Xiaoyan Wang

Auction-Based Frameworks for Secure Communications in Static and Dynamic Cognitive Radio Networks

This paper investigates the secure communication issue for both static and dynamic cognitive radio networks (CRNs), where multiple nonaltruistic primary users (PUs), secondary users (SUs), and eavesdroppers exist. The design objective is to provide secure communications for PUs and, meanwhile, to ease the starvation of transmission opportunities for SUs. To achieve this goal, we propose a barter-like trading model to incentivize the cooperation among nonaltruistic users. Specifically, PUs leverage the assistance of SUs in the form of cooperative relaying or friendly jamming and, in return, yield certain licensed spectrum accessing time to the aided SUs. We propose a truthful nonmonetary double auction framework (FONDA) toward secure communications for static CRN where PUs and SUs interact in a single round. Then, we extend our framework to d-FONDA for dynamic CRN, where SUs who have patience (tolerant of traffic delay) arrive and leave the network dynamically. We prove that both FONDA and d-FONDA preserve nice economic properties, including truthfulness, individual rationality, and budget balance. Simulation results reveal that the proposed frameworks provide substantial performance gains compared with the baseline scheme and suffer acceptable performance degradation over ideal schemes.

eICIC Configuration Algorithm with Service Scalability in Heterogeneous Cellular Networks

Interference management is one of the most important issues in heterogeneous cellular networks with multiple macro and pico cells. The enhanced inter cell interference coordination (eICIC) has been proposed to protect downlink pico cell transmissions by mitigating interference from neighboring macro cells. Therefore, the adaptive eICIC configuration problem is critical, which adjusts the parameters including the ratio of almost blank subframes (ABS) and the bias of cell range expansion (RE). This problem is challenging especially for the scenario with multiple coexisting network services, since different services have different user scheduling strategies and different evaluation metrics. By using a general service model, we formulate the eICIC configuration problem with multiple coexisting services as a general form consensus problem with regularization and solve the problem by proposing an efficient optimization algorithm based on the alternating direction method of multipliers. In particular, we perform local RE bias adaptation at service layer, local ABS ratio adaptation at BS layer, and coordination among local solutions for a global solution at a network layer. To provide the service scalability, we encapsulate the service details into the local RE bias adaptation subproblem, which is isolated from the other parts of the algorithm, and we also introduce some implementation examples of the subproblem for different services. The extensive simulation results demonstrate the efficiency of the proposed algorithm and verify the convergence property.

QoS-aware resource allocation for multicast service over vehicular networks

Quality of service (QoS) constrained multicast service over vehicular networks has considerable benefit for both road safety and entertainment, while the resource allocation problem of it is challenging due to the high mobility of vehicles. In this paper, we manage the vehicle mobility by dividing one scheduling round into multiple segments, and investigate the resource allocation problem to answer the questions of which modulation and coding scheme (MCS) should be adopted for each flow in each segment, and how to schedule the radio resources among all the flows. We consider two kinds of multicast services. For the multicast service to cover all the recipients, we formulate the problem as a resource scheduling problem with fixed MCS profile, and propose a k-commodity packing based approximation algorithm to solve it with low complexity. For the multicast service with adaptive recipients, we notice that the number of valid MCS profiles for each flow is limited, and propose a heuristic algorithm to search for proper MCS profile assignments. The simulation results verify the efficiency of the proposed scheme.

Fingerprint in the Air: Using the RSS Data for Uniqueness Identification

Indoor localization, device identification, and wireless attendance security systems are widely used in recent years. There is always a premise that each person can carry only one wireless device by himself, which is no longer valid nowadays. To detect the uniqueness identification problem, the bio-assisted methods such as fingerprint, face or gait recognition systems are deployed near the entrance which are difficult to use. This paper studies such problem using RF RSS fingerprints which can be collected and modeled as time series. Then we can calculate the similarity of the time series to judge the uniqueness identification problem. Firstly, a naive algorithm using dynamic time warping is presented to simply compute the similarity the asynchronous time series. Then an improved algorithm is proposed to reduce the computational complexity while keeping the robustness. Simulation and experiments results show that our algorithms can perfectly detect the uniqueness identification problem with a reasonable cost.

Population-Aware Relay Placement for Wireless Multi-Hop Based Network Disaster Recovery

Network disaster recovery is one of the greatest concerns for Mobile Network Operators (MNOs) and first responders during large-scale natural disasters such as earth- quakes. In many recent studies, wireless multi-hop networking has been demonstrated as an effective technique to quickly and efficiently extend the network coverage during disasters. In this paper, we specifically address the network deployment problem by proposing the Population-Aware Relay Placement (PARP) solution, which seeks the efficient deployment of a limited number of relays such that population coverage is maximized in the scenario of network disaster recovery. We provide a graph-based modeling and prove its NP-hardness accordingly. In order to efficiently solve this problem, we propose a heuristic solution, which is constructed in two steps. We first design a simple algorithm based on a disk graph to determine the Steiner locations, which is the biggest challenge in this problem. Then, we formulate the problem as an integer programming problem, which is inspired by the formulation of Prize-Collecting Steiner Tree (PCST). Thus, the integer problem is solved by exploring the similarity of the existing algorithm for PCST. To evaluate the proposed solution extensively, we present numerical results on both real-world and random scenarios, which validate the effectiveness of the proposed solution and show substantial improvement by comparing to the previous one.

Pilot-based reference amplitude compensation for ultra-multi-level PAM-SSB-DFTs-OFDM

Ultra-multi-level PAM-SSB-DFTs-OFDM (Single Side Band DFT spreading Orthogonal Frequency Division Multiplexing) was proposed as one of the attractive approaches to achieve higher spectrum efficiency since it has advantages of lower PAPR (Peak to Average Power Ratio) than OFDM and robustness against reference carrier phase error. However, BER performance degradation caused by reference amplitude error is as large as that of conventional QAM signals, especially for ultra-multi-level modulation. To solve this problem, this paper proposes pilot-based reference amplitude compensation to improve BER performance degradation caused by reference amplitude error. This paper describes pilot signal design and BER performance comparison evaluation when non-linear amplifier is used.

Large Scale Environmental Sound Classification Based on Efficient Feature Extraction

In recent years, plenty of studies endeavor to analyze the life auditory scenarios via mining non-speech sounds. Conventional audio recognition schemes clearly bound the feature extraction and recognition stages, such as in speech recognition. However, such separation leads to inconsistency in the purposes at each stage. The recognition stage contributes to portray the global data distribution focusing on relationship between signal samples. However, such consideration can hardly be embedded into feature extraction process which centered on the local structure, thus, the prominent relation information is destroyed. In this paper, we propose a unified acoustic recognition framework taking advantage of primitive feature input without injuring discriminant information and adopting effective classification scheme accordingly. We formulate the sound into subspace representation and initially adopt Grassmannian distance to classify the subspace-indexed non-speech sounds. To validate the proposed framework, we conducted experiments using RWCP Sound Scene Database. The experimental results demonstrated the proposed framework achieved fine recognition performance with high efficiency.

Spatio-temporal data-driven analysis of mobile network availability during natural disasters

The accurate assessment of mobile network availability during large-scale natural disasters is essential for ensuring effective preparation and fast response. However, traditional network availability assessment models are ideal and cannot effectively take into account the spatio-temporal dynamics of mobile network failures in a disaster scenario. Therefore, their evaluation results are generally inaccurate and of coarse granularity, thus not meeting the strict requirements for disaster preparation and response. In this paper, we propose a data-driven analysis framework for the accurate assessment of mobile network availability by integrating essential geographical features from various sources, e.g., seismic intensity data, buildings and land usage data, base station location data, and many other data in related studies. Furthermore, we explore the spatio-temporal inter-correlations and dynamics of several key factors of network failures and their impacts on network availability by associating them with corresponding geographical features in a disaster scenario. We demonstrate our analysis framework with a synthetic earthquake scenario in the Tokyo area and validate our analysis by comparing to existing studies.