A fusion method was recently proposed for ultrasound and magnetic resonance images for endometriosis diagnosis. This method combined the advantages of each modality, i.e., the good contrast and signal to noise ratio of the MR image and the good spatial resolution of the US image. The method was based on an inverse problem, performing a superresolution of the MR image and a denoising of the US image. A polynomial function was introduced to model the relationships between the gray levels of the MR and US images. This paper studies the potential interest of replacing this polynomial function by a non-parametric transformation built using the theory of reproducing kernel Hilbert spaces. Simulations conducted on a phantom and synthetic data allow the performance of the resulting fusion method to be appreciated.

In this work, we investigate the problem of neuralbased error correction decoding, and more specifically, the new so-called syndrome-based decoding technique introduced to tackle scalability in the training phase for larger code sizes. We improve on previous works in terms of allowing full decoding of the message rather than codewords, allowing thus the application to nonsystematic codes, and proving that the single-message training property is still viable. The suggested system is implemented and tested on polar codes of sizes (64,32) and (128,64), and a BCH of size (63,51), leading to a significant improvement in both Bit Error Rate (BER) and Frame Error Rate (FER), with gains between 0.3dB and 1dB for the implemented codes in the high Signal-to-Noise Ratio (SNR) regime.

The world’s forests have undergone substantial changes in the last decades. In the tropics, 17% of moist forests disappeared between 1990 and 2019, through deforestation and forest degradation [7]. These changes contribute greatly to biodiversity loss through habitat destruction, soil erosion, terrestrial water cycle disturbances, and anthropogenic CO2 emissions. Continuous monitoring of global deforestation is a fundamental tool to support preservation actions and to stop further destruction of vegetation. Several forest disturbance detection systems have already been developed, mainly based on space-borne optical remote sensing [4] which is severely limited by cloud coverage in the tropics. Contrarily to optical imagery, SAR products have the great potential of being insensitive to the presence of clouds. In recent years, several SAR-based systems have been developed and are now operational in different dense forest areas across the tropics [2], [3], [5], [6]. Despite the extensive coverage and temporal density of acquisitions, C-band SAR data like Sentinel-1 are not ideal for deforestation monitoring since the returned backscatter can be altered by variations in soil moisture and others. In this work, we investigate a new method to monitor forest loss in a near real-time manner exploiting the principle of Bayesian inference. In particular, forest loss is treated as a change-point detection problem within a univariate time series (i.e. Sentinel-1 single polarization), in which each new observation contributes to the probability of having or not deforestation in a Bayesian-like manner [1]. Detection delay and false alarm reduction have been investigated through the extension of the algorithm to the multivariate case of dual-polarization Sentinel-1 acquisitions. Given the synchronous nature of VV, VH acquisitions, such a modification allows an increase in the equivalent number of looks on a pixel on the ground, hence augmenting the level of confidence of an issued alert. A validation campaign has been conducted to assess the performance of the method. The test sites are located in French Guiana and Brazil where deforestation takes place constantly and near real-time monitoring is fundamental for law enforcement practices. Additionally, a comparison with a well-known deforestation monitoring technique, namely Maximum Likelihood Ratio Test, has been performed to further evaluate the proposed method. Conclusively, the potential of extending the current method to asynchronous data sources such as Sentinel-2 optical data is addressed.

Intrinsic lower bounds on the intrinsic mean square error are of major importance to characterize the best achievable estimation performance of any unbiased estimator on smooth manifold as Lie group (LG). When the parameter is described by a LG model and the observation noise is with unknown variance, it is also necessary to determine a bound both on the LG parameter and this variance. In this communication, we propose an intrinsic generic Fisher information matrix taking into account this problem. To achieve that, we derive an intrinsic Slepian-Bangs formula on the LG product of the unknown parameter of interest and the LG of positive scalar values in which the variance intrinsically lies. The proposed bound is validated on a Gaussian observation model for unknown parameter lying to SE (3) and variance noise on R+.

In this article, we first derive a general intrinsic Barankin bound (IBB) for unknown parameters lying on Lie groups (LGs), and its intrinsic McAulay-Seidman bound (IMSB) approximation. Second, the IMSB expression is used to revisit the intrinsic Cramér-Rao bound (ICRB) on LGs. Indeed, an analytic expression of the ICRB, which is a special IMSB case, is obtained from the latter. Finally, closed-form expressions for both IMSB and ICRB are obtained for Euclidean and LG observation models depending on parameters lying in SO(3) and SE(3). The validity of the these IMSB and ICRB expressions, with respect to the intrinsic mean square error, is shown via numerical simulations to support the discussion.

The study of ground reflections of Global Navigation Satellite System (GNSS) signals, as in GNSS Reflectometry (GNSS-R) can lead to the receiver height estimation. The latter is estimated by comparing the time of arrival difference between the direct and reflected signals, also called path separation. In ground-based scenarios, this path separation can be very small, inducing important interference between paths, which makes it difficult to correctly obtain altimetry products. The path separation estimation can be obtained by a brute force dual source maximum likelihood estimator (2S-MLE), but this solution has a large computational cost. On the other hand, the path separation is so small that a number of approximations can be done. In this study, a third order Taylor approximation of the dual source likelihood criterion is proposed to reduce its complexity. The proposed algorithm performance is compared to the non approximated 2S-MLE for the estimation of the path separation, and to a standard single source processing for the estimation of the direct signal time-delay. These results, along with the corresponding lower bounds, prove that the proposed approach may be of interest for two applications: ground-based GNSS-R altimetry (or radar with low elevation targets) and GNSS multipath mitigation.

This paper investigates a general framework for the registration of 3D magnetic resonance (MR) and 2D ultrasound (US) images. This framework is divided into a rigid slice-tovolume 3D-2D MR/US registration and a 2D-2D US/MRI fusion algorithm to generate an image having a better resolution than the MR image and a better contrast than the US image. The accuracy of the joint registration and fusion method is analyzed by means of quantitative and qualitative tests conducted on experimental phantom and realistic synthetic data generated from an in vivo MRI volume, with a specific attention to endometriosis treatment.

Interferences are an important threat for applications relying on Global Navigation Satellite Systems (GNSS). Interferences degrade GNSS performance, and can lead to denial of service. The most notable intentional interference family is characterized by its constant envelope, e.g. chirp and tone interferences. Due to its simple structure, the space to search the interference contribution yields to complex circles, allowing the introduction of some latent variables related to those circles. In order to mitigate the interference effect, we compute the maximum likelihood estimator of the parameters of interest (time delay and Doppler shift) in presence of those latent variables. Thus, we resort to the Expectation Maximization algorithm which has already been proved to be efficient in such cases. Experiments conducted on synthetic signals highlight the efficiency of the proposed algorithm.