The underwater world offers an immense space with an enormous potential, especially for a country as is Portugal that has one of the biggest marine zones in Europe (Exclusive Economic Zone, the 3º largest in Europe and 11º largest in the world). A fundamental part for withdrawing the full potential of this natural resource is to solve the problem of high-bandwidth underwater communications.

This project aims to develop signal processing schemes that will enable for high bandwidth underwater acoustic (UWA) communications using aggregates of sources and hydrophones (Multiple Input, Multiple Output - MIMO) between mobile emitters and receivers.
In radio communications, MIMO systems allows for considerable gains in the transmitted throughput and reliability, what has recently made several research groups to consider them for UWA. However, the propagation channel in UAC has characteristics substantially different than that of radio channels; since it has a high variability in both time and frequency (i.e., the channel is doubly selective). Besides that, the noise presents impulsive characteristics, for which the normal Gaussian model is inadequate. So being, the use of techniques for the normal radio channel without specific modifications would give rise to high performance degradation in UAC.
This being, the main purpose of this project is to develop and assess the performance of transmission techniques for MIMO-UWA systems. The project will begin by selecting relatively simple channel models to account for the doubly selective specificities of the doubly selective UWA channel model. Adequate models to characterize the impulsive noise specifications of the channel will also be selected, in order to permit adequate analytical processing.
Afterwards, special transmission techniques specially designed for MIMO-UWA will be developed. These techniques will be based on MIMO-OFDM (Orthogonal Frequency Division Multiplex) and MIMO SC-FDE (Single Carrier- Frequency Domain Equalization) [BFDT10], combined with IB-DFE (Iterative Block-Decision Feedback Equalizer) [Benvenuto_Dinis_ProceedingsIEEE2010].

The channel will be estimated by taking advantage of Delay-Doppler Spread Functions (DDSF) [ZGB10_2]. We will exploit the sparsity of DDSFs for the envisaged patterns of motion and their spatial correlation to derive fast MIMO estimation methods using basis pursuit, as well as techniques to compensate for deviations of the carrier’s frequency (CFO – Carrier Frequency Offset) [PDN10].
In the case of OFDM modulation, techniques to reduce the PAPR(Peak-to-Average Power Ratio) will be employed based on non-linear devices [DG04, DSA09], so that a greater amplification efficiency can be attained, as well as adequate receivers for non-linear systems [Guerreiro2012,Guerreiro2013].
The developed MIMO systems will be validated with realistic Matlab simulations, a program yet to be developed and adapted with blocks already implemented by the different persons of the research team. This program will serve as a demonstrator/ prototype for the project.

Não foram encontrados registos.

Não foram encontrados registos.

Não foram encontrados registos.

Não foram encontrados registos.

Não foram encontrados registos.

Não foram encontrados registos.

Não foram encontrados registos.