We address the problem of energy consumption reduction for wireless sensor networks, where each of the sensors has limited power and acquires data that should be transmitted to a central node. The final goal is to have a reconstructed version of the data measurements at the central node, with the sensors spending as little energy as possible, for a given data reconstruction accuracy. In our scenario, sensors in the network have a choice of different coding schemes to achieve varying levels of compression. The compression algorithms considered are based on the lifting factorization of the wavelet transform, and exploit the natural data flow in the network to aggregate data by computing partial wavelet coefficients that are refined as data flows towards the central node. The proposed algorithm operates by first selecting a routing strategy through the network. Then, for each route, an optimal combination of data representation algorithms i.e. assignment at each node, is selected. A simple heuristic is used to determine the data representation technique to use once path merges are taken into consideration. We demonstrate that by optimizing the coding algorithm selection the overall energy consumption can be significantly reduced when compared to the case when data is just quantized and forwarded to the central node. Moreover, the proposed algorithm provides a tool to compare different routing techniques and identify those that are most efficient overall, for given node locations. We evaluate the algorithm using both a second-order autoregressive (AR) model and empirical data from a real wireless sensor network deployment.

In this paper, we address the problem of no-reference quality assessment for digital pictures corrupted with blur. We start with the generation of a large real image database containing pictures taken by human users in a variety of situations, and the conduction of subjective tests to generate the ground truth associated to those images. Based upon this ground truth, we select a number of high quality pictures and artificially degrade them with different intensities of simulated blur (gaussian and linear motion), totalling 6000 simulated blur images. We extensively evaluate the performance of state-of-the-art strategies for no-reference blur quantification in different blurring scenarios, and propose a paradigm for blur evaluation in which an effective method is pursued by combining several metrics and low-level image features. We test this paradigm by designing a no-reference quality assessment algorithm for blurred images which combines different metrics in a classifier based upon a neural network structure. Experimental results show that this leads to an improved performance that better reflects the images' ground truth. Finally, based upon the real image database, we show that the proposed method also outperforms other algorithms and metrics in realistic blur scenarios.

We address the problem of compression for wireless sensor networks, where each of the sensors has limited power, and acquires data that should be sent to a central node. The final goal is to have a reconstructed version of the sampled field at the central node, with the sensors spending as little energy as possible. We propose a distributed compression algorithm for multihop, distributed sensor networks based on the lifting factorization of the wavelet transform that exploits the natural data flow in the network to aggregate data by computing partial wavelet coefficients that are refined as the data flows towards the central node. A key result of our work is that by performing partial computations we greatly reduce unnecessary transmission, significantly reducing the overall energy consumption.

We address a scenario where energy-constrained sensors in a wireless sensor network can choose among different distributed coding schemes to encode their data. We propose a framework where the network is described as a graph, with sensors representing the nodes, and where communication and processing costs are associated to edge weights and the coding schemes associated to states of operation. After describing data transitions and edge costs, we show that a shortest-path algorithm can be used to find the optimum network configuration, i.e., the one that leads to the lowest overall energy consumption.

International audienceThis work deals with the qualitative analysis of a nonlinear integro-differential model of immune competition with special attention to the dynamics of tumor cells contrasted by the immune system. The analysis gives evidence of how initial conditions and parameters influence the asymptotic behavior of the solutions