TCP Performance in Optical Packet-switched Networks

Optical packet assembly is a key function to support inter-working between TCP/IP networks and optical packet-switched networks. It is characterized by the assembly delay and by the segment aggregation needed to form an optical packet. These counter-balancing aspects depend on several environment variables, such as the TCP parameters, the access link speed, the optical packet size whose effects are studied in this paper. Performance evaluations are obtained by extensive simulations in terms of send rate of TCP flows, fairness, efficiency, and assembly delay. Some guidelines in the design of optical packets that take into account the results presented are given.     

A new method for the analysis of integrated transmission lines

A flexible and efficient technique for the analysis of transmission lines is described. The transmission line is considered as the union of « elementary waveguides », where the medium is uniformly stratified. The modes are determined by matching the field at the interfaces using a procedure that leads to a linear eigenvalue problem involving a limited number of variables.     

Bioinspired Security Analysis of Wireless Protocols

Fraglets represent an execution model for communication protocols that resembles the chemical reactions in living organisms. The strong connection between their way of transforming and reacting and formal rewriting systems makes a fraglet program amenable to automatic verification. Grounded on past work, this paper investigates feasibility of adopting fraglets as model for specifying security protocols and analysing their properties. In particular, we give concrete sample analyses over a secure RFID protocol, showing evolution of the protocol run as chemical dynamics and simulating an adversary trying to circumvent the intended steps. The results of our analysis confirm the effectiveness of the cryptofraglets framework for the model and analysis of security properties and eventually show its potential to identify and uncover protocol flaws.     

Algorithms for iterative decoding in the presence of strong phase noise

We present two new iterative decoding algorithms for channels affected by strong phase noise and compare them with the best existing algorithms proposed in the literature. The proposed algorithms are obtained as an application of the sum-product algorithm to the factor graph representing the joint a posteriori probability mass function of the information bits given the channel output. In order to overcome the problems due to the presence in the factor graph of continuous random variables, we apply the method of canonical distributions . Several choices of canonical distributions have been considered in the literature. Well-known approaches consist of discretizing continuous variables or treating them as jointly Gaussian, thus obtaining a Kalman estimator. Our first new approach, based on the Fourier series expansion of the phase probability density function, yields better complexity/performance tradeoff with respect to the usual discretized-phase method. Our second new approach, based on the Tikhonov canonical distribution, yields near-optimal performance at very low complexity and is shown to be much more robust than the Kalman method to the placement of pilot symbols in the coded frame. We present numerical results for binary LDPC codes and LDPC-coded modulation, with particular reference to some phase-noise models and coded-modulation formats standardized in the next-generation satellite Digital Video Broadcasting (DVB-S2). These results show that our algorithms achieve near-coherent performance at very low complexity without requiring any change to the existing DVB-S2 standard.     

Toward automatic phenotyping of developing embryos from videos.

We describe a trainable system for analyzing videos of developing C. elegans embryos. The system automatically detects, segments, and locates cells and nuclei in microscopic images. The system was designed as the central component of a fully automated phenotyping system. The system contains three modules 1) a convolutional network trained to classify each pixel into five categories: cell wall, cytoplasm, nucleus membrane, nucleus, outside medium; 2) an energy-based model, which cleans up the output of the convolutional network by learning local consistency constraints that must be satisfied by label images; 3) a set of elastic models of the embryo at various stages of development that are matched to the label images.     

Characterization and modeling of quantum cascade lasers based on aphoton-assisted tunneling transition

A detailed characterization and modeling of long-wavelength (λ~10 μm) quantum cascade (QC) lasers based on a photon-assisted tunneling transition are presented. In particular, the influence of the finite lifetime of the lower state of the laser transition on the current-voltage and threshold current versus temperature characteristics have been studied both theoretically and experimentally. It is shown that, for our structure, the value of the lower state lifetime can be extracted from the voltage-current curve; the value we found was 2.6 ps. In addition, this model allows to understand the abrupt degradation of the performance of the device for T>150 K. Low temperature (T=10 K) threshold current densities of 1.1 kA/cm² and a tuning range of 85 cm^-1 in pulsed mode are reported. In continuous-wave mode, the emission linewidth of a free-running laser was determined to be 3.9 MHz     

Simplified Soft-Output Detection of CPM Signals Over Coherent and Phase Noise Channels

We consider continuous phase modulations (CPMs) in iteratively decoded serially concatenated schemes. Although the overall receiver complexity mainly depends on that of the CPM detector, almost all papers in the literature consider the optimal maximum a posteriori (MAP) symbol detection algorithm and only a few attempts have been made to design low-complexity suboptimal schemes. This problem is faced in this paper by first considering the case of an ideal coherent detection, then extending it to the more interesting case of a transmission over a typical satellite channel affected by phase noise. In both cases, we adopt a simplified representation of an M-ary CPM signal based on the principal pulses of its Laurent decomposition. Since it is not possible to derive the exact detection rule by means of a probabilistic reasoning, the framework of factor graphs (FGs) and the sum-product algorithm (SPA) is used. In the case of channels affected by phase noise, continuous random variables representing the phase samples are explicitly introduced in the FG. By pursuing the principal approach to manage continuous random variables in a FG, i.e., the canonical distribution approach, two algorithms are derived which do not require the presence of known (pilot) symbols, thanks to the intrinsic differential encoder embedded in the CPM modulator.     

Detection of vesicoureteral reflux using microwave radiometry-system characterization with tissue phantoms

Microwave (MW) radiometry is proposed for passive monitoring of kidney temperature to detect vesicoureteral reflux (VUR) of urine that is externally heated by a MW hyperthermia device and thereafter reflows from the bladder to kidneys during reflux. Here, we characterize in tissue-mimicking phantoms the performance of a 1.375 GHz radiometry system connected to an electromagnetically (EM) shielded microstrip log spiral antenna optimized for VUR detection. Phantom EM properties are characterized using a coaxial dielectric probe and network analyzer (NA). Power reflection and receive patterns of the antenna are measured in layered tissue phantom. Receiver spectral measurements are used to assess EM shielding provided by a metal cup surrounding the antenna. Radiometer and fiberoptic temperature data are recorded for varying volumes (10-30 mL) and temperaturesg (40-46°C) of the urine phantom at 35 mm depth surrounded by 36.5°C muscle phantom. Directional receive pattern with about 5% power spectral density at 35 mm target depth and better than -10 dB return loss from tissue load are measured for the antenna. Antenna measurements demonstrate no deterioration in power reception and effective EM shielding in the presence of the metal cup. Radiometry power measurements are in excellent agreement with the temperature of the kidney phantom. Laboratory testing of the radiometry system in temperature-controlled phantoms supports the feasibility of passive kidney thermometry for VUR detection.     

热湿存储环境下须状晶体的生长机理与影响因素

纯哑锡表面的晶须特性已经成为IC封装中无铅焊接工艺实施的主要影响因素之一。由于其它替代方法,例如预先在引脚上镀Ni/Pd/Au层,会导致成本无法预估以及可靠性下降等问题,因此,目前的关注点是进一步减少纯锡表面须状晶体的生长。