Deep Modulation Research Articles

Automatic Modulation Recognition Based on Multimodal Information Processing: A New Approach and Application

Automatic modulation recognition (AMR) has wide applications in the fields of wireless communications, radar systems, and intelligent sensor networks. The existing deep learning-based modulation recognition models often focus on temporal features while overlooking the interrelations and spatio-temporal relationships among different types of signals. To overcome these limitations, a hybrid neural network based on a multimodal parallel structure, called the multimodal parallel hybrid neural network (MPHNN), is proposed to improve the recognition accuracy. The algorithm first preprocesses the data by parallelly processing the multimodal forms of the modulated signals before inputting them into the network. Subsequently, by combining Convolutional Neural Networks (CNN) and Bidirectional Gated Recurrent Unit (Bi-GRU) models, the CNN is used to extract spatial features of the received signals, while the Bi-GRU transmits previous state information of the time series to the current state to capture temporal features. Finally, the Convolutional Block Attention Module (CBAM) and Multi-Head Self-Attention (MHSA) are introduced as two attention mechanisms to handle the temporal and spatial correlations of the signals through an attention fusion mechanism, achieving the calibration of the signal feature maps. The effectiveness of this method is validated using various datasets, with the experimental results demonstrating that the proposed approach can fully utilize the information of multimodal signals. The experimental results show that the recognition accuracy of MPHNN on multiple datasets reaches 93.1%, and it has lower computational complexity and fewer parameters than other models.

MCLAS adaptively rescues disease-specific sleep and wake phenotypes in neurodegeneration

Sleep alterations are hallmarks of prodromal Alzheimer’s (AD) and Parkinson’s disease (PD), with fundamental neuropathological processes of both diseases showing susceptibility of change upon deep sleep modulation. However, promising pharmacological deep sleep enhancement results are hindered by specificity and scalability issues, thus advocating for noninvasive slow-wave activity (SWA) boosting methods to investigate the links between deep sleep and neurodegeneration. Accordingly, we have recently introduced mouse closed-loop auditory stimulation (mCLAS), which is able to successfully boost SWA during deep sleep in neurodegeneration models. Here, we aim at further exploring mCLAS’ acute effect onto disease-specific sleep and wake alterations in AD (Tg2576) and PD (M83) mice. We found that mCLAS adaptively rescues pathological sleep and wake traits depending on the disease-specific impairments observed at baseline in each model. Notably, in AD mice mCLAS significantly increases NREM long/short bout ratio, decreases vigilance state distances by decreasing transition velocities and increases the percentage of cumulative time spent in NREM sleep in the last three hours of the dark period. Contrastingly, in PD mice mCLAS significantly decreases NREM sleep consolidation, by potentiating faster and more frequent transitions between vigilance states, decreases average EMG muscle tone during REM sleep and increases alpha power in WAKE and NREM sleep. Overall, our results indicate that mCLAS selectively prompts an acute alleviation of neurodegeneration-associated sleep and wake phenotypes, by either potentiating sleep consolidation and vigilance state stability in AD or by rescuing bradysomnia and decreasing cortical hyperexcitability in PD. Further experiments assessing the electrophysiological, neuropathological and behavioural long-term effects of mCLAS in neurodegeneration may majorly impact the clinical establishment of sleep-based therapies.

An analytic, efficient and optimal readout algorithm for compact interferometers based on deep frequency modulation

Compact laser interferometers with large dynamic range are one of the core emerging tools to improve low frequency performance in gravitational wave detectors by providing local displacement sensing with sub 1 pmHz-0.5\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {pm Hz}^{-0.5}$$\\end{document} precision. Strong sinusoidal frequency modulations are used in such laser interferometers to create heterodyne-like photodetector signals from which the phase and other parameters, such as the absolute distance, can be extracted. The nested sinusoidal function in such signals is a challenge for the real-time parameter estimation in low-noise applications. In this article, we present an algorithm to calculate exact signal parameters in a non-iterative way from such interferometric signals. The algorithm makes use of a recurrence relation between Bessel functions to enable a direct extraction of modulation parameters from the signal. Additionally, the algorithm is capable of dealing with high phase dynamics where the Doppler-shift of the signal becomes relevant and can limit the range and precision of the parameter estimation, if not accounted for. Simulations show that the algorithm is computationally efficient, can be well parallelised and the phase estimation is close to optimal precision given by the Cramer–Rao lower bound of the signal parameters.

Acoustic deep brain modulation: Enhancing neuronal activation and neurogenesis

BackgroundNon-invasive deep brain modulation (DBM) stands as a promising therapeutic avenue to treat brain diseases. Acoustic DBM represents an innovative and targeted approach to modulate the deep brain, employing techniques such as focused ultrasound and shock waves. Despite its potential, the optimal mechanistic parameters, the effect in the brain and behavioral outcomes of acoustic DBM remains poorly understood. ObjectiveTo establish a robust protocol for the shock wave DBM by optimizing its mechanistic profile of external stimulation, and to assess its efficacy in preclinical settings. MethodsWe used shockwaves due to their capacity to leverage a broader spectrum of peak intensity (10–127 W/mm2) in contrast to ultrasound (0.1–5.0 W/mm2), thereby enabling a more extensive range of neuromodulation effects. We established various types of shockwave pressure profiles of DBM and compared neural and behavioral responses. To ascertain the anticipated cause of the heightened neural activity response, numerical analysis was employed to examine the mechanical dynamics within the brain. ResultsAn optimized profile led to an enhancement in neuronal activity within the hypothalamus of mouse models. The optimized profile in the hippocampus elicited a marked increase in neurogenesis without neuronal damage. Behavioral analyses uncovered a noteworthy reduction in locomotion without significant effects on spatial memory function. ConclusionsThe present study provides an optimized shock wave stimulation protocol for non-invasive DBM. Our optimized stimulation profile selectively triggers neural functions in the deep brain. Our protocol paves the way for new non-invasive DBM devices to treat brain diseases.

A Physiological Marker for Deep Brain Ultrasonic Neuromodulation

ObjectivesTranscranial ultrasound neuromodulation (TUSN) is a noninvasive and spatially specific therapy that promises to deliver treatments tailored to the specific needs of individuals. To fulfill this promise, each treatment must be modified to adequately correct for variation across individual skulls and neural anatomy. This study examines the use of ultrasound-induced voltage potentials (measured with electroencephalography [EEG]) to guide TUSN therapies. Materials and MethodsWe measured EEG responses in two awake nonhuman primates during sonication of 12 targets surrounding two deep brain nuclei, the left and right lateral geniculate nucleus. ResultsWe report reliable ultrasound evoked potentials measured with EEG after the deep brain ultrasonic modulation in nonhuman primates. Robust responses are observed after just ten repetitions of the ultrasonic stimuli. Moreover, these potentials are only evoked for specific deep brain targets. Furthermore, a behavioral study in one subject shows a direct correspondence between the target with maximal EEG response and ultrasound-based modulation of visual choice behavior. Thus, this study provides evidence for the feasibility of EEG-based guidance for ultrasound neuromodulation therapies.

Optical Simulation and Development towards Compact Sensor Heads using Deep Frequency Modulation Interferometry

Zusammenfassung Die Tiefenfrequenzmodulationsinterferometrie (DFMI) bietet eine vielversprechende Alternative zur klassischen Heterodyn-Interferometrie, indem sie die optische Komplexität reduziert und gleichzeitig eine dynamische Auslesung über mehrere Laser-Wellenlängen sicherstellt. DFMI hat auch das Potenzial absolute Entfernung zu messen, indem die optische Laufzeit in die Modulationstiefe enkodiert wird, was diese Technik attraktiv für Hilfssensoren in Gravitationswellendetektoren und für industrielle Anwendungen macht. Diese Studie untersucht DFMI in einem Michelson-Interferometer zur Messung von Oberflächenverformungen mit einem Sensornetzwerk. Wir adressieren Herausforderungen bei der Entwicklung kompakter, drahtloser Sensoren mit kostengünstigen Laserdioden, die unerwünschte Amplitudenmodulationen einführen. Wir bewerten die Effizienz und Phasenfehler von DFMI-Anpassungsalgorithmen bei starker residualer Amplitudenmodulation und zeigen die Notwendigkeit neuer Anpassungsmethoden, um die Fehler selbst bei schwachen Amplitudenmodulationen zu reduzieren. Um den Dynamikbereich eines kompakten Sensors abzuschätzen, simulieren wir ein Michelson-Interferometer mit IFOCAD, um die interferometrischen Signale in Abhängigkeit der Testmassen-Bewegung vorherzusagen. Unsere Ergebnisse zeigen, dass die Verwendung eines Retroreflektors als Testmasse robuste Translationsmessungen auch bei hohen Winkeldynamiken ermöglicht und eine genaue Bewegungserkennung senkrecht zur Laserstrahlachse durch die Nutzung einer Quadrant-Photodiode mit differentieller Wellenfront- oder Leistungsmessung erlaubt. Wir vergleichen diese Ergebnisse mit der Leistung eines Planspiegels als Testmasse.

Complex phase modulation of liquid crystal devices with deep learning.

A deep learning-based phase modulation method for liquid crystal (LC) devices was demonstrated. For LC devices with a single-electrode structure, achieving complex phase distributions is highly challenging. Meanwhile, multi-electrode LC devices, as pixel resolution increases and electrode size decreases, encounter issues of cumbersome modulation steps and reduced modulation accuracy during the phase modulation process. This method uses the concept of field to modulate the phase of the LC device, providing an effective phase modulation scheme. By establishing a deep learning model, it maps the phase retardation distribution of LC devices onto the electric field distribution. This method effectively mitigates the phase modulation issues arising from the fringe field effect, enabling an accurate and precise phase modulation distribution.

All-optical modulator with photonic topological insulator made of metallic quantum wells.

All-optical modulators hold significant prospects for future information processing technologies for they are able to process optical signals without the electro-optical convertor which limits the achievable modulation bandwidth. However, owing to the hardly-controlled optical backscattering in the commonly-used device geometries and the weak optical nonlinearities of the conventional material systems, constructing an all-optical modulator with a large bandwidth and a deep modulation depth in an integration manner is still challenging. Here, we propose an approach to achieving an on-chip ultrafast all-optical modulator with ultra-high modulation efficiency and a small footprint by using photonic topological insulators (PTIs) made of metallic quantum wells (MQWs). Since PTIs have attracted significant attention because of their unidirectional propagating edge states, which mitigate optical backscattering caused by structural imperfections or defects. Meanwhile, MQWs have shown a large Kerr nonlinearity, facilitating the development of minimally sized nonlinear optical devices including all-optical modulators. The proposed photonic topological modulator shows a remarkable modulation depth of 15 dB with a substantial modulation bandwidth above THz in a tiny footprint of only 4 × 10 µm2, which manifests itself as one of the most compact optical modulators compared with the reported ones possessing a bandwidth above 100 GHz. Such a high-performance optical modulator could enable new functionalities in future optical communication and information processing systems.

Comprehensive Treatment of Knee Osteoarthritis Pain through Cryoneurolysis: A Promising Approach for Deep and Superficial Genicular Nerve Modulation - Case Report

Background Knee osteoarthritis is a prevalent and debilitating condition characterized by joint pain, stiffness, and limited mobility. Conventional non-operative treatment approaches often do not provide adequate relief of the patient’s symptoms. Cryoneurolysis of the superficial genicular nerves is a growing and accepted technique used to treat knee pain, however, there is limited literature demonstrating the effectiveness of treating both the superficial and deep genicular nerves, as well as the use of subsequent repeat treatments. Case Report This case study reports on a 65 year-old male with Kellgren-Lawrence grade 2-3 tricompartmental bilateral knee osteoarthritis who underwent cryoneurolysis of the superficial and deep genicular nerves with the iovera° cryoneurolysis device followed by two subsequent repeat cryoneurolysis treatments. Conclusion The results demonstrated a complete resolution of the patient’s pain and enhanced quality of life following cryoneurolysis. Furthermore, subsequent repeat treatments sustained the pain reduction, offering the patient an extended period of improved mobility and reduced discomfort. This case study demonstrates that cryoneurolysis holds promise as a valuable addition to the armamentarium of interventions available for managing knee pain and improving the quality of life for patients suffering from osteoarthritis when they are unable or decide not to pursue surgical intervention.

Deep Learning-Based Multi-Feature Fusion for Communication and Radar Signal Sensing

Recent years witness the rapid development of communication and radar technologies, and many transmitters are equipped with both communication and radar functionalities. To keep track of the working state of a target dual-functional transmitter, it is crucial to sense the modulation mode of the emitted signals. In this paper, we propose a deep learning-based intelligent modulation sensing technique for dual-functional transmitters. Different from existing modulation sensing methods, which usually focus on communication signals, we take both communication and radar signals into consideration. Typically, the dominant features of communication signals lie in the time domain, while those of radar signals lie in both time and frequency domains. To enhance the sensing accuracy, we first exploit real and complex value convolution operations to extract both time-domain and frequency-domain features of emitted signals from the target transmitter. Then, we fuse the extracted features by assigning proper weights with the attention mechanism. Simulation results reveal that the proposed technique can improve the sensing accuracy by up to 4% on average compared with benchmarks.

Deep Convolution Modulation for Image Super-Resolution

Deep Convolution Modulation for Image Super-Resolution

Communication signal detection based on high-order cumulants time-frequency analysis: on the application of deep learning YOLOV5 network

The detection of communication signals in heterogeneous electromagnetic environments currently relies primarily on a one-dimensional statistical feature threshold method. However, this approach is highly sensitive to dynamic changes in the environment, fluctuations in signal-to-noise ratios, and complex noise. To address these limitations, this paper proposes a novel time-frequency diagram based on high-order accumulation for signal detection. Traditional time-frequency diagrams suffer from poor noise suppression ability and unclear features. However, higher-order cumulants can effectively overcome these shortcomings. Currently, methods based on higher-order cumulants are typically limited to one-dimensional signals. Yet, two-dimensional time-frequency signal diagrams can represent a broader array of features. This paper employs higher-order accumulation to extract time-frequency features from the received signal, thereby transforming the conventional radio detection problem into an image recognition challenge. By merging the advantages of higher-order accumulations and time-frequency diagrams, we propose the use of higher-order accumulation time-frequency diagrams for signal detection. Extensive experimental simulations demonstrate that the proposed time-frequency diagram exhibits strong anti-noise performance and effectively suppresses frequency bias from multiple perspectives. The performance of the Higher-Order Cumulant-Time Frequency (HOC-TF) indicated lower Root Mean Square Error (RMSE) compared with the Short-Time Fourier Transform-Time Frequency (STFT-TF) and Wavelet Transform-Time Frequency (WT-TF). Additionally, compared to the STFT-TF and WT-TF methodologies, the novel time-frequency diagram introduced demonstrates superior stability using the Singular Value Decomposition (SVD) method. Moreover, by combining the new time-frequency diagram with the deep learning YOLOV5 network, signal detection and modulation identification of communication signals can be achieved.

Wireless Universal Adversarial Attack and Defense for Deep Learning-Based Modulation Classification

Wireless Universal Adversarial Attack and Defense for Deep Learning-Based Modulation Classification

Towards explainability for AI-based edge wireless signal automatic modulation classification

With the development of artificial intelligence technology and edge computing technology, deep learning-based automatic modulation classification (AI-based AMC) deployed at edge devices using centralised or distributed learning methods for optimisation has emerged in recent years, and has made great progress in the recognition accuracy and recognisable range of wireless signals. However, the lack of sufficient explanation of these models leads to low accuracy and training efficiency of model training, and their applications and further improvements are limited. Researchers have started to propose interpretable methods for technical analysis of deep learning-based AMC. In this paper, based on the research and application development of interpretable methods in recent years, we review the applicable methods and existing research challenges of interpretable automatic modulation classification. And an interpretable AI-based automatic modulation classification framework is proposed to map the interpretability of automatic modulation classification results by obtaining the contribution of wireless signal features to deep learning network training. Experimental results show that the proposed method possesses the ability to explore the classification mechanism of non-transparent auto-modulated classification networks and has the potential to help edge devices train networks with lower energy consumption and higher accuracy.

Deep learning-based modulation recognition with constellation diagram: A case study

Automatic modulation recognition is a promising solution for identifying and classifying signals received in heterogeneous wireless networks. In dynamic and autonomous environments, receivers must extract the relevant signal from various modulated signals to enable further communication procedures. Machine learning, including its sub-branches for classification problems, offers promising operational capabilities. This study utilized the ResNet-50 deep learning method for modulation classification. A dataset consisting of eight digital modulation techniques was generated, with constellation diagrams created as image data over the additive white Gaussian noise (AWGN) channel at signal-to-noise ratios (SNR) of 5 dB, 10 dB, and 20 dB. The deep learning algorithm's performance metrics were evaluated using a confusion matrix, and F1 scores were compared to those of the AlexNet deep learning algorithm. The simulation results clearly indicate the superior performance of ResNet-50 over AlexNet. In terms of average F1 scores, ResNet-50 exhibits a significant advantage, surpassing AlexNet by approximately 67%, 29%, and 10% at SNR values of 5 dB, 10 dB, and 20 dB, respectively.

Thermo-optical liquid-crystal phase modulator for ultrafast optics, driven by neural network

We propose a new spatial light modulator (SLM) concept, relying on a local thermal modification of a thick liquid crystal layer, that is optically-induced through the absorption of a control beam. This innovative thermo-optically addressed SLM, coined TOA-SLM, has shown dynamic phase control capabilities over multi-octave light spectrum, as a promising candidate for spatial or temporal manipulation of ultrafast pulses. In addition to being ultra-broadband and programmable, such a device is low-cost, large-aperture and un-segmented with a high number of control points. The construction and training of a neural network-based statistical model provides configurable design of a prototype TOA-SLM. This step, together with the ultra-broadband acceptance of the device and its ability to introduce continuous and deep phase modulation over a large aperture, opens the way for ultrafast laser aberration compensation using this new technology.

Deep Unintentional Modulation Feature Extraction Framework Based on Decomposition Reconstruction and Metric Learning

Deep Unintentional Modulation Feature Extraction Framework Based on Decomposition Reconstruction and Metric Learning

Modulation recognition method of mixed signals based on cyclic spectrum projection

The signal in the receiver is mainly a combination of different modulation types due to the complex electromagnetic environment, which makes the modulation recognition of the mixed signal a hot topic in recent years. In response to the poor adaptability of existing mixed signals recognition methods, this paper proposes a new recognition method for mixed signals based on cyclic spectrum projection and deep neural network. Firstly, through theoretical derivation, we prove the feasibility of using cyclic spectrum for mixed communication signal identification. Then, we adopt grayscale projections on the two-dimensional cyclic spectrum as identifying representation. And a new nonlinear piecewise mapping and directed pseudo-clustering method are used to enhance the above-mentioned grayscale images, which reduces the impact of energy ratios and symbol rates on signal identification. Finally, we use deep neural networks to extract deep abstract modulation information to achieve effective recognition of mixed signals. Simulation results show that the proposed method is robust against noise. When signal-to-noise ratio is not less than 0 dB, the average recognition rate is greater than 95%. Furthermore, this method exhibits good robustness towards the changes in signal symbol rates and energy ratios between mixed signals.

Defending Wireless Receivers Against Adversarial Attacks on Modulation Classifiers

Deep learning has been adopted for a wide range of wireless communication tasks, including modulation classification, because of its great classification capability. However, deep learning models have been shown to also introduce risks and vulnerabilities. For instance, adversarial attacks craft and introduce imperceptible perturbations that compromise the accuracy of deep learning-based modulation classifiers on wireless receivers. Therefore, in this paper, we propose a novel wireless receiver architecture that enhances deep learning-based modulation classifiers to defend them against adversarial attacks. Our experimental results show that our defense technique significantly diminishes the accuracy reduction that is caused by adversarial attacks by protecting modulation classifiers at least 18% more than existing defense techniques.

Adversarial Example Detection and Restoration Defensive Framework for Signal Intelligent Recognition Networks

Deep learning-based automatic modulation recognition networks are susceptible to adversarial attacks, posing significant performance vulnerabilities. In response, we introduce a defense framework enriched by tailored autoencoder (AE) techniques. Our design features a detection AE that harnesses reconstruction errors and convolutional neural networks to discern deep features, employing thresholds from reconstruction error and Kullback–Leibler divergence to identify adversarial samples and their origin mechanisms. Additionally, a restoration AE with a multi-layered structure effectively restores adversarial samples generated via optimization methods, ensuring accurate classification. Tested rigorously on the RML2016.10a dataset, our framework proves robust against adversarial threats, presenting a versatile defense solution compatible with various deep learning models.