Cooperative jamming is deemed as a promising physical layer-based approach to secure wireless transmissions in the presence of eavesdroppers. In this paper, we investigate cooperative jamming in a two-tier 5G heterogeneous network (HetNet), where the macrobase stations (MBSs) at the macrocell tier are equipped with large-scale antenna arrays to provide space diversity and the local base stations (LBSs) at the local cell tier adopt non-orthogonal multiple access (NOMA) to accommodate dense local users (LUs). In the presence of imperfect channel state information, we propose three robust secrecy transmission algorithms that can be applied to various scenarios with different security requirements. The first algorithm employs robust beamforming (RBA) that aims to optimize the secrecy rate of a marcouser (MU) in a macrocell. The second algorithm provides robust power allocation (RPA) that can optimize the secrecy rate of an LU in a local cell. The third algorithm tackles a robust joint optimization (RJO) problem across tiers that seek the maximum secrecy sum rate of a target MU and a target LU robustly. We employ convex optimization techniques to find feasible solutions to these highly non-convex problems. The numerical results demonstrate that the proposed algorithms are highly effective in improving the secrecy performance of a two-tier HetNet.

Multiple-Input Multiple-Output Cooperative Cognitive Radio Networks (MIMO-CCRNs) have been proposed recently to further improve the spectrum efficiency. In MIMO-CCRNs, secondary users (SUs) equipped with multiple antennas can cooperatively relay the primary signals for the primary users (PUs) while concurrently accessing the same spectrum to send the secondary traffic for themselves. This communication model is intrinsically vulnerable to eavesdropping as more wiretapping opportunities can be exploited by attackers. In this paper, we propose to employ cooperative jamming at physical (PHY) layer to achieve secure transmissions. In cooperative jamming, certain SUs are employed as helpers to send jamming signals to jam the eavesdropper without interfering the legitimate receivers. We optimize the design of the beamformers (including the information beamformer and the jamming beamformer) as well as the power allocation vector to maximize the secrecy capacity in two scenarios. Simulation results demonstrate that the secrecy capacity is remarkably enhanced in our cooperative jamming scheme. To the best of our knowledge, this work is the first one to investigate the PHY layer security issue for MIMO-CCRNs.

This paper considers energy and information cooperation between a single-antenna primary user (PU) pair and a multiple-antennae secondary user (SU) pair in a cognitive radio system. The secondary transmitter (ST) harvests energy from the primary signal and gains opportunity to transmit its own signal in return for helping relay the primary transmitter's (PT) traffic. A time-divided power splitting scheme is proposed to enable the energy and information cooperation with the objective of maximizing the throughput of the SU pair under the energy constraint of the ST and the received signal-to-inference plus noise ratio (SINR) constraint of the primary receiver (PR). Simulation results demonstrate the influence of the time division proportion and the power splitting parameter on the throughput of the SU pair and the PU pair.

In real world, wireless heterogeneous sensor network (HSN) design and information integration are necessary in different applications. Traditionally, wireless sensor networks information integration is set up to passively fuse all received data. Such an approach is computationally challenging and operationally ineffective because improvements in information accuracy are not guaranteed. Opportunistic Sensing (OS) refers to a paradigm for signal and information processing in which a network of sensing systems can automatically discover and select sensor platforms based on an operational scenario. In this paper, we propose theory and algorithms of OS to simplify the HSN design and promote more efficient information integration. We propose an information theoretical criterion for opportunistic sensing in HSN, and show that HSN with correlated modalities needs less number of codewords than that with independent modalities. Our OS algorithm advances autonomous sensing that not only ensures effective utilization of sensing assets but also provides robust optimal performance. We apply our OS algorithm to radar sensor networks for surveillance and monitoring, and show that our approach works very well and much better than other approaches.

Body Area Networks (BANs) are expected to play a major role in patient health monitoring in the near future. Providing an efficient key agreement with the prosperities of plug-n-play and transparency to support secure inter-sensor communications is critical especially during the stages of network initialization and reconfiguration. In this paper, we present a novel key agreement scheme termed Ordered-Physiological-Feature-based Key Agreement (OPFKA), which allows two sensors belonging to the same BAN to agree on a symmetric cryptographic key generated from the overlapping physiological signal features, thus avoiding the pre-distribution of keying materials among the sensors embedded in the same human body. The secret features computed from the same physiological signal at different parts of the body by different sensors exhibit some overlap but they are not completely identical. To overcome this challenge, we detail a computationally efficient protocol to securely transfer the secret features of one sensor to another such that two sensors can easily identify the overlapping ones. This protocol possesses many nice features such as the resistance against brute force attacks. Experimental results indicate that OPFKA is secure, efficient, and feasible. Compared with the state-of-the-art PSKA protocol, OPFKA achieves a higher level of security at a lower computational overhead.

In many practical applications of wireless sensor networks, it is crucial to accomplish the localization of sensors within a given time bound. We find that the traditional definition of relative localization is inappropriate for evaluating its actual overhead in localization time. To address this issue, we define a novel problem called essential localization and present the first rigorous study on the essential localizability of a wireless sensor network within a given time bound. Additionally, we propose an efficient distributed algorithm for time-bounded essential localization over a sensor network and evaluate the performance of the algorithm with analysis and extensive simulation studies.

Motivated by recent advances on Compressive Sensing (CS), we study the sparsity of sense-through-foliage radar signals. Based on CLEAN method, we obtain the impulse response for sense-through-foliage communication channels for three different radars, 200MHz, 400MHz, and UWB radars. Channel impulse responses for the above three different kinds of channels demonstrate that the sense-through-foliage signals are very sparse, which means CS is possible to be applied to sense-through-foliage radar signals to tremendously reduce the sampling rate. We apply CS and linear programming to sparse signal compression and recovery, and it turns out that we could achieve compression ratio of 32:1 with perfect recovery for the UWB radar signals.

This paper addresses the joint selection and power assignment of a largest set of given links which can communicate successfully at the same time under the physical interference model in the duplex (i.e. bidirectional) mode. For the special setting in which all nodes have unlimited maximum transmission power, Halldorsson and Mitra [5] developed an approximation algorithm with a huge constant approximation bound. For the general setting in which all nodes have bounded maximum transmission power, the existence of constant approximation algorithm remains open. In this paper, we resolve this open problem by developing an approximation algorithm which not only works for the general setting of bounded maximum transmission power, but also has a much smaller constant approximation bound.

Target counting in sensor-based surveillance systems is an interesting task that potentially could have many important applications in practice. In such a system, each sensor outputs the number of targets in its sensing region, and the problem is how one can combine all the reported numbers from sensors to provide an estimate of the total number of targets present in the entire monitored area. The main challenge of the problem is how to handle different sensors' outputs that contain some counts of the same targets falling into the overlapped area from these sensors' sensing regions. This paper introduces a statistical approach to estimate the target count in such a surveillance system. Our approach avoids direct handling of the overlapping issue by adopting statistical methods. First, depending on whether or not certain prior knowledge is available regarding the target distribution, the procedure in minimizing the residual sum of squares or kernel regression is used to estimate the distribution of targets. Then the estimated count of the total targets is obtained by the method of likelihood estimation based on a sequence of binomial distributions that are derived from a sampling procedure. Comparisons based on simulations show that our proposed counting approach outperform the state of art counting algorithms. Extensive simulations also show that our proposed approach is very fast and very promising in estimating the target count in sensor-based surveillance systems.

Opportunistic Sensing (OS) refers to a paradigm for signal and information processing in which a network of sensing systems can automatically discover and select sensor platforms based on an operational scenario. In this paper, we propose theory and algorithms of OS for advancing autonomous sensing that not only ensures effective utilization of sensing assets but also provides robust optimal performance. We propose an information theoretical criterion for opportunistic sensing in wireless sensor networks and derive an algorithm for efficient OS. We apply our OS algorithm to radar sensor networks for sense-through-foliage target detection, and show that our OS algorithm works very well and much better than other approach.

This paper addresses the problem of target counting based on the Monte Carlo simulation. We rely on an Accept-Reject process to guide the placement of virtual targets in a virtual sensor field, which has exactly the same sensor layout as the real one. The objective of this construction is to generate a virtual target energy landscape whose shape is close enough to an energy landscape estimated from the real sensor readings. Based on the number of virtual targets placed on the virtual field and the total virtual and real target energy volumes, the number of real targets can be estimated. We consider both single-epoch and multi-epoch sensor readings and our theoretical analysis indicates that by exploiting the information from multiple epochs, our approach yields a target count that approximately converges to the true target count when the number of epochs is large enough. Extensive comparison based simulation study has been performed and the results verify the effectiveness of our target counting algorithms.

In many practical applications of wireless sensor networks, it is crucial to accomplish the localization of sensors within a given time bound. We find that the traditional definition of relative localization is inappropriate for evaluating its actual overhead. To address this problem, we define a novel problem called essential localization, and present the first rigorous study on the essential localizability of a wireless sensor network within a given time bound. We propose an efficient distributed algorithm for time-bounded essential localization over a sensor network, and evaluate the performance of our algorithm with extensive simulations.

BACKGROUND: Most genomic data have ultra-high dimensions with more than 10,000 genes (probes). Regularization methods with L₁ and L(p) penalty have been extensively studied in survival analysis with high-dimensional genomic data. However, when the sample size n << m (the number of genes), directly identifying a small subset of genes from ultra-high (m > 10, 000) dimensional data is time-consuming and not computationally efficient. In current microarray analysis, what people really do is select a couple of thousands (or hundreds) of genes using univariate analysis or statistical tests, and then apply the LASSO-type penalty to further reduce the number of disease associated genes. This two-step procedure may introduce bias and inaccuracy and lead us to miss biologically important genes. RESULTS: The accelerated failure time (AFT) model is a linear regression model and a useful alternative to the Cox model for survival analysis. In this paper, we propose a nonlinear kernel based AFT model and an efficient variable selection method with adaptive kernel ridge regression. Our proposed variable selection method is based on the kernel matrix and dual problem with a much smaller n x n matrix. It is very efficient when the number of unknown variables (genes) is much larger than the number of samples. Moreover, the primal variables are explicitly updated and the sparsity in the solution is exploited. CONCLUSIONS: Our proposed methods can simultaneously identify survival associated prognostic factors and predict survival outcomes with ultra-high dimensional genomic data. We have demonstrated the performance of our methods with both simulation and real data. The proposed method performs superbly with limited computational studies.

Though the TNM (Tumor, Lymph Node, Metastasis) is a widely used staging system for predicting the outcome of cancer patients, it is limited in prediction mainly because it does not integrate multiple prognostic factors. Expanding the TNM now becomes possible due to availability of large cancer patient datasets. In this paper, we introduce a group testing algorithm that can be used to add new prognostic factors while preserving the TNM staging system. Our approach starts with survival and evaluates its relation to potential prognostic factors individually and in various combinations. A demonstration is given for lung cancer.

Accurate prediction of survival rates of cancer patients is often key to stratify patients for prognosis and treatment. Survival prediction is often accomplished by the TNM system that involves only three factors: tumor extent, lymph node involvement, and metastasis. This prediction from the TNM has been limited, mainly because other potential prognostic factors are not used in the system. Based on availability of large cancer datasets, it is possible to establish powerful prediction systems by using machine learning procedures and statistical methods. In this paper, we present a clustering based approach to develop prognostic systems of cancer patients. Our method starts with grouping combinations that are formed using levels of factors recorded in the data. The dissimilarity measure between combinations is obtained through a sequence of data partitions produced by multiple clusterings. This dissimilarity measure is then used with a hierarchical clustering method in order to find clusters of combinations. Prediction of survival is made simply by using the survival function derived from each cluster. Our approach admits multiple factors and provides a practical and useful tool in outcome prediction of cancer patients. A demonstration of use of the proposed method is given for lung cancer patients.

The TNM (tumor, lymph node, metastasis) is a widely used staging system for predicting the outcome of cancer patients. However, the TNM is not accurate in prediction, partially due to the fact of deficient staging within and between stages. Based on the availability of large cancer patient datasets, there is a need to expand the TNM. In this paper, we present a general clustering-based approach to accomplish this task of expansion. This approach admits multiple factors. One major advantage of the approach is that patients within each generated group are homogeneous in terms of survival, so that a more accurate prediction of outcome of patients can be made. A demonstration of use of the proposed method is given for breast cancer patients.

Abstract — Though destructive to network functions, insider attackers are not detectable with only the classic cryptographybased techniques. Many mission-critic sensor network applications demand an effective, light, flexible algorithm for internal adversary identification with only localized information available. The insider attacker detection scheme proposed in this paper meets all the requirements by exploring the spatial correlation existent among the networking behaviors of sensors in close proximity. Our work is exploratory in that the proposed algorithm considers multiple attributes simultaneously in node behavior evaluation, with no requirement on a prior knowledge about normal/malicious sensor activities. Moreover, it is applicationfriendly, which employs original measurements from sensors and can be employed to monitor many aspects of sensor networking behaviors. Our algorithm is purely localized, fitting well to the large-scale sensor networks. Simulation results indicate that internal adversaries can be identifiedwithahighaccuracyand a low false alarm rate when as many as 25 % sensors are misbehaving. I.

Spatial outliers are the spatial objects with distinct features from their surrounding neighbors. Detection of spatial outliers helps reveal valuable information from large spatial data sets. In many real applications, spatial objects can not be simply abstracted as isolated points. They have different boundary, size, volume, and location. These spatial properties affect the impact of a spatial object on its neighbors and should be taken into consideration. In this paper, we propose two spatial outlier detection methods which integrate the impact of spatial properties to the outlierness measurement. Experimental results on a real data set demonstrate the effectiveness of the proposed algorithms.

This paper targets the identification of faulty sensors and detection of the reach of events in sensor networks with faulty sensors. Typical applications include the detection of the transportation front line of a contamination and the diagnosis of network health. We propose and analyze two novel algorithms for faulty sensor identification and fault-tolerant event boundary detection. These algorithms are purely localized and thus scale well to large sensor networks. Their computational overhead is low, since only simple numerical operations are involved. Simulation results indicate that these algorithms can clearly detect the event boundary and can identify faulty sensors with a high accuracy and a low false alarm rate when as many as 20% sensors become faulty. Our work is exploratory in that the proposed algorithms can accept any kind of scalar values as inputs, a dramatic improvement over existing works that take only 0/1 decision predicates. Therefore, our algorithms are generic. They can be applied as long as the "events" can be modelled by numerical numbers. Though designed for sensor networks, our algorithms can be applied to the outlier detection and regional data analysis in spatial data mining.

Fault-tolerant target detection and localization is a challenging task in collaborative sensor networks. The paper introduces our exploratory work toward identifying a stationary target in sensor networks with faulty sensors. We explore both spatial and temporal dimensions for data aggregation to decrease the false alarm rate and improve the target position accuracy. To filter out extreme measurements, the median of all readings in the close neighborhood is used to approximate the local observation to the target. The sensor whose observation is a local maximum computes a position estimate at each epoch. Results from multiple epochs are combined to decrease the false alarm rate further and improve the target localization accuracy. Our algorithms have low computation and communication overheads. A simulation study demonstrates the validity and efficiency of our design.

Gene expression profiles have been successfully applied to class prediction. Due to a large number of genes (features) and a small number of samples in gene expression data, feature selection is essential when performing the prediction task. Many methods have been proposed to select features in microarray data analysis, but there is no unique method which performs uniformly well for all the learning algorithms. It is then practical to find a feature selection method and a learning algorithm that give superior performance. In this paper, we present an integrated scheme to perform the task of class prediction based on gene expression profiles. The scheme incorporates a simple novel feature selection procedure into naive Bayes models. Each selected gene has a high score of discriminatory power determined by the Brown-Forsythe test statistics. Any pair of selected genes have a low correlation. This facilitates the use of the conditional independence among genes assumed by the naive Bayes models. To demonstrate the effectiveness, the proposed scheme was applied to three commonly used expression data sets COLON, OVARIAN, and LEUKEMIA. The results show that the numbers of misclassified samples are 0, and 4, respectively.

We present a novel time-based positioning scheme (TPS) for efficient location discovery in outdoor sensor networks. TPS relies on TDoA (time-difference-of-arrival) of RF signals measured locally at a sensor to detect range differences from the sensor to three base stations. These range differences are averaged over multiple beacon intervals before they are combined to estimate the sensor location through trilateration. A nice feature of this positioning scheme is that it is purely localized: sensors independently compute their positions. We present a statistical analysis of the performance of TPS in noisy environments. We also identify possible sources of position errors with suggested measures to mitigate them. Our scheme requires no time synchronization in the network and minimal extra hardware in sensor construction. TPS induces no communication overhead for sensors, as they listen to three beacon signals passively during each beacon interval. The computation overhead is low, as the location detection algorithm involves only simple algebraic operations over scalar values. TPS is not adversely affected by increasing network size or density and thus offers scalability. We conduct extensive simulations to test the performance of TPS when TDoA measurement errors are normally distributed or uniformly distributed. The obtained results show that TPS is an effective scheme for outdoor sensor self-positioning.

A spatial outlier is a spatially referenced object whose non-spatial attribute values are significantly different from the values of its neighborhood. Identification of spatial outliers can lead to the discovery of unexpected, interesting, and useful spatial patterns for further analysis. One drawback of existing methods is that normal objects tend to be falsely detected as spatial outliers when their neighborhood contains true spatial outliers. We propose a suite of spatial outlier detection algorithms to overcome this disadvantage. We formulate the spatial outlier detection problem in a general way and design algorithms which can accurately detect spatial outliers. In addition, using a real-world census data set, we demonstrate that our approaches can not only avoid detecting false spatial outliers but also find true spatial outliers ignored by existing methods.

A spatial outlier is a spatially referenced object whose non-spatial attribute values are significantly different from the values of its neighborhood. Identification of spatial outliers can lead to the discovery of unexpected, interesting, and useful spatial patterns for further analysis. Previous work in spatial outlier detection focuses on detecting spatial outliers with a single attribute. In the paper, we propose two approaches to discover spatial outliers with multiple attributes. We formulate the multi-attribute spatial outlier detection problem in a general way, provide two effective detection algorithms, and analyze their computation complexity. In addition, using a real-world census data, we demonstrate that our approaches can effectively identify local abnormality in large spatial data sets.