Accumulated Polar Feature-Based Deep Learning for Efficient and Lightweight Automatic Modulation Classification With Channel Compensation Mechanism

Abstract

In next-generation communications, massive machine-type communications (mMTC) induce severe burden on base stations. To address such an issue, automatic modulation classification (AMC) can help to reduce signaling overhead by blindly recognizing the modulation types without handshaking. Thus, it plays an important role in future intelligent modems. The emerging deep learning (DL) technique stores intelligence in the network, resulting in superior performance over traditional approaches. However, conventional DL-based approaches suffer from heavy training overhead, memory overhead, and computational complexity, which severely hinder practical applications for resource-limited scenarios, such as Vehicle-to-Everything (V2X) applications. Furthermore, the overhead of online retraining under time-varying fading channels has not been studied in the prior arts. In this work, an accumulated polar feature-based DL with a channel compensation mechanism is proposed to cope with the aforementioned issues. Firstly, the simulation results show that learning features from the polar domain with historical data information can approach nearoptimal performance while reducing training overhead by 99.8 times. Secondly, the proposed neural network-based channel estimator (NN-CE) can learn the channel response and compensate for the distorted channel with 13% improvement. Moreover, in applying this lightweight NN-CE in a time-varying fading channel, two efficient mechanisms of online retraining are proposed, which can reduce transmission overhead and retraining overhead by 90% and 76%, respectively. Finally, the performance of the proposed approach is evaluated and compared with prior arts on a public dataset to demonstrate its great efficiency and lightness. The lightweight and efficient learning features of the proposed mechanism will be very attractive for future resourceconstrained/aware IoT/V2X applications.

Machine Learning Based Automatic Modulation Recognition for Wireless Communications: A Comprehensive Survey

“…Teng et al [5] extend their previous work in [91] and proposed an accumulated polar feature-based DL with a channel compensation mechanism. In [91], they have shown that learning features from polar coordinates, which can be obtained from Cartesian coordinates, can achieve higher recognition accuracy.…”

“…In [91], they have shown that learning features from polar coordinates, which can be obtained from Cartesian coordinates, can achieve higher recognition accuracy. Then in [5], they add a new temporal axis to accumulate historical information of symbols in such dimension. In the proposed method, the polar coordinates were projected to grid-like images.…”

Automatic Modulation Classification Based on One-Dimensional Convolution Feature Fusion Network

“…In this paper, the I/Q format of the original complex sample is mainly converted to A/P format; in other words, the original sample is converted from I/Q coordinates to polar coordinates [7]. In literature [15], the author directly mapped the received complex symbols to the constellation map on the complex plane as features and achieved good performance.…”

“…To cope with a more complex realistic environment and reduce the influence of channels on transmitted signals, an improved CNN method is proposed in [6] to correct signal distortion that may occur in wireless channels. In [7], a channel estimator based on neural network is designed to find the inverse channel response and improve the accuracy of the network by reducing the influence of channel fading [8]. Based on the theoretical knowledge of signal parameter estimation, a parameter estimator is introduced to extract information related to phase offset and transform phase parameters.…”

Deep Learning-Based Automatic Modulation Classification With Blind OFDM Parameter Estimation

“…Hu et al [21] proposed a robust modulation classifier in an uncertain noise channel environment by adopting the attention mechanism to the output of an LSTM layer. To compensate for the fading channel characteristics, Teng et al [32] proposed a CNN technique to estimate the channel response for improving modulation classification using the accumulated amplitude and phase data.…”