arXiv subjects

Qi Dong

Publications and source records attributed to Qi Dong.

28 records · Page 2Linked to original sources

arXiv2019

Time Varying Channel Tracking for Multi-UAV Wideband Communications with Beam Squint

Unmanned aerial vehicle (UAV) has become an appealing solution for a wide range of commercial and civilian applications because of its high mobility and flexible deployment. Due to the continuous UAV navigation, the channel between UAV and base station (BS) is subject to the Doppler effect. Meanwhile, when the BS is equipped with massive number of antennas, the non-negligible propagation delay across the array aperture would cause beam squint effect. In this paper, we first investigate the channel of UAV communications under both Doppler shift effect and beam squint effect. Then, we design a gridless compressed sensing (GCS) based channel tracking method, where the high dimension uplink channel can be derived by estimating a few physical parameters such as the direction of arrival (DOA), Doppler shift, and the complex gain information. Besides, with the Doppler shift reciprocity and angular reciprocity, the downlink channel can be derived by only one pilot symbol, which greatly decreases the downlink channel training overhead. Various simulation results are provided to verify the effectiveness of the proposed methods.

cs.IT↗

arXiv2019

Unsupervised Deep Learning by Neighbourhood Discovery

Deep convolutional neural networks (CNNs) have demonstrated remarkable success in computer vision by supervisedly learning strong visual feature representations. However, training CNNs relies heavily on the availability of exhaustive training data annotations, limiting significantly their deployment and scalability in many application scenarios. In this work, we introduce a generic unsupervised deep learning approach to training deep models without the need for any manual label supervision. Specifically, we progressively discover sample anchored/centred neighbourhoods to reason and learn the underlying class decision boundaries iteratively and accumulatively. Every single neighbourhood is specially formulated so that all the member samples can share the same unseen class labels at high probability for facilitating the extraction of class discriminative feature representations during training. Experiments on image classification show the performance advantages of the proposed method over the state-of-the-art unsupervised learning models on six benchmarks including both coarse-grained and fine-grained object image categorisation.

cs.CV↗

arXiv2019

Single-Label Multi-Class Image Classification by Deep Logistic Regression

The objective learning formulation is essential for the success of convolutional neural networks. In this work, we analyse thoroughly the standard learning objective functions for multi-class classification CNNs: softmax regression (SR) for single-label scenario and logistic regression (LR) for multi-label scenario. Our analyses lead to an inspiration of exploiting LR for single-label classification learning, and then the disclosing of the negative class distraction problem in LR. To address this problem, we develop two novel LR based objective functions that not only generalise the conventional LR but importantly turn out to be competitive alternatives to SR in single label classification. Extensive comparative evaluations demonstrate the model learning advantages of the proposed LR functions over the commonly adopted SR in single-label coarse-grained object categorisation and cross-class fine-grained person instance identification tasks. We also show the performance superiority of our method on clothing attribute classification in comparison to the vanilla LR function.

cs.CV↗

arXiv2018

Constructing Trustworthy and Safe Communities on a Blockchain-Enabled Social Credits System

The emergence of big data and Artificial Intelligence (AI) technology is reshaping the world. While the technological revolution improves the quality of our life, new concerns are triggered. The superhuman capability enables AI to outperform human workers in many data- and/or computing-intensive tasks. Also, digital superpowers are showing arrogance towards individuals, which erodes the trust foundation of the society. In this position paper, we suggest to construct trustworthy and safe communities based on a BLockchain-Enabled Social credits System (BLESS) that rewards the residents who commit in socially beneficial activities. Human being's true value lies in serving other people. The BLESS system is considered as an efficient approach to promote the value and dignity in efforts focused on enhancing our communities and regulating business and private behaviors. The BLESS system leverages the decentralized architecture of the blockchain network, which not only allows grassroots individuals to participate rating process of a social credit system (SCS), but also provides tamper proof of transaction data in the trustless network environment. The anonymity in blockchain records also protects individuals from being targeted in the fight against powerful enterprises. Smart contract enabled authentication and authorization strategy prevents any unauthorized entity from accessing the credit system. The BLESS scheme is promising to offer a secure, transparent and decentralized SCS.

cs.CY↗

arXiv2018

A Survey on Simulation Tools and Testbeds for Cognitive Radio Networks Study

Efficient utility of radio spectrum has been a hot topic as the wireless communication spectrum is a precious resource. The past decade has witnessed intensive research in spectrum sharing techniques. Most of the techniques are based on cognitive radio networks (CRNs) because cognitive capabilities are essential for optimizing spectrum efficiency and guaranteeing safe coexistence in the presence of the spectrum uncertainty. However, due to the high complexity of the problem, most research has been limited to theoretical analysis. It is non-trivial to build a simulator that is capable of carrying out a comprehensive experimental study. In this paper, a survey is conducted to provide a big picture of the available simulators in CRNs research. By illustrating their major functionalities, the insight enables researchers to select tools that match their needs. In addition, with a better understanding of the advantages and constraints, this survey aims at providing a guideline for simulator designers who have been trying to meet the requirements from the CRN research community.

cs.NI↗

arXiv2018

Explore Recurrent Neural Network for PUE Attack Detection in Practical CRN Models

The proliferation of the Internet of Things (IoTs) and pervasive use of many different types of mobile computing devices make wireless communication spectrum a precious resource. In order to accommodate the still fast increasing number of devices requesting wireless connection, more efficient, fine-grained spectrum allocation and sharing schemes are badly in need. Cognitive radio networks (CRNs) have been widely recognized as one promising solution, in which the secondary users (SUs) are allowed to share channels with licensed primary users (PUs) as long as bringing no interference to the normal operations of the PUs. However, malicious attackers or selfish SUs may mimic the behavior of PUs to occupy the channels illegally. It is nontrivial to accurately, timely detect such kind of primary user emulation (PUE) attacks. In this paper, an efficient PUE attacked detection method is introduced leveraging the recurrent neural network (RNN). After a fundamental algorithm using basic RNN, an advanced version taking advantage of long-short-term-memory (LSTM) is proposed, which is more efficient on processing time series with long term memory. The experimental study has provided deeper insights about the different performances the RNNs achieved and validated the effectiveness of the proposed detectors.

cs.NI↗

arXiv2018

Imbalanced Deep Learning by Minority Class Incremental Rectification

Model learning from class imbalanced training data is a long-standing and significant challenge for machine learning. In particular, existing deep learning methods consider mostly either class balanced data or moderately imbalanced data in model training, and ignore the challenge of learning from significantly imbalanced training data. To address this problem, we formulate a class imbalanced deep learning model based on batch-wise incremental minority (sparsely sampled) class rectification by hard sample mining in majority (frequently sampled) classes during model training. This model is designed to minimise the dominant effect of majority classes by discovering sparsely sampled boundaries of minority classes in an iterative batch-wise learning process. To that end, we introduce a Class Rectification Loss (CRL) function that can be deployed readily in deep network architectures. Extensive experimental evaluations are conducted on three imbalanced person attribute benchmark datasets (CelebA, X-Domain, DeepFashion) and one balanced object category benchmark dataset (CIFAR-100). These experimental results demonstrate the performance advantages and model scalability of the proposed batch-wise incremental minority class rectification model over the existing state-of-the-art models for addressing the problem of imbalanced data learning.

cs.CV↗

arXiv2018

An Adaptive Primary User Emulation Attack Detection Mechanism for Cognitive Radio Networks

The proliferation of advanced information technologies (IT), especially the wide spread of Internet of Things (IoTs) makes wireless spectrum a precious resource. Cognitive radio network (CRN) has been recognized as the key to achieve efficient utility of communication bands. Because of the great difficulty, high complexity and regulations in dynamic spectrum access (DSA), it is very challenging to protect CRNs from malicious attackers or selfish abusers. Primary user emulation (PUE) attacks is one type of easy-to-launch but hard-to-detect attacks in CRNs that malicious entities mimic PU signals in order to either occupy spectrum resource selfishly or conduct Denial of Service (DoS) attacks. Inspired by the physical features widely used as the fingerprint of variant electronic devices, an adaptive and realistic PUE attack detection technique is proposed in this paper. It leverages the PU transmission features that attackers are not able to mimic. In this work, the transmission power is selected as one of the hard-to-mimic features due to the intrinsic discrepancy between PUs and attackers, while considering constraints in real implementations. Our experimental results verified the effectiveness and correctness of the proposed mechanism.

cs.NI↗

arXiv2017

Class Rectification Hard Mining for Imbalanced Deep Learning

Recognising detailed facial or clothing attributes in images of people is a challenging task for computer vision, especially when the training data are both in very large scale and extremely imbalanced among different attribute classes. To address this problem, we formulate a novel scheme for batch incremental hard sample mining of minority attribute classes from imbalanced large scale training data. We develop an end-to-end deep learning framework capable of avoiding the dominant effect of majority classes by discovering sparsely sampled boundaries of minority classes. This is made possible by introducing a Class Rectification Loss (CRL) regularising algorithm. We demonstrate the advantages and scalability of CRL over existing state-of-the-art attribute recognition and imbalanced data learning models on two large scale imbalanced benchmark datasets, the CelebA facial attribute dataset and the X-Domain clothing attribute dataset.

cs.CV↗

arXiv2016

Multi-Task Curriculum Transfer Deep Learning of Clothing Attributes

Recognising detailed clothing characteristics (fine-grained attributes) in unconstrained images of people in-the-wild is a challenging task for computer vision, especially when there is only limited training data from the wild whilst most data available for model learning are captured in well-controlled environments using fashion models (well lit, no background clutter, frontal view, high-resolution). In this work, we develop a deep learning framework capable of model transfer learning from well-controlled shop clothing images collected from web retailers to in-the-wild images from the street. Specifically, we formulate a novel Multi-Task Curriculum Transfer (MTCT) deep learning method to explore multiple sources of different types of web annotations with multi-labelled fine-grained attributes. Our multi-task loss function is designed to extract more discriminative representations in training by jointly learning all attributes, and our curriculum strategy exploits the staged easy-to-complex transfer learning motivated by cognitive studies. We demonstrate the advantages of the MTCT model over the state-of-the-art methods on the X-Domain benchmark, a large scale clothing attribute dataset. Moreover, we show that the MTCT model has a notable advantage over contemporary models when the training data size is small.

cs.CV↗