arXiv subjects

Nan Yang

Publications and source records attributed to Nan Yang.

At least 217 records · Page 12Linked to original sources

arXiv2017

Simplified Cooperative Detection for Multi-Receiver Molecular Communication

Diffusion-based molecular communication (MC) systems experience significant reliability losses. To boost the reliability, an MC scheme where multiple receivers (RXs) work cooperatively to decide the signal of a transmitter (TX) by sending the same type of molecules to a fusion center (FC) is proposed in this paper. The FC observes the total number of molecules received and compares this number with a threshold to determine the TX's signal. The proposed scheme is more bio-realistic and requires relatively low computational complexity compared to existing cooperative schemes where the RXs send and the FC recognizes different types of molecules. Asymmetric and symmetric topologies are considered, and closed-form expressions are derived for the global error probability for both topologies. Results show that the trade-off for simplified computations leads to a slight reduction in error performance, compared to the existing cooperative schemes.

cs.IT↗

arXiv2017

Effect of Local Population Uncertainty on Cooperation in Bacteria

Bacteria populations rely on mechanisms such as quorum sensing to coordinate complex tasks that cannot be achieved by a single bacterium. Quorum sensing is used to measure the local bacteria population density, and it controls cooperation by ensuring that a bacterium only commits the resources for cooperation when it expects its neighbors to reciprocate. This paper proposes a simple model for sharing a resource in a bacterial environment, where knowledge of the population influences each bacterium's behavior. Game theory is used to model the behavioral dynamics, where the net payoff (i.e., utility) for each bacterium is a function of its current behavior and that of the other bacteria. The game is first evaluated with perfect knowledge of the population. Then, the unreliability of diffusion introduces uncertainty in the local population estimate and changes the perceived payoffs. The results demonstrate the sensitivity to the system parameters and how population uncertainty can overcome a lack of explicit coordination.

q-bio.CB↗

arXiv2017

On the Design of Secure Non-Orthogonal Multiple Access Systems

This paper proposes a new design of non-orthogonal multiple access (NOMA) under secrecy considerations. We focus on a NOMA system where a transmitter sends confidential messages to multiple users in the presence of an external eavesdropper. The optimal designs of decoding order, transmission rates, and power allocated to each user are investigated. Considering the practical passive eavesdropping scenario where the instantaneous channel state of the eavesdropper is unknown, we adopt the secrecy outage probability as the secrecy metric. We first consider the problem of minimizing the transmit power subject to the secrecy outage and quality of service constraints, and derive the closed-form solution to this problem. We then explore the problem of maximizing the minimum confidential information rate among users subject to the secrecy outage and transmit power constraints, and provide an iterative algorithm to solve this problem. We find that the secrecy outage constraint in the studied problems does not change the optimal decoding order for NOMA, and one should increase the power allocated to the user whose channel is relatively bad when the secrecy constraint becomes more stringent. Finally, we show the advantage of NOMA over orthogonal multiple access in the studied problems both analytically and numerically.

cs.IT↗

arXiv2017

Ambiguity set and learning via Bregman and Wasserstein

Construction of ambiguity set in robust optimization relies on the choice of divergences between probability distributions. In distribution learning, choosing appropriate probability distributions based on observed data is critical for approximating the true distribution. To improve the performance of machine learning models, there has recently been interest in designing objective functions based on Lp-Wasserstein distance rather than the classical Kullback-Leibler (KL) divergence. In this paper, we derive concentration and asymptotic results using Bregman divergence. We propose a novel asymmetric statistical divergence called Wasserstein-Bregman divergence as a generalization of L2-Wasserstein distance. We discuss how these results can be applied to the construction of ambiguity set in robust optimization.

stat.ML↗

arXiv2017

Selective Encoding for Abstractive Sentence Summarization

We propose a selective encoding model to extend the sequence-to-sequence framework for abstractive sentence summarization. It consists of a sentence encoder, a selective gate network, and an attention equipped decoder. The sentence encoder and decoder are built with recurrent neural networks. The selective gate network constructs a second level sentence representation by controlling the information flow from encoder to decoder. The second level representation is tailored for sentence summarization task, which leads to better performance. We evaluate our model on the English Gigaword, DUC 2004 and MSR abstractive sentence summarization datasets. The experimental results show that the proposed selective encoding model outperforms the state-of-the-art baseline models.

cs.CL↗

arXiv2017

Neural Question Generation from Text: A Preliminary Study

Automatic question generation aims to generate questions from a text passage where the generated questions can be answered by certain sub-spans of the given passage. Traditional methods mainly use rigid heuristic rules to transform a sentence into related questions. In this work, we propose to apply the neural encoder-decoder model to generate meaningful and diverse questions from natural language sentences. The encoder reads the input text and the answer position, to produce an answer-aware input representation, which is fed to the decoder to generate an answer focused question. We conduct a preliminary study on neural question generation from text with the SQuAD dataset, and the experiment results show that our method can produce fluent and diverse questions.

cs.CL↗

arXiv2017

Existence and stability of circular orbits in general static and spherically symmetric spacetimes

The existence and stability of circular orbits (CO) in static and spherically symmetric (SSS) spacetime are important because of their practical and potential usefulness. In this paper, using the fixed point method, we first prove a necessary and sufficient condition on the metric function for the existence of timelike COs in SSS spacetimes. After analyzing the asymptotic behavior of the metric, we then show that asymptotic flat SSS spacetime that corresponds to a negative Newtonian potential at large $r$ will always allow the existence of CO. The stability of the CO in a general SSS spacetime is then studied using the Lyapunov exponent method. Two sufficient conditions on the (in)stability of the COs are obtained. For null geodesics, a sufficient condition on the metric function for the (in)stability of null CO is also obtained. We then illustrate one powerful application of these results by showing that an SU(2) Yang-Mills-Einstein SSS spacetime whose metric function is not known, will allow the existence of timelike COs. We also used our results to assert the existence and (in)stabilities of a number of known SSS metrics.

gr-qc↗

arXiv2016

User Load Analysis and Pilot Sequence Design for Multi-Cell Massive MIMO Networks

We propose a novel algorithm to design user load-achieving pilot sequences that mitigate pilot contamination in multi-cell massive multiple-input multiple-output (MIMO) networks. To this end, we first derive expressions for the user load and the load region of the network considering both small-scale and large-scale propagation effects. We then develop the pilot sequence algorithm for multi-cell massive MIMO networks as per the rules of generalized Welch bound equality design. Notably, we find that our algorithm and the corresponding downlink power allocation ensure that the user load is achieved when the signal-to-interference-plus-noise ratio (SINR) requirements for the users lie within the load region. Furthermore, we demonstrate the performance advantage of our proposed design relative to the existing designs, in terms of a larger load region and a higher maximum permitted SINR. Finally, we show that our proposed design can satisfy the pre-defined SINR requirements for users with a finite number of antennas at the base station (BS), while the existing designs cannot satisfy the same requirements even with an infinite number of antennas at the BS.

cs.IT↗

arXiv2016

Artificial-Noise-Aided Secure Transmission in Wiretap Channels with Transmitter-Side Correlation

This work for the first time examines the impact of transmitter-side correlation on the artificial-noise-aided secure transmission, based on which a new power allocation strategy for artificial noise (AN) is devised for physical layer security enhancement. Specifically, we design a correlation-based power allocation (CPA) for AN, of which the optimality in terms of achieving the minimum secrecy outage probability is analytically proved in the large system regime with the number of transmit antennas approaching infinity. In order to fully reveal the benefits of the CPA, we derive easy-to-evaluate expressions for the secrecy outage probability achieved by the CPA. Our study demonstrates that the CPA is nearly optimal and significantly outperforms the widely-used uniform power allocation (UPA) even for a moderately small number of correlated transmit antennas. Furthermore, our numerical results reveal a fundamental difference between the CPA and UPA. That is when the number of correlated transmit antennas increases, the secrecy outage probability of the CPA always reduces while the secrecy outage probability of the UPA suffers from a saturation point.

cs.IT↗

arXiv2016

Mitigating Pilot Contamination Through Location-Aware Pilot Assignment in Massive MIMO Networks

We propose a novel location-aware pilot assignment scheme to mitigate pilot contamination in massive multiple-input multiple-output (MIMO) networks, where the channels are subjected to Rician fading. Our proposed scheme utilizes the location information of users as the input to conduct pilot assignment in the network. Based on the location information, we first determine the line of sight (LOS) interference between the intended signal and the interfering signal. Our analysis reveals that the LOS interference converges to zero as the number of antennas at the base station (BS) goes to infinity, whereas for finite number of antennas at the BS the LOS interference indeed depends on specific pilot allocation strategies. Following this revelation, we assign pilot sequences to all the users in the massive MIMO network such that the LOS interference is minimized for finite number of antennas at the BS. Our proposed scheme outperforms the random pilot assignment in terms of achieving a much higher uplink sum rate for reasonable values of the Rician K-factor. Moreover, we propose a new performance metric, which measures the strength of the LOS interference and demonstrate that it is a good metric to analyze the performance of pilot assignment schemes. Theoretical analysis is provided and verified through numerical simulations to support the proposed scheme.

cs.IT↗

arXiv2016

Multi-Cell Multiuser Massive MIMO Networks: User Capacity Analysis and Pilot Design

We propose a novel pilot sequence design to mitigate pilot contamination in multi-cell multiuser massive multiple-input multiple-output networks. Our proposed design generates pilot sequences in the multi-cell network and devises power allocation at base stations (BSs) for downlink transmission. The pilot sequences together with the power allocation ensure that the user capacity of the network is achieved and the pre-defined signal-to-interference-plus-noise ratio (SINR) requirements of all users are met. To realize our design, we first derive new closed-form expressions for the user capacity and the user capacity region. Built upon these expressions, we then develop a new algorithm to obtain the required pilot sequences and power allocation. We further determine the minimum number of antennas required at BSs to achieve certain SINR requirements of all users. Numerical results are presented to corroborate our analysis and to examine the impact of key parameters, such as the pilot sequence length and the total number of users, on the network performance. A pivotal conclusion is reached that our design achieves a larger user capacity region than the existing designs and needs less antennas at the BS to fulfill the pre-defined SINR requirements of all users in the network than the existing designs.

cs.IT↗

arXiv2016

Inflationary models with non-minimally derivative coupling

We derive the general formulae for the the scalar and tensor spectral tilts to the second order for the inflationary models with non-minimally derivative coupling without taking the high friction limit. The non-minimally kinetic coupling to Einstein tensor brings the energy scale in the inflationary models down to be sub-Planckian. In the high friction limit, the Lyth bound is modified with an extra suppression factor, so that the field excursion of the inflaton is sub-Planckian. The inflationary models with non-minimally derivative coupling are more consistent with observations in the high friction limit. In particular, with the help of the non-minimally derivative coupling, the quartic power law potential is consistent with the observational constraint at 95\% CL.

gr-qc↗

arXiv2016

Gravitational lensing of massive particles in Schwarzschild gravity

Both massless light ray and objects with nonzero mass experience trajectory bending in a gravitational field. In this work the bending of trajectories of massive objects in a Schwarzschild spacetime and the corresponding gravitational lensing (GL) effects are studied. A particle sphere for Schwarzschild black hole (BH) is found with its radius a simple function of the particle velocity and proportional to the BH mass. A single master formula for both the massless and massive particle bending angle is found, in the form of an elliptic function depending only on the velocity and impact parameter. This bending angle is expanded in both large and small velocity limits and large and small impact parameter limits. The corresponding deflection angle for weak and strong GL of massive particles are analyzed, and their corrections to the light ray deflection angles are obtained. The dependence of the deflection angles on the source angle and the particle speed is investigated. Finally we discuss the potential applications of the results in hypervelocity star observations and in determining mass/mass hierarchy of slow particles/objects.

gr-qc↗

arXiv2016

Artificial-Noise-Aided Transmission in Multi-Antenna Relay Wiretap Channels with Spatially Random Eavesdroppers

We design a new secure transmission scheme in the relay wiretap channel where a source communicates with a destination through a decode-and-forward relay in the presence of spatially random-distributed eavesdroppers. For the sake of practicality, we consider a general antenna configuration in which the source, relay, destination, and eavesdroppers are equipped with multiple antennas. In order to confuse the eavesdroppers, we assume that both the source and the relay transmit artificial noise signals in addition to information signals. We first derive a closed-form expression for the transmission outage probability and an easy-to-compute expression for the secrecy outage probability. Notably, these expressions are valid for an arbitrary number of antennas at the source, relay, and destination. We then derive simple yet valuable expressions for the asymptotic transmission outage probability and the asymptotic secrecy outage probability, which reveal the secrecy performance when the number of antennas at the source grows sufficiently large. Using our expressions, we quantify a practical performance metric, namely the secrecy throughput, under a secrecy outage probability constraint. We further determine the system and channel parameters that maximize the secrecy throughput, leading to analytical security solutions suitable for real-world deployment.

cs.IT↗

arXiv2016

Distributed Cooperative Detection for Multi-Receiver Molecular Communication

In this paper, a cooperative diffusion-based molecular communication system is considered where distributed receivers collaboratively determine a transmitter's signal. In this system, the receivers first make local hard decisions about the current transmitted bit and then report these decisions to a fusion center (FC). The FC combines the local hard decisions to make a global decision using an N-out-of-K fusion rule. Asymmetric and symmetric topologies are considered and for each topology, two reporting scenarios, namely, perfect reporting and noisy reporting, are addressed. Closed-form analytical expressions for the expected global error probability are derived for all considered topologies and scenarios. Numerical and simulation results show that system reliability can be greatly improved by combining the detection information of distributed receivers.

cs.IT↗

arXiv2016

Brains and pseudorandom generators

In a pioneering classic, Warren McCulloch and Walter Pitts proposed a model of the central nervous system; motivated by EEG recordings of normal brain activity, Chv\' atal and Goldsmith asked whether or not this model can be engineered to provide pseudorandom number generators. We supply evidence suggesting that the answer is negative.

math.DS↗

arXiv2016

McCulloch-Pitts brains and pseudorandom functions

In a pioneering classic, Warren McCulloch and Walter Pitts proposed a model of the central nervous system. Motivated by EEG recordings of normal brain activity, Chvátal and Goldsmith asked whether or not these dynamical systems can be engineered to produce trajectories which are irregular, disorderly, apparently unpredictable. We show that they cannot build weak pseudorandom functions.

math.DS↗

arXiv2015

Jointly Modeling Topics and Intents with Global Order Structure

Modeling document structure is of great importance for discourse analysis and related applications. The goal of this research is to capture the document intent structure by modeling documents as a mixture of topic words and rhetorical words. While the topics are relatively unchanged through one document, the rhetorical functions of sentences usually change following certain orders in discourse. We propose GMM-LDA, a topic modeling based Bayesian unsupervised model, to analyze the document intent structure cooperated with order information. Our model is flexible that has the ability to combine the annotations and do supervised learning. Additionally, entropic regularization can be introduced to model the significant divergence between topics and intents. We perform experiments in both unsupervised and supervised settings, results show the superiority of our model over several state-of-the-art baselines.

cs.CL↗