Signal, Image and Telecom
The SITe research team includes 3 permanent members (2 HDRs, 1 PhD) and, currently, 4 PhD students and two research engineers. The research activities are related to three main axes:
- Signal and image processing, in various application fields, especially in medical imaging.
- Image and video acquisition, processing and encoding.
- Cellular and ad-hoc networks.
Signal and Image Processing
The team has developed a strong collaboration with the Clinical Investigation Centre of the Quinze-Vingts hospital, which is specialized in the study and the treatment of eye diseases. Two main research axes have been investigated: the study of blinding diseases, resulting in the progressive loss of the photoreceptors, the study of the retinal vascularization and of its modification in case of abnormalities. The team conducts also research activities in collaboration with the Institut Gustave Roussy (IGR), to characterize the quantification methods of tumor vasculature by dynamic contrast-enhanced ultrasonography (DCE-US).
The team is also involved in collaborative projects related to other application fields: sustainable agriculture with the analysis of multimedia data acquired through machine to machine (M2M) networks, digital reedition of documents acquired in image mode and optical music recognition (OMR).
In the last few years we have witnessed revolutionary advances in video and image technology. Storage capacity, as well as the communication links, has witnessed an enormous technological development. The work into video processing field tries thus to cope with nowadays exigencies: reduced acquisition cost of high dimensional video signals and video adaptation to heterogeneous bandwidth and display devices’ requirements.
In order to reduce the cost of video acquisition (e.g., multi-view or multi-view videos), compressed-sensed acquisition methods have been investigated; these schemes ensure reduction of both memory and computational power within the sensing/acquisition device, and considerably lower the transmission bandwidth.
As bandwidth requirements can have huge fluctuations from one user to another, efficient video compression is still a challenge. Therefore, an important part of the work on this direction has been focused on new techniques for efficient image/ video coding.
Cellular and Ad-Hoc Networks
With the massive growth of wireless connectivity demand, the radio frequency spectrum is becoming more and more a scarce resource. The spatial dimensions introduced by the Multiple Input Multiple Output (MIMO) systems are expected to become a must to achieve high individual data rates with high reliability on one hand, and to enable the multiplexing of multiple devices on the same radio resource on the other hand.
The LTE-advanced (LTE-A) standard supports Single-User SU-MIMO and Multi-User MU-MIMO transmission modes in the uplink as well as in the downlink.
In the context of LTE-A MIMO networks, we studied:
- the conception of radio resource allocation algorithms that distribute in an efficient way the time-frequency-spatial resources among the active user equipments.
- the dimensioning problem to statistically predict the number of time-frequency-spatial radio resource blocks, required to minimize the occurrence of the radio resource outage event.