08h45 - 09h10
Control and Stabilization of Discrete-time Networked Control Systems under Denial of Service Attack
This paper presents stability analysis and control design for linear discrete time networked control systems due to packet drop. The feedback loop is closed over a network that is susceptible to packet drop due to denial-of-service (DoS) attack or network congestion. Packet drop at the network interface is modeled as an i.i.d Bernoulli random process such that when a packet is dropped, the controller uses the most recent successfully transmitted packet for feedback control. Criterion for closed-loop mean square stability is derived which shows that for a given controller, the closed loop NCS is stable if packet drop probability remains within a certain threshold. This provides a way of validating whether or not the controller would maintain the expected performance under DoS attacks. A Riccati type equation is also derived for minimization of quadratic cost state and control for a given packet drop probability. Numerical results are presented to demonstrate stability and optimum performance of the closed loop system.
09h10 - 09h35
Manifold Constrained Joint Sparse Sensing with K-SVD for Image sets based Face Recognition
Sparse representation based face recognition has recently become an important research topic in computer vision community. However, as the coding process suffers from the perturbations caused by the variations of the input samples, since the consistence of the filters for similar input samples are not well addressed in the existing literature. In this paper, we will address an efficient technique for dictionary learning by fusing multiple features from different visual aspects. In particular, the proposed method merges manifold constraints into the standard dictionary learning framework to impose an orthogonality constraint on the basis matrix and force this optimization process to satisfy the structure preservation requirement of the coefficient matrix. On the other hand, this framework can consistently integrate the manifold constraints during the optimization process. Simultaneously, with regularized data structure constraint, the proposed approach contributes a better solution which is robust to the variance of the input samples against over-complete filters. Extensive experiments on several popular face databases reveal a consistent performance improvement in comparison to some of state-of-the-art algorithms.
09h35 - 10h00
Optimization of sparse beamforming systems
Beamforming is a spatial filtering technique. It is used to enhance the required signal in a particular area via the use of multiple sensors. When a target response is defined, the beamforming design problem can be formulated as an optimization problem and various optimization techniques can be applied. However, when the number of microphone increases and the filters are long, the complexity can grow significantly. It is advantageous if many of the filter coefficients are zeroes so that the implementation can be executed with less computation. In this paper, the sparse design of beamforming systems is studied. The trade-off between the performance of the designed frequency responses and the number of zero elements in the design will be investigated. Numerical results show that sparsity of the designed beamformers can be reduced significantly without affecting very much of the performance.