Jintae Park

Maximin Distributed Detection in the Presence of Impulsive Alpha-Stable Noise

The distributed detection problem in wireless sensor networks is studied under the impulsive α-stable noise assumption. Since symmetric α-stable density does not have a closed form, its approximation, the bi-parameter Cauchy Gaussian mixture model, is used to describe the impulsive behavior of α-stable noises. With this model, we propose a low-complexity robust fusion rule by taking the maximin setting with respect to the detection probability. An explicit formula for the detection probability is derived. Robustness of the proposed maximin fusion rule is justified by numerical and simulation results for α-stable noises.

Distributed Detection and Fusion of Weak Signals in Fading Channels with Non-Gaussian Noises

Distributed detection and information fusion have received recent research interest due to the success of emerging wireless sensor network (WSN) technologies. For the problem of distributed detection in WSNs under energy constraints, a weak signal model in the canonical parallel fusion scheme with additive non-Gaussian noises and fading channels is considered. To solve this problem in the Neyman-Pearson setting, a unified asymptotic fusion rule generalizing the maximum ratio combiner (MRC) fusion rule is proposed. Explicit formulas for the threshold and detection probability applicable for wide classes of fading channels and noise distributions are written out. Both asymptotic analysis and Monte Carlo modeling are used to examine the performance of the proposed detection fusion rule.

Large-Signal Robustness of the Chair-Varshney Fusion Rule Under Generalized-Gaussian Noises

The Chair-Varshney rule (CVR) has been used to provide a large signal-to-noise ratio (SNR) approximation of the optimal fusion rule under Gaussian noise. For more practical use in sensor networks, this paper extends CVR to Generalized-Gaussian noise channels, along with verification of the suboptimality and robustness of CVR under the Generalized-Gaussian channel noise through the use of Monte Carlo simulations.

Fusion of Decisions Modeled as Weak Signals in Wireless Sensor Networks

Distributed detection has newly received research interest due to the success of the emerging wireless sensor network (WSN) technology. To deal with the problem of distributed detection for the WSN having the energy constraint, the fusion of decisions modeled as weak signals is studied. By using the weak signal model and additive non-Gaussian noise channels in the canonical parallel fusion scheme, we propose an asymptotic fusion rule applicable for wide classes of noise probability density functions (pdfs). In the particular case of a known pdf, an optimal detection rule is given. Both asymptotic analysis and Monte Carlo simulation are used to examine the performance of the proposed detection fusion rule.

Robust Distributed Detection with Total Power Constraint in Large Wireless Sensor Networks

In practical problems of signal detection, it is quite common that the underlying noise distribution is not Gaussian and may vary in a wide range from light- to heavy-tailed forms. To design a robust fusion rule for distributed detection in wireless sensor networks, an asymptotic maximin approach is used by introducing weak signals in the canonical parallel fusion model. Explicit formulas for the detection and false alarm probabilities are derived. The analytic results are written out for the classes of nondegenerate, with a bounded variance and contaminated Gaussian noise distributions. Numerical and simulation results are obtained to justify robustness and asymptotic characteristics of the proposed fusion rule.

A robust fusion rule using Piece-Wise Linear function in wireless sensor networks

The problem of decision fusion in wireless sensor networks is investigated in this paper. Based on the parallel fusion model under fading and the noise channels of the generalized Gaussian and the Cauchy models, we develop a robust fusion rule. By utilizing high and low signal-to-noise ratio (SNR) approximations, we obtain both high and low SNR alternatives respectively to the optimum likelihood ratio based fusion statistic for both noise models. To overcome the near-optimality of both alternatives for only limited ranges of SNR values, we propose the Piece-Wise Linear fusion statistic (PWL-FS) that combines both high and low SNR results by using a piece-wise linear function. Performance evaluation is performed through Mote Carlo simulations.

SAMOP: Synchronization avoiding modification of outgoing packets in wireless sensor networks

In wireless sensor networks (WSNs), time synchronization is basically required to provide time-stamp for the reported events, active-sleep MAC protocols to extend network lifetime, etc. Most synchronization schemes assume that the send time of a packet is captured when packet transmission is just started, and the packet is modified to save the send time before packet transmission is not complete. However, it is difficult for system-performance limited sensor nodes to complete the packet modification in time. In this paper, we propose a synchronization scheme avoiding modification of the outgoing packet (SAMOP) based on the CSMA-CA protocol, where a transmitter and a receiver measure only time instance for the packet with their own local clocks and then a transmitter transfers the time measurements with a subsequent packet. Based on multiple time measurements, SAMOP synchronizes local clocks of both nodes. Simulation results show our SAMOP provides improved accuracy with respect to a synchronizing period.

A Note on the Channel Aware Decision Fusion in Wireless Sensor Networks

Decision fusion in wireless sensor networks under non-Gaussian noise channels is studied. Based on the parallel fusion model, the likelihood ratio (LR) based fusion rule is represented and considered as an optimal fusion rule. From this rule, we obtain suboptimum rules by utilizing high and low signal-to-noise ratio (SNR) approximations corresponding to tail and central parts of noise distributions, respectively. For the high SNR case two wide classes of distributions corresponding to two types of tails, exponentially-tailed and Pareto type distributions, and for low SNR case smooth densities in their central part are considered and studied to derive alternative rules to the LR rule with symmetric and unimodal assumptions. Performance evaluation for several fusion rules is performed through simulation