MACC · Multi-application advanced channel coding

Final Activity Report Summary - MACC (Multi-application advanced channel coding)

Communication, i.e. the exchange of information from one or several senders to one or several receivers, is at the basis of all modern cultures. It now seems impossible to conceive our lives without mobile phones, satellite links, or computers, to make a few examples. In all these cases, there is some probability that the received message will not be identical to the transmitted one: errors can occur during the communication process. The following question naturally arises: how can we achieve reliable communication over an imperfect communication channel? Information theory and coding theory have provided the answer to this question. The principle, known as error correction coding, is to add to the user data some redundancy in a smart way so that this redundancy can be used in the receiver side to correct errors introduced by the channel. Today, error correction codes are an essential element of any communications system. The main goal of this project was the analysis and design of advanced error correction coding techniques targeting several key applications that will drive the future of communications: satellite communications, wireless communications and packet-wise communications (Internet). A particular emphasis was directed towards turbo-like codes, also known as concatenated codes, considered among the most powerful error correction schemes of today. The proposed research fell into the category of fundamental research but it was strongly motivated by practical applications. The project made several significant contributions to the theory of concatenated codes. We gave a theoretical analysis of the finite-length and the asymptotic behaviour of concatenated codes for the Gaussian channel (characteristic of e.g. satellite communications) and the Binary Erasure Channel (BEC, e.g. Internet). We also provided a unified framework for the analysis of concatenated codes, which generalizes and includes all existing concatenations. Our theoretical results contributed to a better understanding of concatenated codes, determining their potential and establishing their limits. Furthermore, our theoretical investigations proved that certain concatenated codes are asymptotically good (a property which is desirable) for both the Gaussian and the BEC channels, and established the basis for the design of powerful low complexity concatenated coding schemes with application to a number of crucial applications. Based on our theoretical analysis, we proposed novel low-complexity, versatile and powerful code structures for wireless and satellite communications, in the framework of a European Space Agency (ESA) project. As a significant achievement, one of the proposed structures was adopted for the implementation of a satellite broadband system for triple play services (high-speed internet, television and telephony). Moreover, the proposed structure is being currently commercialised and it will also be proposed for the incoming satellite standards. Another code structure will be considered (supported by a European telecom company) for the future standard for mobile TV 'Digital Video Broadcasting-Newt Generation Hand Held' (DVB-NGH). The scientific relevance of the project is demonstrated from the numerous publications that it generated. 6 journal papers and 10 conference papers were published (or were accepted for publication) in the most prestigious journals/conferences of the research field during the duration of the project. Furthermore, 4 journal papers and 2 conference papers were submitted, and 4 journal papers are now in preparation. During this project, research collaborations were established with several prestigious research labs in Europe, the US and Australia, as well as with European industrial partners. All of these collaborations are still on-going and generated new lines of research. From this point of view, besides its theoretical contributions, this project had also a direct impact on commercial business and on industrial products and standards. As an example, two projects with European companies are now on-going for the exploitation and extension of the results obtained in this project.

Data: CORDIS, © European Union