various practical ways to optimize resource utilization and hence increase the lifespan of the sensor network. In this article, an effort has been made to present a broad overview of several machine learning approaches that may be utilized to address various challenges in WSNs, with specific emphasis on routing problems and localization strategies and QoS.

Machine learning-based security-aware spatial modulation for heterogeneous radio-optical networks

In this article, we propose a physical layer security (PLS) technique, namely security-aware spatial modulation (SA-SM), in a multiple-input multiple-output-based heterogeneous network, wherein both optical wireless communications and radio-frequency (RF) technologies coexist. In SA-SM, the time-domain signal is altered prior to transmission using a key at the physical layer for combating eavesdropping. Unlike conventional PLS techniques, SA-SM does not rely on channel characteristics for securing the information, as its perception is self-imposed, which allows its adoption in radio-optical networks. Additionally, a novel periodical key selection algorithm is proposed. Instead of having multiple keys stored in the nodes, by using off-the-shelf and low-complexity machine learning (ML) methods, including a support vector machine, logistic regression and a single-layer neural network, SA-SM nodes can estimate the used key. Results show that a positive secrecy capacity can be achieved for both the RF and optical links by using 1000 different keys, with a minimal signal-to-noise ratio penalty of less than 5 dB for the legitimate user using SA-SM versus conventional transmission at a bit-error-rate of 10−4. The analysis also includes computational time and classification accuracy evaluation of the various proposed ML techniques using different hardware architectures.

An interpretable low-complexity machine learning framework for robust exome-based in-silico diagnosis of Crohn's disease patients.

Whole exome sequencing (WES) data are allowing researchers to pinpoint the causes of many Mendelian disorders. In time, sequencing data will be crucial to solve the genome interpretation puzzle, which aims at uncovering the genotype-to-phenotype relationship, but for the moment many conceptual and technical problems need to be addressed. In particular, very few attempts at the in-silico diagnosis of oligo-to-polygenic disorders have been made so far, due to the complexity of the challenge, the relative scarcity of the data and issues such as batch effects and data heterogeneity, which are confounder factors for machine learning (ML) methods. Here, we propose a method for the exome-based in-silico diagnosis of Crohn’s disease (CD) patients which addresses many of the current methodological issues. First, we devise a rational ML-friendly feature representation for WES data based on the gene mutational burden concept, which is suitable for small sample sizes datasets. Second, we propose a Neural Network (NN) with parameter tying and heavy regularization, in order to limit its complexity and thus the risk of over-fitting. We trained and tested our NN on 3 CD case-controls datasets, comparing the performance with the participants of previous CAGI challenges. We show that, notwithstanding the limited NN complexity, it outperforms the previous approaches. Moreover, we interpret the NN predictions by analyzing the learned patterns at the variant and gene level and investigating the decision process leading to each prediction.

Real-Time Cognitive Workload Monitoring Based on Machine Learning Using Physiological Signals in Rescue Missions.

High levels of cognitive workload decreases human's performance and leads to failures with catastrophic outcomes in risky missions. Today, reliable cognitive workload detection presents a common major challenge, since the workload is not directly observable. However, cognitive workload affects several physiological signals that can be measured non-invasively. The main goal of this work is to develop a reliable machine learning algorithm to identify the cognitive workload induced during rescue missions, which is evaluated through drone control simulation experiments. In addition, we aim to minimize the computing resources usage while maximizing the cognitive workload detection accuracy for a reliable real-time operation. We perform an experiment in which 24 subjects played a rescue mission simulator while respiration, electrocardiogram, photoplethysmogram, and skin temperature signals were measured. State-of-the-art feature-based machine learning algorithms are investigated for cognitive workload characterization using learning curves, data augmentation, and cross-validation techniques. The best classification algorithm is selected, optimized, and the most informative features are selected. Finally, the generalization power of the optimized model is evaluated on an unseen test set. We obtain an accuracy level of 86% on the new unseen datasets using the proposed and optimized eXtreme Gradient Boosting (XGB) algorithm. Then, we reduce the complexity of the machine learning model for future implementation on resource-constrained wearable embedded systems, by optimizing the model and selecting the 26 most important features. Overall, a generalizable and low-complexity machine learning model for cognitive workload detection based on physiological signals is presented for the first time in the literature.

A Low-Complexity Machine Learning Nitrate Loss Predictive Model–Towards Proactive Farm Management in a Networked Catchment

With the advent of wireless sensor networks, the ability to predict nutrient-rich discharges, using on-node prediction models, offers huge potential for enabling real-time water reuse and management within an agriculturally dominated catchment. Existing discharge models use multiple parameters and large historical data which are difficult to extract and this, coupled with constraints on network nodes (battery life, computing power, and sensor availability), makes it necessary to develop low-dimensional models. This paper investigates a data-driven model for predicting daily total oxidized nitrate fluxes and reduces the number of model parameters used to 5-a reduction of at least 50%. Trained on only a 12-month training dataset derived from the published measured data, results for the model generated using an M5 decision tree, giving an R 2 of 0.92 and a relative root-mean-square error of 26%. The 80% of the residuals for test data falls within +/-0.05 Kgůha -1 ůday -1 error range, which is minimal, offering an improvement over results obtained by the contemporary research.

Sparse identification for nonlinear optical communication systems: SINO method.

We introduce low complexity machine learning method method (based on lasso regression, which promotes sparsity, to identify the interaction between symbols in different time slots and to select the minimum number relevant perturbation terms that are employed) for nonlinearity mitigation. The immense intricacy of the problem calls for the development of "smart" methodology, simplifying the analysis without losing the key features that are important for recovery of transmitted data. The proposed sparse identification method for optical systems (SINO) allows to determine the minimal (optimal) number of degrees of freedom required for adaptive mitigation of detrimental nonlinear effects. We demonstrate successful application of the SINO method both for standard fiber communication links (over 3 dB gain) and for few-mode spatial-division-multiplexing systems.

Predicting discharge using a low complexity machine learning model

This paper reports on the validation of a simplified discharge prediction model that is suitable for implementation on a resourced constrained system such as a wireless sensor network, which will allow their operation to become more proactive rather than reactive. The data-driven model, utilising an M5 decision tree modelling technique, is validated using a 12-month training data set derived from published measured data. Daily runoff and drainage is predicted, and the results are compared with existing data-driven models developed in this domain. Results for the model give an R2 of 0.82 and Root Relative Mean Square Error (RRMSE) of 35.9%. 80% of the residuals for the predicted test values fall within a ±2mm discharge depth/day error range. The main significance is that the proposed model gives comparable results with fewer samples and simpler parameters when compared to previous published research, which offers the potential for implementation in resource constrained monitoring and control systems.