Quadrature Samples Research Articles

Adaptive Uplink Data Compression in Spectrum Crowdsensing Systems

Understanding spectrum activity is challenging when attempted at scale. The wireless community has recently risen to this challenge in designing spectrum monitoring systems that utilize many low-cost spectrum sensors to gather large volumes of sampled data across space, time, and frequencies. These crowdsensing systems are limited by the uplink bandwidth available to backhaul the raw in-phase and quadrature (IQ) samples and power spectrum density (PSD) data needed to run various applications. This paper presents, a framework based on the Walsh-Hadamard transform to compress spectrum data collected from distributed and low-cost sensors for real-time applications. This transformation allows sensors to significantly save uplink bandwidth thanks to its inherent properties both when it is applied to IQ and PSD data. Additionally, by leveraging a feedback loop between the sensor and the edge device it connects to, carefully adapts the compression ratio over time to changes in the spectrum and different applications, jointly considering data size, application performance, and spectrum variations. We experimentally evaluate in several applications. Our results show that is particularly suitable for IoT transmissions and signals close to the noise floor. Compared with prior work, provides up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$7\times$</tex-math> </inline-formula> more reduction of uplink data size for signal detection based on PSD data, and reduces up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$6\times$</tex-math> </inline-formula> to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$8\times$</tex-math> </inline-formula> the number of undecodable messages for IQ sample decoding.

Improved radar range and velocity resolution using the fast orthogonal search

AbstractThe authors show improved radar range and velocity resolution is achieved using fast orthogonal search in place of the standard fast Fourier transform. The method reliably detects targets that are close either in range or in velocity, which is relevant today given recent advances in target swarms. The method presented also allows radars to detect slower targets closer to the zero‐Doppler region, like uncrewed aerial systems (drones) moving slowly to escape detection. The authors also show a novel method for estimating the velocity spectrum of a pulse‐Doppler radar using the in‐phase and quadrature samples directly by presenting unique terms to the fast orthogonal search. This eliminates the need to convert into complex samples for processing. The authors compared the performance of the new fast orthogonal search method against the fast Fourier transform using both simulated pulse‐Doppler radar data and real frequency‐modulated continuous wave radar data taken in an anechoic chamber. The simulated pulse‐Doppler scenarios included targets closely spaced in velocity and targets near the zero‐Doppler null, all at a variety of signal‐to‐interference ratios. In all cases, the fast orthogonal search method was shown capable of detecting both the closely‐spaced and near‐zero‐Doppler targets a greater percentage of time compared to the fast Fourier transform. Furthermore, when the methods were compared using real frequency‐modulated continuous wave radar data, the stationary targets were resolved at 0.4 m using the fast orthogonal search as compared to 0.6 m for the fast Fourier transform. Thus, this new method is capable of greater radar range and velocity resolution than the fast Fourier transform.

Time–frequency fusion for enhancement of deep learning-based physical layer identification

Existing studies on deep learning-based physical layer identification have mainly exploited raw in-phase/ quadrature (IQ) samples or power spectral density (PSD) samples as inputs independently. The raw IQ and PSD samples represent the information in the time and frequency domains, respectively. It has been observed from the results of existing studies that identification using raw IQ samples outperforms that using PSD in low signal-to-noise ratio (SNR) regimes, and that identification using PSD outperforms that using raw IQ in high SNR regimes. In this paper, we propose to use the fusion of raw IQ and PSD samples to enhance deep learning-based physical layer identification. In particular, we design three general fusion frameworks, i.e., input, feature, and decision fusions, and integrate them with three typical deep neural network architectures of fully connected neural network, convolutional neural network, and recurrent neural network to form fusion identification schemes. We conduct experiments using 50 off-the-shelf Wi-Fi devices to validate the concerned fusion schemes and investigate their performance gains in identification and model training. Our experimental results verify that the proposed fusion identification schemes can achieve comparable or superior identification performances to the state-of-the-art schemes in the entire SNR regime. Moreover, for the considered fusion schemes, we further investigate the impacts of fusion strategies, deep-learning networks, and SNR conditions on the identification performance and training time.

Fronthaul Compression for Uplink Massive MIMO Using Matrix Decomposition

Massive multiple-input-multiple-output (MIMO) is a key enabler for obtaining higher data rates in the next generation wireless technology. While it has the power to transform cellular communication, with potential for spatial diversity and multiplexing, a bottleneck that often gets overlooked is the fronthaul capacity. The fronthaul link that connects a massive MIMO Remote Radio Head (RRH) and carries in-phase and quadrature (IQ) samples to the Baseband Unit (BBU) of the base station can throttle the network capacity/speed if appropriate data compression techniques are not applied, particularly in the uplink. This paper proposes an iterative technique for fronthaul load reduction in the uplink for massive MIMO systems utilizing the convolution structure of the received signals. The proposed algorithm provides compression ratios of about 30-50×. This work provides extensive analysis of the performance of the proposed method for a plethora of practical scenarios and constraints, such as different channel parameters and models, receive antenna correlation, and under imperfect channel information. It also discusses the numerical convergence and complexity of the proposed algorithm and compares the performance against other existing compression techniques.

MIMO-OFDM Modulation Classification Using Three-Dimensional Convolutional Network

Automatic modulation classification (AMC) plays a vital role in cognitive radio to improve spectrum utilization efficiency, however, most of the existing works have focused on single-carrier communications in single-input single-output systems. In this paper, we propose an efficient AMC method for multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) communication systems with the assumption of unknown frequency-selective fading channels and signal-to-noise ratio. At the receiver, the complex envelope samples of a burst signal acquired by multiple antennas are decomposed into in-phase and quadrature samples, which are then structured into a high-dimensional data array. To learn the modulation patterns from received signals, we develop a deep network, namely three-dimensional MIMO-OFDM convolutional neural network (MONet). With cuboidal convolution filters, the proposed MONet allows the network to capture underlying features as intra- and inter-antenna correlations at multi-scale signal representations. Relying on simulations, MONet achieves the classification accuracy of over 95% at 0 dB SNR under various channel impairments and shows the robustness with different MIMO antenna configurations.

SignalNet: A Low Resolution Sinusoid Decomposition and Estimation Network

The detection and estimation of sinusoids is a fundamental signal processing task for many applications related to sensing and communications. While algorithms have been proposed for this setting, quantization is a critical, but often ignored modeling effect. In wireless communications, estimation with low resolution data converters is relevant for reduced power consumption in wideband receivers. Similarly, low resolution sampling in imaging and spectrum sensing allows for efficient data collection. In this work, we propose SignalNet, a neural network architecture that detects the number of sinusoids and estimates their parameters from quantized in-phase and quadrature samples. We incorporate signal reconstruction internally as domain knowledge within the network to enhance learning and surpass traditional algorithms in mean squared error and Chamfer error. We introduce a worst-case learning threshold for comparing the results of our network relative to the underlying data distributions. This threshold provides insight into why neural networks tend to outperform traditional methods and into the learned relationships between the input and output distributions. In simulation, we find that our algorithm is always able to surpass the threshold for three-bit data but often cannot exceed the threshold for one-bit data. We use the learning threshold to explain, in the one-bit case, how our estimators learn to minimize the distributional loss, rather than learn features from the data.

RF Fingerprint Classification With Combinatorial-Randomness-Based Power Amplifiers and Convolutional Neural Networks: Secure analog/RF electronics and electromagnetics

The growth of the IoT requires more comprehensive security measures than ever. RF fingerprinting (RFF) utilizes features in the signals and waveforms from transmitters’ physical-layer imperfections to classify and authenticate devices. To prevent attacks from impersonators, combinatorial randomness is exploited to augment the RF fingerprints with a high-efficiency PA for IoT applications. By enabling different subsets of thinly sliced PA elements, the transmitter can be reconfigured with 220 subsets that exhibit distinctive RF fingerprints for signal analysis at the edge. In this work, a combinatorial-randomness-based PA was implemented in a BLE system. The BLE packets’ in-phase and quadrature samples transmitted from each configuration are collected with different SNRs to emulate the environmental changes in communication channels. A lightweight convolutional neural network (CNN) classifier demonstrates the possibility of accurate and fast inference of unique features in the IoT environment, which our approach exploits to enable on-chip time-varying RF fingerprints.

Radio Frequency Fingerprint Identification for LoRa Using Deep Learning

Radio frequency fingerprint identification (RFFI) is an emerging device authentication technique that relies on the intrinsic hardware characteristics of wireless devices. This paper designs a deep learning-based RFFI scheme for Long Range (LoRa) systems. Firstly, the instantaneous carrier frequency offset (CFO) is found to drift, which could result in misclassification and significantly compromise the stability of the deep learning-based RFFI system. CFO compensation is demonstrated to be effective mitigation. Secondly, three signal representations for deep learning-based RFFI are investigated in time, frequency, and time-frequency domains, namely in-phase and quadrature (IQ) samples, fast Fourier transform (FFT) results and spectrograms, respectively. For these signal representations, three deep learning models are implemented, i.e., multilayer perceptron (MLP), long short-term memory (LSTM) network and convolutional neural network (CNN), in order to explore an optimal framework. Finally, a hybrid classifier that can adjust the prediction of deep learning models with the estimated CFO is designed to further increase the classification accuracy. The CFO will not change dramatically over several continuous days, hence it can be used to correct predictions when the estimated CFO is much different from the reference one. Experimental evaluation is performed in real wireless environments involving 25 LoRa devices and a Universal Software Radio Peripheral (USRP) N210 platform. The spectrogram-CNN model is found to be optimal for classifying LoRa devices which can reach an accuracy of 96.40% with the least complexity and training time.

Robust Digital Signal Recovery for LEO Satellite Communications Subject to High SNR Variation and Transmitter Memory Effects

This paper proposes a robust digital signal recovery (DSR) technique to tackle the high signal-to-noise ratio (SNR) variation and transmitter memory effects for broadband power efficient down-link in next-generation low Earth orbit (LEO) satellite constellations. The robustness against low SNR is achieved by concurrently integrating magnitude normalization and noise feature filtering using a filtering block built with one batch normalization (BN) layer and two bidirectional long short-term memory (BiLSTM) layers. Moreover, unlike existing deep neural network-based DSR techniques (DNN-DSR), which failed to effectively take into account the memory effects of radio-frequency power amplifiers (RF-PAs) in the model design, the proposed BiLSTM-DSR technique can extracts the sequential characteristics of the adjacent in-phase (I) and quadrature (Q) samples, and hence can obtain superior memory effects compensation compared with the DNN-DSR technique. Experimental validation results of the proposed BiLSTM-DSR with a 100 MHz bandwidth OFDM signal demonstrate an excellent performance of 11.83 dB and 9.4% improvement for adjacent channel power ratio (ACPR) and error vector magnitude (EVM), respectively. BiLSTM-DSR also outperforms the existing DNN-DSR technique in terms of the ACPR and EVM by 2.4 dB and 0.9%, which provides a promising solution for developing deep learning-assisted receivers for high-throughput LEO satellite networks.

Joint multislice and cooperative detection aided RFID method based on deep learning

Radio frequency fingerprint identification (RFID) provokes many promising applications of internet of things. Deep learning is considered as one of powerful tools to empower RFID techniques. Recently, several deep learning based RFID methods have been proposed. However, their deep neural networks were trained from the limited length of RF fingerprint samples due to the very high training cost. Hence, existing deep learning based RFID methods are hard to extract full features from the RF fingerprint (RFF) datasets. In order to solve this problem, we propose a joint multislice and cooperative detection aided RFID method based on deep learning. Firstly, acquisition equipment is used to collect RFF signals from seven power amplifiers, which are composed of in-phase and quadrature (IQ) samples with 200,000 sampling points. The IQ samples are sliced and turned into slices with fewer sampling points, which makes them use less training resources, and then they are input into neural network for training. The convolutional neural network (CNN) and convolutional long short-term deep neural networks (CLDNN) are used for training. Secondly, a cooperative detection algorithm is proposed to classify all slices from the same signal to determine the signal category. Finally, experiment results are given to confirm the proposed method in different scenarios. It shows that this method can greatly reduce the training resources, and at the same time maintain a high accuracy.

FPGA-Assisted Real-Time RF Wireless Data Analytics System: Design, Implementation, and Statistical Analyses

A wide range of new ultra reliable low latency communication (URLLC) applications in next generation (NG) wireless systems demand real-time radio frequency (RF) data analytics of channel utilization (CU) that can help in making proactive resource allocation decisions. However, such real-time RF data analytics require processing of tens of millions of in-phase and quadrature (IQ) samples per second and sending huge quantities of samples to a resource allocating entity is not practical. We present design and implementation of an RF data analytics system which utilizes field-programmable gate arrays (FPGAs) at the network edge to process real-time streaming IQ samples from RF transceiver. FPGAs process millions of samples per second and output low-overhead descriptive statistics of wireless CU, such as mean CU values, maximum CU values, and entire histograms to obtain probability distribution of CU values, to a resource controller server where a quantile estimation based technique is used to detect congestion in CU in real-time. The FPGA-based modules are implemented on Xilinx's Zynq-7000 devices mounted with RF transceivers. We evaluate the performance of the implemented analytics system using extensive measurements, testing, and statistical analyses that are performed in both laboratory and over-the-air environments.

Specific Emitter Identification Against Unreliable Features Interference Based on Time-Series Classification Network Structure

Specific emitter identification (SEI) aims to distinguish different emitter individuals based on the subtle differences in the signals. The technology can be widely applied to various wireless communication systems. Existing individual identification methods often ignore the radio frequency (RF) fingerprint information carried by the waveforms and thus can be interfered by unreliable RF features. To alleviate these issues, we devise the deep bidirectional long short-term memory (DBi-LSTM) and the one-dimensional residual convolution network with dilated convolution and squeeze-and-excitation block (Conv-OrdsNet). We exploit the combination of DBi-LSTM and Conv-OrdsNet (CoBONet) to extract temporal structure features directly from baseband in-phase and quadrature (I/Q) samples. The proposed network is able to capture the fine-grained details of signals and combine different information extracted by two networks. Moreover, we propose a data augmentation method to solve the interference of unreliable features by randomly changing the values of noise, power, frequency offset (FO), and phase offset (PO). It is worth noting that our method only needs a short slice of a steady-state signal without complex preprocessing, which reduces the cost of acquisition and calculation. Extensive experiments show that our method can effectively extract reliable RF fingerprinting features from I/Q samples and the classification results are better than most existing methods.

Vibrometry and Sound Reproduction of Acoustic Sources on Moving Platforms Using Millimeter Wave Pulse-Doppler Radar

This paper presents millimeter wave (MMW) pulse-Doppler radar for the remote sensing of acoustic vibration for targets in motion. A key advance in this work is the development of precision motion compensation for MMW vibrometry, making it possible for a monostatic radar to extract and reproduce small-scale vibrations on platforms undergoing large-scale motion. The motion compensation methodology uses a hierarchical approach combining direct and indirect estimation for the time dependent variation of target motion parameters across coherent samples in radar fast time and slow time. Additionally, the wide bandwidth commonly available at MMW allows vibrations to be selectively detected and disambiguated in range across the length of moving targets. Stretch processing compresses the received radar bandwidth by more than 10x, so that Hilbert sampling can be used to acquire quadrature samples using a single analog-to-digital converter. The resulting complex baseband response directly reproduces the target’s acoustic signature. To demonstrate the technique, a 94 GHz pulsed linear frequency modulated (LFM) radar accurately reproduces the pitch of audio waveforms generated by a speaker in the rear of an accelerating automobile at an outdoor test range. These results should have major consequences for the development of MMW vibrometry as a remote sensing technique.

Evaluating Adversarial Evasion Attacks in the Context of Wireless Communications

Recent advancements in radio frequency machine learning (RFML) have demonstrated the use of raw in-phase and quadrature (IQ) samples for multiple spectrum sensing tasks. Yet, deep learning techniques have been shown, in other applications, to be vulnerable to adversarial machine learning (ML) techniques, which seek to craft small perturbations that are added to the input to cause a misclassification. The current work differentiates the threats that adversarial ML poses to RFML systems based on where the attack is executed from: direct access to classifier input, synchronously transmitted over the air (OTA), or asynchronously transmitted from a separate device. Additionally, the current work develops a methodology for evaluating adversarial success in the context of wireless communications, where the primary metric of interest is bit error rate and not human perception, as is the case in image recognition. The methodology is demonstrated using the well known Fast Gradient Sign Method to evaluate the vulnerabilities of raw IQ based Automatic Modulation Classification and concludes RFML is vulnerable to adversarial examples, even in OTA attacks. However, RFML domain specific receiver effects, which would be encountered in an OTA attack, can present significant impairments to adversarial evasion.

Data-Driven Deep Learning for Automatic Modulation Recognition in Cognitive Radios

Automatic modulation recognition (AMR) is an essential and challenging topic in the development of the cognitive radio (CR), and it is a cornerstone of CR adaptive modulation and demodulation capabilities to sense and learn environments and make corresponding adjustments. AMR is essentially a classification problem, and deep learning achieves outstanding performances in various classification tasks. So, this paper proposes a deep learning-based method, combined with two convolutional neural networks (CNNs) trained on different datasets, to achieve higher accuracy AMR. A CNN is trained on samples composed of in-phase and quadrature component signals, otherwise known as in-phase and quadrature samples, to distinguish modulation modes, that are relatively easy to identify. We adopt dropout instead of pooling operation to achieve higher recognition accuracy. A CNN based on constellation diagrams is also designed to recognize modulation modes that are difficult to distinguish in the former CNN, such as 16 quadratic-amplitude modulation (QAM) and 64 QAM, demonstrating the ability to classify QAM signals even in scenarios with a low signal-to-noise ratio.

Enhancing the Coexistence of LTE and Wi-Fi in Unlicensed Spectrum Through Convolutional Neural Networks

Over the last years, the ever-growing wireless traffic has pushed the mobile community to investigate solutions that can assist in more efficient management of the wireless spectrum. Towards this direction, the long-term evolution (LIE) operation in the unlicensed spectrum has been proposed. Targeting a global solution that respects the regional requirements, 3GPP announced the standard of LIE licensed assisted access (LAA). However, LIE LAA may result in unfair coexistence with Wi-Fi, especially when Wi-Fi does not use frame aggregation. Targeting a technique that enables fair channel access, the mLTE-U scheme has been proposed. According to mLTE-U, LTE uses a variable transmission opportunity, followed by a variable muting period that can be exploited by other networks to transmit. For the selection of the appropriate mLTE-U configuration, information about the dynamically changing wireless environment is required. To this end, this paper proposes a convolutional neural network (CNN) that is trained to perform identification of LIE and Wi-Fi transmissions. In addition, it can identify the hidden terminal effect caused by multiple LTE transmissions, multiple Wi-Fi transmissions, or concurrent LIE and Wi-Fi transmissions. The designed CNN has been trained and validated using commercial off-the-shelf LIE and Wi-Fi hardware equipment and for two wireless signal representations, namely, in-phase and quadrature samples and frequency domain representation through fast Fourier transform. The classification accuracy of the two resulting CNNs is tested for different signal to noise ratio values. The experimentation results show that the data representation affects the accuracy of CNN. The obtained information from CNN can be exploited by the mLTE-U scheme in order to provide fair coexistence between the two wireless technologies.

LOW COST SDR RECEIVER IMPLEMENTATION USING RTL-SDR DONGLE

The advent of Software Defined Radio (SDR) has made it possible to replace large parts of classical radio communication equipment with digital and software components to easily achieve configurable and multiplatform communication systems. SDR provides many advantages for research and development. In this work, the low-cost Realtek Software Defined Radio (RTL-SDR) dongle can serve as a cheap SDR platform to receive and digitize exchanged RF signals. In case of RTL-SDR dongle, in-phase and quadrature samples (I/Q) are sent as a stream through simple Universal Serial Bus interface (USB) to host computer. This paper describes a SDR application that can be used for many functionalities such as monitoring and scanning of frequency bands with RTL-SDR dongle.

Deep Learning-Based Signal Modulation Identification in OFDM Systems

Signal modulation identification (SMI) plays a very important role in orthogonal frequency-division multiplexing (OFDM) systems. Currently, SMI methods are often implemented via feature extraction based on machine learning. However, the traditional methods encounter a bottleneck where the probability of correct classification (PCC) is very limited and hence it is hard to implement in practical OFDM systems due to the fact that traditional methods are difficult to extract feature of the OFDM signals. In order solve these problems, we propose a deep learning (DL) based SMI method for identifying OFDM signals. Specifically, convolutional neural network (CNN) is adopted to train in-phase and quadrature (IQ) samples for OFDM signals. Then we choose dropout layer to prevent overfitting and improve its identification accuracy. In addition, datasets with different modulation modes are adopted to verify our trained CNN. Experiments are conducted to show that our proposed method achieves higher accuracy and better consistency than traditional methods. Moreover, extensive results confirm that the proposed method performs robustly in different datasets.

Deep Learning Aided Method for Automatic Modulation Recognition

Automatic modulation recognition (AMR) is considered one of most important techniques in the non-cooperative wireless communication systems. Traditional algorithms, e.g., support vector machine (SVM) based on high order cumulants (HOC), are hard to achieve the reliable performance. In this paper, we propose an effective AMR algorithm based on deep learning (DL) with capabilities of automatically extracting representative and effective features. Our proposed method resorts to in-phase and quadrature (IQ) samples which are IQ components of received baseband signal, respectively. We adopt convolutional neural networks (CNN) and recurrent neural networks (RNN) to classify six types of signal modulations over additive white Gaussian noise (AWGN) channel and Rayleigh fading channel, respectively. Simulation results show that DL-AMR is much better than traditional algorithms under two fading channels.

An FPGA-Based Implementation of a Multifunction Environment Sensing Device for Shared Access With Rotating Radars

To protect radar receivers and to facilitate shared access in radar bands, regulatory bodies have recommended the use of spectrum monitoring devices called environmental sensing capability (ESC). High-speed and low-cost ESC devices are required to process in real time the large amount of data (in-phase and quadrature samples) for the detection of radar signals and to differentiate them from secondary users (SUs) signals. In this paper, we present a field-programmable gate array (FPGA)-based design and implementation of a multifunction ESC device that can detect radar pulses and can also differentiate them from SU signals in microsecond time scales. The proposed ESC device performs the following tasks in parallel: 1) it detects and differentiates between radar and SU signals; 2) it measures received signal strength from SUs for radar protection; and 3) it also measures SUs’ airtime utilization (ATU) in a channel, which can be used to perform load balancing (based on ATU) of SUs on different channels for efficient access. Detection of signals requires threshold setting. We present a novel minimum-based threshold setting technique, which is suitable for real-time operation of energy detectors. We implement a prototype of the proposed ESC device design on a Wireless Open-Access Research Platform node, which is equipped with a Xilinx FPGA. We evaluate the performance of the implemented device and show that with very high probability (close to 100%), it detects and differentiates between radar and SU signals. Moreover, it also accurately measures the ATU of SUs.