Haotian Zhang

Haotian Zhang

Auto-generated profile, based on publicly available information

Haotian Zhang

School of Electronics and Communication Engineering Sun Yat‐Sen University Shenzhen China

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 October 2019
Verified
Conference Paper
Closed access
HZHaotian Zhang
Haotian Zhang
LZLin Zhang
Lin Zhang
YJYuan Jiang
Yuan Jiang
In this paper, we study the deep learning (DL) based end- to-end transmission systems, then we present the analysis for the underfitting and overfitting phenomena which happen during the training of the neural networks (NNs). Different from the DL based image processing, the transmission systems require the bit error rate to be as low as 10– 4 to achieve high reliability, thus the training errors induced by the underfitting and overfitting may greatly degrade the transmission reliability performances of DL-based communications. In considered DL-based communication systems, we propose to use the average, variance and minimum of transmitted signals' minimum Euclidean distance to estimate the effects of underfitting and overfitting on the error rate performances in terms of the energy per bit to noise power spectral ratio E b / N o of signals. Furthermore, we propose to apply the regularization scheme to alleviate the overfitting issue. Simulations are performed to demonstrate the underfitting and overfitting analysis for transmission systems over the additive white Gaussian noise (AWGN) channel, and validate the improved performances by applying the regularization.

7 September 2020 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
LZLin Zhang
Lin Zhang
HZHaotian Zhang
Haotian Zhang
YJYuan Jiang
Yuan Jiang
ZWZhiqiang Wu
Zhiqiang Wu
Chaos communications have widely been applied to provide secure and anti-jamming transmissions by exploiting the irregular chaotic behavior. However, the real-valued chaotic sequences imposed on the information induce interferences to the user data, thereby leading to reliability performance degradations. To address this issue, in this paper, we propose to utilize the intelligent and feature extraction capability of the deep neural network (DNN) to learn the transmission patterns to demodulate the received signals. In our design, we propose to construct the long short-term memory (LSTM) unit aided intelligent DNN-based deep learning (DL) demodulator for orthogonal frequency division multiplexing-aided differential chaos shift keying (OFDM-DCSK) systems. After learning and extracting features of information-bearing chaotic transmissions at the training stage, at the deployment stage, the received signals can be recovered efficiently and reliably. Thanks to the recursive LSTM-aided DL design, correlations between information-bearing chaotic modulated signals can be exploited to enhance reliability performances. Simulation results demonstrate with the proposed DL demodulation design, and the intelligent OFDM-DCSK system can achieve more reliable performances over additive white Gaussian noise (AWGN) channel and fading channels compared with benchmark systems.

25 May 2022 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
HZHaotian Zhang
Haotian Zhang
LZLin Zhang
Lin Zhang
YJYuan Jiang
Yuan Jiang
ZWZhiqiang Wu
Zhiqiang Wu
Chaos-based communications can be applied to high-speed vehicular information transmissions thanks to the anti-jamming and anti-interference capabilities of chaotic transmissions. In traditional practical chaotic systems, reference chaotic signals are required to be delivered to remove complex chaotic synchronization circuits. However, the direct transmission of reference signals will degrade the security performances, while interferences and noises imposed on the reference signals due to imperfect channel conditions will deteriorate the reliability performances. In order to enhance the reliability and security performances over vehicular channels such as the railway channel and the channels undergoing fast fadings, in this paper, we propose a deep learning (DL) aided intelligent OFDM-DCSK transceiver. In this design, no reference chaotic signals are delivered, and we propose to utilize the time-delay neural network (TDNN) to learn the chaotic maps, followed by the long short-term memory (LSTM) units to extract and exploit the correlations between chaotic modulated signals, and multiple fully connected layers (FCLs) to estimate the user bit data. With the aid of the constructed deep neural network (DNN), after the offline neural network training, the receiver can recover the transmitted information with lower bit error rate (BER) and enhance security performances. Theoretical performance is then analyzed for the proposed intelligent transceiver. Simulation results validate the proposed design, and demonstrate that the intelligent DL-based OFDM-DCSK system can achieve better BER and security performances over fast fading and railway channels compared with the benchmark systems.

13 September 2021 in IEEE Transactions on Cognitive Communications and Networking
Verified
Journal Article
Closed access
HZHaotian Zhang
Haotian Zhang
LZLin Zhang
Lin Zhang
Julian ChengJulian Cheng
Julian Cheng
YJYuan Jiang
Yuan Jiang
ZWZhiqiang Wu
Zhiqiang Wu
Multilevel code shifted ${M}$ -ary differential chaos shift keying (MCS-MDCSK) system provides higher data rate chaotic information transmission by applying multilevel code shifting aided ${M}$ -ary modulation. However, the real-valued chaotic sequences induce interferences to signals while higher-order modulation shortens the Euclidean distance between adjacent symbols, thereby leading to performance degradation. To improve the bit-error rate (BER) performances, we propose an intelligent detector to achieve the joint demodulation and de-spreading at the receiver. In this design, we construct the recursive long short-term memory (LSTM) unit to extract features from the correlated chaotic modulated signals. Then we concatenate the LSTM unit with multiple full connection layers (FCLs) and compose the deep neural network (DNN) to recover the information. Owing to the serial concatenated LSTM-aided DNN, the intelligent detector can learn the joint chaotic modulation and spreading pattern, and achieve the joint demodulation and de-spreading. Consequently, larger performance gain can be attained and the reliability performances will be improved. Simulation results validate the proposed design. Moreover, for practical systems undergoing multiplicative fading, the intelligent MCS-MDCSK detector exhibits better BER performances than the benchmark systems.

12 December 2023 in IEEE Internet of Things Journal
Verified
Journal Article
Closed access
HZHaotian Zhang
Haotian Zhang
LZLei Zhao
Lei Zhao
YJYuan Jiang
Yuan Jiang
By exploiting the inherent hardware characteristics of wireless devices, radio frequency fingerprint identification (RFFI) has been widely applied in device authentication and spoofing attack detection to improve the security. However, due to the dependence on the training sample size, the state-of-the-art deep learning (DL)-based identification methods will face serious overfitting problem with inadequate training samples. Besides, in noncooperative scenarios, most existing methods cannot tackle the challenge of zero-shot identification, i.e., identifying the objects outside of the training data with no prior samples, which obstructs their practical applications. To solve these two problems, in this article, we propose a novel identification method that combines the deep neural network (DNN) and the unsupervised clustering. In this design, after offline training, the triplet loss convolutional neural network (TLCNN) can be utilized to extract the features of the radio frequency signals outside of the training set. Then, the $K$ -means++ clustering algorithm is applied to the extracted features to realize zero-shot RFFI. Moreover, to improve the identification performances of the proposed design under small sample conditions, the random integration (RI) augmentation and the variational mode decomposition (VMD) are exploited to preprocess the input signals, and a $K$ -value estimation method for the $K$ -means++ clustering is proposed to ensure the effectiveness of the proposed design. Experiment on digital mobile radio (DMR) portable radios validates that the proposed design can achieve the best identification performances and the minimal time and space costs compared with the benchmark schemes.

18 July 2024 in Electronics Letters
Verified
Journal Article
Open access
HZHaotian Zhang
Haotian Zhang
YJYuan Jiang
Yuan Jiang
LZLei Zhao
Lei Zhao
BPBo Peng
Bo Peng
As a potential solution to improve wireless security, specific emitter identification is a lightweight access authentication technology. However, the existed deep learning‐based specific emitter identification methods are highly dependent on the training sample size, leading to serious overfitting problem when the training samples are inadequate, which obstructs their practical applications. To address this issue, an innovative data augmentation method to effectively expand the sample size is proposed. In this design, after data preprocessing, a random integration based data augmentation is applied to integrate several initial samples and generate new samples. Furthermore, compared with the existed methods, chaotic sequences are utilized to randomly set the integration weight of each initial sample, and thus enhancing the diversity of augmented samples. The superiority of the proposed chaotic integration‐based data augmentation method in accuracy, generalization ability and robustness is validated by the hardware implementation on digital mobile radio portable radios.