20 June 2018 in IEEE Transactions on Geoscience and Remote Sensing
The supervised deep networks have shown great potential in improving the classification performance. However, training these supervised deep networks is very challenging for hyperspectral image given the fact that usually only a small amount of labeled samples are available. In order to overcome this problem and enhance the discriminative ability of the network, in this paper, we propose a deep network architecture for a super-resolution (SR)-aided hyperspectral image classification with classwise loss (SRCL). First, a three-layer SR convolutional neural network (SRCNN) is employed to reconstruct a high-resolution image from a low-resolution image. Second, an unsupervised triplet-pipeline CNN (TCNN) with an improved classwise loss is built to encourage intraclass similarity and interclass dissimilarity. Finally, SRCNN, TCNN, and a classification module are integrated to define the SRCL, which can be fine-tuned in an end-to-end manner with a small amount of training data. Experimental results on real hyperspectral images demonstrate that the proposed SRCL approach outperforms other state-of-the-art classification methods, especially for the task in which only a small amount of training data are available.
28 October 2021 in IEEE Access
This paper studies a non-linear energy harvesting (EH) enabled full-duplex (FD) cooperative non-orthogonal multiple access (C-NOMA) system, where the near user acts as an FD relay to assist the transmission from BS to far user with power-splitting (PS) protocol. We derive the close-form expressions of outage probability in this considered system, and also give the asymptotic outage probability under high signal-to-noise ratio (SNR) regime. Additionally, we obtain the optimal power allocation parameter to minimize the outage probability, and discuss the impact of self-interference (SI) on outage performance. Moreover, it is observed from simulation results that the presence of SI improve the outage performance, which is caused by the so-called self-energy recycling.
8 October 2020 in Journal of the American Medical Informatics Association
Objective: Normalizing clinical mentions to concepts in standardized medical terminologies, in general, is challenging due to the complexity and variety of the terms in narrative medical records. In this article, we introduce our work on a clinical natural language processing (NLP) system to automatically normalize clinical mentions to concept unique identifier in the Unified Medical Language System. This work was part of the 2019 n2c2 (National NLP Clinical Challenges) Shared-Task and Workshop on Clinical Concept Normalization.Materials and Methods: We developed a hybrid clinical NLP system that combines a generic multilevel matching framework, customizable matching components, and machine learning ranking systems. We explored 2 machine leaning ranking systems based on either ensemble of various similarity features extracted from pretrained encoders or a Siamese attention network, targeting at efficient and fast semantic searching/ranking. Besides, we also evaluated the performance of a general-purpose clinical NLP system based on Unstructured Information Management Architecture.Results: The systems were evaluated as part of the 2019 n2c2 challenge, and our original best system in the challenge obtained an accuracy of 0.8101, ranked fifth in the challenge. The improved system with newly designed machine learning ranking based on Siamese attention network improved the accuracy to 0.8209.Conclusions: We demonstrate the successful practice of combining multilevel matching and machine learning ranking for clinical concept normalization. Our results indicate the capability and interpretability of our proposed approach, as well as the limitation, suggesting the opportunities of achieving better performance by combining general clinical NLP systems.
3 May 2018 in IEEE Systems Journal
In this paper, we consider a two-way relay with multirelays. We first review the joint decoding (JD), decoding by parts (DP), bit-level exclusive-or (XOR), and superposition coding (SC) schemes for the two-way decode-and-forward relay (TWDR), and then propose a new TWDR selection scheme for the two mentioned coding methods. In particular, we focus on the outage performance analysis of the relay selection strategy using the aforementioned coding and decoding methods. Based on an approximate analysis of the outage performance in the high signal-to-noise ratio, we present the diversity-multiplexing tradeoff for the considered decoding and coding schemes. Moreover, we present an optimal power allocation for relays with the SC mechanism. Our simulation results show that our proposed relay selection strategy is outage optimal for all the coding and decoding methods mentioned in this paper. Our results also show that the outage performance under the JD-Xor and JD-SC schemes is more competitive for the case in which the channel statistical characteristics satisfy γ̅ α σ kα 2 ≈ γ̅ b σ kb 2 (γ̅i = P i /N 0 , i ∈ {α, b}, where a and b denote two sources exchanging information with each other, Pi is the average transmitting power from the i source, N 0 is the variance of additive white Gaussian noise obtained by the relay k, and σ ki 2 is the variance of the Rayleigh fading channel between k and i). When γ̅ α σ kα 2 ≫ γ̅ b σ kb 2 or γ̅ α σ kα 2 ≪ γ̅ α σ kα 2 , the outage performance using DP-Xor or DP-SC mechanisms is much better.
8 June 2022 in IEEE Systems Journal
In this article, a novel two-phase transmission protocol is proposed to help the bidirectional transmission in a user-aid two-way cooperative nonorthogonal multiple access system, where superposition coding, bit-level Exclusive-Or operation, and successive interference cancellation (SIC) techniques are employed for improving the spectrum efficiency and throughput. Considering the residual interference generated in the practice process of decoding, both the perfect SIC and the imperfect SIC scenarios are discussed. Besides, the power allocation factor $\alpha _{1}$ and $\alpha _{2}$ are optimized under the perfect SIC to reduce the total outage probability, whose closed-form expression is derived and verified by simulations. Also, the approximated outage performance is derived in imperfect SIC scenario. Furthermore, two other conventional two-phase protocols and one three-phase protocol are provided as the benchmark to compare the outage probability and throughput with the proposed protocol, which shows the superiority of the proposed two-phase protocol and power allocation.
23 September 2024 in IEEE Transactions on Vehicular Technology
Aiming at the serious threat of untrusted users to the system security in the non-orthogonal multiple access (NOMA) downlink system, where an untrusted near user assists a far user to receive data, a cooperative transmission scheme is proposed to improve the physical layer security (PLS) for our presented wiretap NOMA system model. The cooperative transmission scheme is implemented in two communication phases. In the first phase, the reverse power allocation method is used, in which the higher power is allocated to the data transmitted to the near user, and the far user is introduced as a cooperative jammer to send the artificial noise (AN), so as to improve the security when the base station sends data. Considering the situation that the near user decodes its data, two information forwarding processes are designed in the second phase. If the near user can decode its data, it will employ the amplify-and-forward (AF) protocol to forward the remaining signal after removing its own data. Otherwise, two relay selection (RS) strategies are explored to improve reliability of the considered system. In this case, the selected relay adopts the physical-layer network coding (PNC) technique, and the AN sent by the far user is regarded as the key to participating in the bit-level exclusive-or (XOR) operation, which ensures the absolute security of the data transmitted to the far user. The exact closed-form expressions of the outage probability (OP) and intercept probability (IP) are derived respectively, and the approximate results are also given under high signal-to-noise ratio (SNR). Furthermore, the influence of relevant system parameters is discussed concretely. The numerical simulation results indicate that substantial performance gain is achieved over orthogonal multiple access (OMA) and other schemes, and the two RS strategies involved in this paper have the same reliability.