Section: Overall Objectives

Overall Objectives

Abstract: The Cairn project-team researches new architectures, algorithms and design methods for flexible and energy efficiency domain-specific system-on-chip (SoC). As performance and energy-efficiency requirements of SoCs are continuously increasing, they become difficult to fulfill using only programmable processors solutions. To address this issue, we promote/advocate the use of reconfigurable hardware, i.e. hardware structures whose organization may change before or even during execution. Such reconfigurable SoCs offer high performance at a low energy cost, while preserving a high-level of flexibility. The group studies these SoCs from three angles: (i) The invention and design of new reconfigurable platforms with an emphasis on flexible arithmetic operator design, dynamic reconfiguration management and low-power consumption. (ii) The development of their corresponding design flows (compilation and synthesis tools) to enable their automatic design from high-level specifications. (iii) The interaction between algorithms and architectures especially for our main application domains (wireless communications, wireless sensor networks and digital security).

 

The scientific goal of the Cairn group is to research new hardware architectures of Reconfigurable System-on-Chips (RSoC) along with their associated design flows. RSoCs chips integrate reconfigurable blocks whose hardware structure may be adjusted before or even during a program execution. They originate from the possibilities opened up by Field Programmable Gate Arrays (FPGA) technology and by reconfigurable processors [89] , [99] . Recent evolutions in technology and modern hardware systems confirm that reconfigurable systems are increasingly used in recent applications or embedded into more general system-on-chip (SoC) [104] . This architectural model has received a lot of attention in academia over the last decade [94] , and is now considered for industrial use. One reason is the rapidly changing standards in communications and information security that require frequent device modifications. In many cases, software updates are not sufficient to keep devices on the market, while hardware redesigns remain too expensive. The need to continuously adapt the system to changing environments (e.g. cognitive radio) is another incentive to use dynamic reconfiguration at runtime. Last, with technologies at 65 nm and below, manufacturing problems strongly influence electrical parameters of transistors, and transient errors caused by particles or radiations will also appear more and more often during execution: error detection and correction mechanisms or autonomic self-control can benefit from reconfiguration capabilities.

Standard processors or system-on-chips enable flexible software on fixed hardware, whereas reconfigurable platforms make possible flexible software on flexible hardware.

As chip density increases [116] , power efficiency has become “the Grail” of all chip architects, be they designing circuits for portable devices or for high-performance general-purpose processors. Indeed, power (or energy) constraints are now as equally important as performance constraints. Moreover, this power issue can often only be addressed through the use of a complete application-specific architecture, or by incorporating some application-specific components within a programmable SoC. Designers hence face a very difficult choice between the flexibility and short design time of programmable architectures and the power efficiency of specialized architecture. In this context, reconfigurable architectures are acknowledged for providing the best trade-off between power, performance, cost and flexibility. This efficiency stems from the fact that their hardware structure can be adapted to the application requirements [115] , [99] .

However, designing reconfigurable systems poses several challenges : first, the definition of the architecture structure itself along with its dynamic reconfiguration capabilities, and then, its corresponding compilation/synthesis tools. The scientific goal of Cairn is therefore to leverage the background and past experience of its members to tackle these challenges. We therefore propose to approach energy efficient reconfigurable architectures from three angles: (i) the invention of new reconfigurable platforms, (ii) the development of their corresponding design and compilation tools, and (iii) the exploration of the interaction between algorithms and architectures.

Wireless Communication is our privileged application domain, and it builds on our experience in 3G. Our research includes the prototyping of (subsets of) such applications on reconfigurable and programmable platforms. For this application domain, the high computational complexity of the Next-Generation (4G) Wireless Communication Systems calls for the design of highly specialized high-performance architectures. In Wireless Sensor Networks (WSN), where each wireless node is expected to operate without battery replacement for significant periods of time, energy consumption is the most important constraint. In this context, our research focuses on energy-efficient architectures and wireless cooperative techniques for WSN and wireless transmission in Intelligent Transportation Systems (ITS). Other important fields such as automotive, digital security and multimedia processing are also considered.

Members of the Cairn team have collaborations with large companies like STMicroelectronics (Grenoble), Technicolor (Rennes), Thales (Paris), Alcatel (Lannion), France-Telecom Orange Labs (Lannion), Atmel (Nantes), Xilinx (USA), SME like Geensys (Nantes), R-interface (Marseille), TeamCast/Ditocom (Rennes), Sensaris (Grenoble), Envivio (Rennes), InPixal (Rennes), Sestream (Paris), Ekinops (Lannion) and Institute like DGA (Rennes), CEA (Saclay, Grenoble). They are involved in several national or international funded projects (FP7 Alma, FP7 Flextiles, Nano2012 S2S4HLS and RECMOTIF projects, ANR funded Pavois, Ardyt, Defis, Faon, Compa, Greco, Ocelot and "Images&Networks Competitiveness Cluster" funded 100Gflex).