Deep Learning Models for Wireless Signal Classification With Distributed Low-Cost Spectrum Sensors

Abstract

This paper looks into the modulation classification problem for a distributed wireless spectrum sensing network. First, a new data-driven model for Automatic Modulation Classification (AMC) based on long short term memory (LSTM) is proposed. The model learns from the time domain amplitude and phase information of the modulation schemes present in the training data without requiring expert features like higher order cyclic moments. Analyses show that the proposed model yields an average classification accuracy of close to 90% at varying SNR conditions ranging from 0dB to 20dB. Further, we explore the utility of this LSTM model for a variable symbol rate scenario. We show that a LSTM based model can learn good representations of variable length time domain sequences, which is useful in classifying modulation signals with different symbol rates. The achieved accuracy of 75% on an input sample length of 64 for which it was not trained, substantiates the representation power of the model. To reduce the data communication overhead from distributed sensors, the feasibility of classification using averaged magnitude spectrum data and on-line classification on the low-cost spectrum sensors are studied. Furthermore, quantized realizations of the proposed models are analyzed for deployment on sensors with low processing power.

Data Augmentation for Deep Learning-Based Radio Modulation Classification

“…Considering channel interference, the CNN structure also achieves a considerable classification accuracy [13]. In addition to the CNN-based models, the Long Short-Term Memory (LSTM) architecture with time-dependent amplitude and phase information can achieve the state-of-theart classification accuracy [18]. To reduce the training time of deep learning models, different subsampling techniques are investigated in [17] which reduce the dimensions of the input signals.…”

Deep Learning for Physical-Layer 5G Wireless Techniques: Opportunities, Challenges and Solutions

“…is an appealing technique for environment identification and transmitter identification. Inspired by the strengths of the long short term memory (LSTM), deep learning-based signal classification is developed [14]. The framework is trained to classify 11 typical modulation Rectified Linear Unit (ReLU) function is introduced to activate all the hidden layers.…”

Real-Time Interference Identification via Supervised Learning: Embedding Coexistence Awareness in IoT Devices

“…The effect of digital RSSI filtering, here obtained by an 802.15.4-emulator, is most devastating in terms of information loss. Due to low-pass filtering (LPF) effect, not only it wipes out the information on signal envelopeinhibiting modulation-based identification (such as [10]), but also caps time-resolution-dampening the information gain of sampling frequencies above f r = 7.8 kHz. This, in turn, reflects on the inability to capture short interference bursts and inter-frame spaces, such as the 802.11 DCF inter-frame space (DIFS).…”

“…Using specialized hardware, such as WLAN cards and SDRs, for protocol-free IDI has also been investigated in many studies, e.g., [10], [18], [19] and references therein. As this hardware can ensure high sampling frequency and highresolution I/Q data, the benefit of more complex classifiers (e.g., deep learning-based [10]) increases and higher accuracy is generally achieved. However, we have shown that, even with limited sensing resolution and lightweight supervised learning, COTS IoT nodes can reach the same-level of accuracy in realtime.…”