With a vast majority of Internet connections shorter than 10 segments, designing a new fast start-up TCP mechanism is a major concern. While enlarging the Initial Window (IW) up to 10 segments is the fastest solution to deal with a short-lived connection in uncongested networks, numerous researchers are concerned about the impact of the large initial burst on congested networks.

New mobile technology generations succeed in achieving high goodput, which results in diverse applications profiles exploiting various resource providers (Wifi, 4G, 5G, . . . ). Badly set parameters on one of the network component may severely impact on the transmission delay and reduce the quality of experience. The cross layer impact should be investigated on to assess the origin of latency. To run cross-layer (from physical layer to application layers) simulations, two approaches are possible: (1) use physical layer models that may not be exhaustive enough to drive consistent analysis or (2) use real physical traces. Driving realistic measurements by using real physical (MAC/PHY) traces inside network simulations is a complex task. We propose to cope with this problem by introducing Cross Layer InFormation Tool (CLIFT), that translates real physical events from a given trace in order to be used inside a network simulator such as ns-2. Our proposal enables to accurately perform analysis of the impact of link layer reliability schemes (obtained by the use of real physical traces) on transport layer performance and on the latency. Such approach enables a better understanding of the interactions between the layers. The main objective of CLIFT is to let us study the protocols introduced at each layer of the OSI model and study their interaction. We detail the internal mechanisms and the benefits of this software with a running example on 4G satellite communications scenarios.

Leon Chua introduced memristors in 1971 [1] as an ideal two-terminal circuit element in complement to the already known three basic circuit elements: resistor, inductor and capacitor (RLC). Memristors are defined by a non-linear memristance that relates the flux (or integral of voltage across the device) to the charge (or integral of current in the device). Because of this definition the memristor will generate passive intermodulation products and their power will depend on the memory of the past current that is contained in the device.

In the context of satellite communications, random access methods can significantly increase throughput and reduce latency over the network. The recent random access methods are based on multi-user multiple access transmission at the same time and frequency followed by iterative interference cancellation and decoding at the receiver. Generally, it is assumed that perfect knowledge of the interference is available at the receiver. In practice, the interference term has to be accurately estimated to avoid performance degradation. Several estimation techniques have been proposed lately in the case of superimposed signals. In this paper, we present an overview on existing channel estimation methods and we propose an improved channel estimation technique that combines estimation using an autocorrelation based method and the Expectation-Maximization algorithm, and uses pilot symbol assisted modulation to further improve the performance and achieve optimal interference cancellation.

Delay/Doppler radar altimetry is a new technology that has been receiving an increasing interest, especially since the launch of Cryosat-2 in 2010 , the first altimeter using this technique. The Delay/Doppler technique aims at reducing the measurement noise and increasing the along-track resolution in comparison with conventional pulse limited altimetry. A new semi-analytical model with five parameters has been recently introduced for this new technology. However, two of these parameters are highly correlated resulting in bad estimation performance when estimating all parameters. This paper proposes a new strategy improving estimation performance for delay/Doppler altimetry. The proposed strategy exploits all the information contained in the delay/Doppler domain. A comparison with other classical algorithms (using the temporal samples only) allows to appreciate the gain in estimation performance obtained when using both temporal and Doppler data.

This paper presents an algorithm based on sparse representation for fusing hyperspectral and multispectral images. The observed images are assumed to be obtained by spectral or spatial degradations of the high resolution hyperspectral image to be recovered. Based on this forward model, the fusion process is formulated as an inverse problem whose solution is determined by optimizing an appropriate criterion. To incorporate additional spatial information within the objective criterion, a regularization term is carefully designed, relying on a sparse decomposition of the scene on a set of dictionaries. The dictionaries and the corresponding supports of active coding coefficients are learned from the observed images. Then, conditionally on these dictionaries and supports, the fusion problem is solved by iteratively optimizing with respect to the target image (using the alternating direction method of multipliers) and the coding coefficients. Simulation results demonstrate the efficiency of the proposed fusion method when compared with the state-of-the-art.

In this paper, we address the design of a cooperative protocol for a hybrid satellite/terrestrial emergency system. We want to perform energy savings compared to the case where all the terrestrial relay nodes are forwarding satellite messages to ground receivers. This is done via the selection of relevant relay nodes. The parameterization of the protocol phases has been done through simulations and takes into account the duration of the selection process, the number of selected nodes, and the signaling overhead. The selection process based on a node identifier (ID) appears to provide greater energy savings compared to the selection process based on the signal to interference and noise ratio (SINR). The solutions have been implemented in the real case scenario of forest fire that has been thoroughly documented by the US administration. According to the scenario parameters, 100% of the masked nodes are reached after cooperation.

This article addresses the problem of signal reconstruction, spectral estimation and linear filtering directly from irregularly-spaced samples of a continuous signal (or autocorrelation function in the case of random signals) when signal spectrum is assumed to be bounded. The number 2L of samples is assumed to be large enough so that the variation of the spectrum on intervals of width π/L is small. Reconstruction formulas are based on PNS (Periodic Nonuniform Sampling) schemes. They allow for reconstruction schemes not requiring regular resampling and suppress two stages in classical computations. The presented method can also be easily generalized to spectra in symmetric frequency bands (bandpass signals).

In this paper, we derive an asymptotic analysis for a capacity approaching design of serially concatenated turbo schemes that works for both systematic generalized irregular repeat accumulate (GIRA) and low density generator matrix (LDGM) codes concatenated with a continuous phase modulation (CPM). The proposed design is based on a semi analytic EXIT chart optimization method. By considering a particular scheduling, inserting partial interleavers between GIRA accumulator and CPM and allowing degree-1 variable nodes, we show that designed rates perform very close to the maximum achievable rate (R∗).

A powerful approach to sparse representation, dictionary learning consists in finding a redundant frame in which the representation of a particular class of images is sparse. In practice, all algorithms performing dictionary learning iteratively estimate the dictionary and a sparse representation of the images using this dictionary. However, the numerical complexity of dictionary learning restricts its use to atoms with a small support. A way to alleviate these issues is introduced in this paper, consisting in dictionary atoms obtained by translating the composition of K convolutions with S-sparse kernels of known support. The dictionary update step associated with this strategy is a non-convex optimization problem, which we study here. A block-coordinate descent or Gauss-Seidel algorithm is proposed to solve this problem, whose search space is of dimension KS, which is much smaller than the size of the image. Moreover, the complexity of the algorithm is linear with respect to the size of the image, allowing larger atoms to be learned (as opposed to small patches). An experiment is presented that shows the approximation of a large cosine atom with K = 7 sparse kernels, demonstrating a very good accuracy.