An interval estimate is provided for the Herfindahl-Hirschman Index (HHI) when the knowledge about the market is incomplete, and we know just the largest market shares. Two rigorous bounds are provided for the HHI, wi...
详细信息
An interval estimate is provided for the Herfindahl-Hirschman Index (HHI) when the knowledge about the market is incomplete, and we know just the largest market shares. Two rigorous bounds are provided for the HHI, without any further assumptions. Though the interval gets wider as the sum of the known market shares gets smaller, the estimate proves to be quite tight even when the fraction of the market that we do not know in detail is as high as 30%. This robustness is shown through three examples, considering respectively a set of real data and two sets of synthetic data, with the company sizes (a proxy for market shares) following respectively a Zipf law and a Pareto distribution.
A compact diplexer fabricated on substrate-integrated waveguide (SIW) technology is presented. The diplexer is a sixth order coupled-resonator network with no additional elements such as power dividers, composed exclu...
详细信息
A compact diplexer fabricated on substrate-integrated waveguide (SIW) technology is presented. The diplexer is a sixth order coupled-resonator network with no additional elements such as power dividers, composed exclusively of three dual-mode square SIW cavities. The coupling between the two orthogonal resonant modes of each cavity is generated by means of additional vias perturbing the cavity symmetry, while inductive irises implement the inter-cavity couplings. The coupling topology of the whole diplexer is restricted by the orientation of the resonant modes. A fabricated prototype with passbands around 10 GHz proves that this approach is feasible.
In this paper we utilize Bayesian modeling and inference to learn a softmax classification model which performs Supervised Classification and Active Learning. For p < 1, l_p-priors are used to impose sparsity on th...
详细信息
ISBN:
(纸本)9781479946037
In this paper we utilize Bayesian modeling and inference to learn a softmax classification model which performs Supervised Classification and Active Learning. For p < 1, l_p-priors are used to impose sparsity on the adaptive parameters. Using variational inference, all model parameters are estimated and the posterior probabilities of the classes given the samples are calculated. A relationship between the prior model used and the independent Gaussian prior model is provided. The posterior probabilities are used to classify new samples and to define two Active Learning methods to improve classifier performance: Minimum Probability and Maximum Entropy. In the experimental section the proposed Bayesian framework is applied to Image Segmentation problems on both synthetic and real datasets, showing higher accuracy than state-of-the-art approaches.
With the advent and rapid spread of microblogging services, web information management finds a new research topic. Although classical information retrieval methods and techniques help search engines and services to pr...
详细信息
A compact diplexer fabricated on substrate-integrated waveguide (SIW) technology is presented. The diplexer is a coupled-resonator network with no additional elements such as power dividers, composed exclusively of si...
详细信息
ISBN:
(纸本)9781479938704
A compact diplexer fabricated on substrate-integrated waveguide (SIW) technology is presented. The diplexer is a coupled-resonator network with no additional elements such as power dividers, composed exclusively of six SIW resonators coupled through inductive irises, with one transmission zero at each channel that improves the selectivity. The close physical proximity of the resonators and the leakage associated to via-hole walls characteristic of the SIW technology are the main challenge of this design. However, this effect can be taken into account, as proved by a fabricated prototype with pass bands around 8 GHz.
Focal plane arrays (FPAs) find applications in dish-antenna based receivers having multiple wideband beams. Such beams have a plurality of off-axes directions from the main beam. Maximally-decimated multirate 3-D digi...
详细信息
ISBN:
(纸本)9781479920365
Focal plane arrays (FPAs) find applications in dish-antenna based receivers having multiple wideband beams. Such beams have a plurality of off-axes directions from the main beam. Maximally-decimated multirate 3-D digital FIR frustum filterbank can be employed in an FPA receiver to reduce directional jamming/clutter and other interference, noise, mutual electromagnetic coupling effects, and out-of-system noise/interference from warm bodies close to the antenna. A 3-D frustum filter requires high arithmetic complexity in the DSP hardware and sophisticated wide-band RF-front ends featuring high dynamic range. The temporal channelization of FPA elements using microwave triplexers achieves frequency channelization in continuous-time. Channelization is moved before the LNA stage, thereby allowing considerable improvement in dynamic range requirements for the RF-front end. The channelization reduces LNA bandwidth for each channel. The signals within the microwave channelizer subbands can either be down-converted through mixers, or directly sub-sampled by means of bandpass sampling A/D converters. Integration of multi-dimensional signal processing theory - which, traditionally is built around DSP methods - with microwave engineering, yields mixed-mode microwave-digital realizations of 3-D frustum filters having applications in FPA receivers. Simulation results of 3-D microwave-digital FIR mixed-domain three-channel frustum filters show SNR improvement of 21.5 dB and SIR improvement for 24.2 dB, for wideband applications in the range 1-4 GHz using an FPA having 32 × 32 elements with a dish of aperture 5m and focal length 7m.
Many real classification tasks are oriented to sequence (neighbor) labeling, that is, assigning a label to every sample of a signal while taking into account the sequentiality (or neighborhood) of the samples. This is...
详细信息
ISBN:
(纸本)9781479957521
Many real classification tasks are oriented to sequence (neighbor) labeling, that is, assigning a label to every sample of a signal while taking into account the sequentiality (or neighborhood) of the samples. This is normally approached by first filtering the data and then performing classification. In consequence, both processes are optimized separately, with no guarantee of global optimality. In this work we utilize Bayesian modeling and inference to jointly learn a classifier and estimate an optimal filterbank. Variational Bayesian inference is used to approximate the posterior distributions of all unknowns, resulting in an iterative procedure to estimate the classifier parameters and the filterbank coefficients. In the experimental section we show, using synthetic and real data, that the proposed method compares favorably with other classification/filtering approaches, without the need of parameter tuning.
The main contributions of this paper are an automated approach for applying the ABCDE rules in a digital dermoscopy platform with fixed settings and a new registration method specially designed for aligning and compar...
详细信息
The main contributions of this paper are an automated approach for applying the ABCDE rules in a digital dermoscopy platform with fixed settings and a new registration method specially designed for aligning and comparing follow-up digital dermoscopy images in order to evaluate the evolution over time parameter E. Experimental evaluations of the registration method are reported for image pairs acquired during follow-up examinations.
This paper aims to present an approach in order to share knowledge between the business actors of the extended enterprise. This approach is based on a semantic wiki that access the organizational memory of our knowled...
详细信息
This paper aims to present an approach in order to share knowledge between the business actors of the extended enterprise. This approach is based on a semantic wiki that access the organizational memory of our knowledge management system OCEAN. This principle allows us, firstly to share knowledge in order to facilitate the individual work, and secondly to reuse capitalized knowledge automatically.
IEEE 802.11x CSMA/CA DCF MAC protocol supports that wireless nodes have statistically impartial probabilities of wireless channel access through fair competition. However, there is greedy node problem that maliciously...
详细信息
IEEE 802.11x CSMA/CA DCF MAC protocol supports that wireless nodes have statistically impartial probabilities of wireless channel access through fair competition. However, there is greedy node problem that maliciously increasing the transmission rates of mobile nodes altering their MAC operation disturbs fair transmissions between wireless nodes. This paper addresses how to find misbehavior greedy nodes. Previous works inspect the operation of DCF MAC protocol by the MAC frame to detect greedy nodes. In this paper, a greedy node detection algorithm using Kolmogorov-Smirnov test is proposed. The algorithm classifies wireless nodes with similar probability distributions of transmission intervals and draws a comparison between groups to find a group of greedy nodes. This paper evaluates the proposed algorithm through simulation and the simulation results shows that the algorithm can accurately detect greedy nodes in the congestion condition.
暂无评论