Distributed detection fusion via Monte Carlo importance sampling

  title={Distributed detection fusion via Monte Carlo importance sampling},
  author={Hang Rao and Xiaojing Shen and Yunmin Zhu and Jianxin Pan},
  journal={2016 35th Chinese Control Conference (CCC)},
Distributed detection fusion with high-dimension conditionally dependent observations is known to be a challenging problem. When a fusion rule is fixed, this paper attempts to make progress on this problem for the large sensor networks by proposing a new Monte Carlo framework. Through the Monte Carlo importance sampling, we derive a necessary condition for optimal sensor decision rules in the sense of minimizing the approximated Bayesian cost function. Then, a Gauss-Seidel/person-by-person… 

Figures from this paper

Distributed detection fusion with nonideal channels under Monte Carlo framework
It is proved that the discretized algorithm is finitely convergent and the proposed method allows us to design the sensor networks with a higher dimensional joint probability density function of the sensor observations.
Analytic Sensor Rules for Optimal Distributed Decision Given $K$-Out-of-$L$ Fusion Rule Under Monte Carlo Approximation
A Monte Carlo importance sampling method to reduce the computational complexity of the distributed decision fusion of conditionally dependent and high dimensional sensor observations and a group of analytic sensor rules based on the necessary and sufficient condition of the optimal sensor rules.
2 Decentralize CFAR Detection Fusion Algorithm 2 . 1 False Alarm Likelihood Function ( FALF )
A fusion detection algorithm that focuses on decentralized CFAR (Constant False Alarm Rate) signal detection problem without prior information is proposed and an almost optimal detection performance is obtained.
A Visual Approach to Data Fusion in Sensor Networks
A novel approach for data fusion in sensor networks that processes anonymous data from distance based sensors such as explosive detectors to detect the presence of an attacker carrying an explosive at a given time and analyze movement over time in an learning environment.


A New Framework for Distributed Detection With Conditionally Dependent Observations
A new framework for distributed detection that builds on a hierarchical conditional independence model is proposed that unifies distributed detection with dependent or independent observations through the introduction of a hidden variable that induces conditional independence among the sensor observations.
Decentralized detection by a large number of sensors
  • J. Tsitsiklis
  • Computer Science, Mathematics
    Math. Control. Signals Syst.
  • 1988
This work considers the decentralized detection problem, in which N independent, identical sensors transmit a finite-valued function of their observations to a fusion center which then decides which one ofM hypotheses is true, and shows how the optimal number of sensors in each group may be determined by solving a mathematical programming problem.
Unexpected properties and optimum-distributed sensor detectors for dependent observation cases
A discretized iterative algorithm is suggested that can provide approximate solutions to the necessary conditions for optimum distributed sensor decision rules under a fixed fusion rule, and the algorithm is shown to converge in a finite number of iterations.
Distributed Detection and Fusion in a Large Wireless Sensor Network of Random Size
A decision fusion rule is proposed that uses the total number of detections reported by local sensors as a statistic for hypothesis testing and can achieve a very good system-level detection performance even at very low signal-to-noise ratio (SNR).
Decentralized detection in sensor networks
A binary decentralized detection problem in which a network of wireless sensors provides relevant information about the state of nature to a fusion center, and it is shown that having a set of identical binary sensors is asymptotically optimal, as the number of observations per sensor goes to infinity.
Optimum distributed detection of weak signals in dependent sensors
Locally optimum (LO) distributed detection is considered for observations that are dependent from sensor to sensor and solutions indicate that the LO sensor detector nonlinearities, in general, contain a term proportional to f'/f, where f is the noise probability density function.
A Near-Optimal Iterative Algorithm via Alternately Optimizing Sensor and Fusion Rules in Distributed Decision Systems
A computationally efficient iterative algorithm is proposed to simultaneously/alternately search for a fusion rule and sensor compression rules by combining both Zhu's and Chair and Varshney's methods and is extended to multiple bit compression and Network decision systems.
Further Results on the Optimality of the Likelihood-Ratio Test for Local Sensor Decision Rules in the Presence of Nonideal Channels
It is shown that the optimal detection performance can be achieved by employing likelihood-ratio quantizers (LRQ) as local decision rules under both the Bayesian criterion and Neyman-Pearson (NP) criterion even for the cases where the channels between the fusion center and local sensors are dependent and noisy.
Optimum multisensor fusion of correlated local decisions
A distributed detection system consisting of a number of local detectors and a fusion center and the optimum decision fusion rule in the Neyman-Pearson sense is derived and analyzed.
The good, bad and ugly: distributed detection of a known signal in dependent Gaussian noise
Bayes-optimal binary quantization for the detection of a shift in mean in a pair of dependent Gaussian random variables is studied, and it is seen that in certain situations, an XOR fusion rule is optimal, and in these cases, the implied decision rule is bizarre.