Error-correction codes are used in the vast majority of communication systems because they allow a significant reduction of the transmitter power. Although many different coding schemes that can approach the Shannon limit on transmit power are known, the capability of practical systems is often limited by the energy consumption of the decoder. This constraint is being mentioned as a key performance metric in the ongoing 5G standardization process in order to deal with the great increase in the number of users and in the throughput while keeping a constant energy budget. In addition to the practical need for low-energy receivers, recent theoretical results have shown that in order to get arbitrarily close to the Shannon limit, the decoding circuit must consume an arbitrarily large amount of energy. This shows that optimizing the trade-off of coding gain versus decoding energy is fundamental in the channel coding problem. EF-FECtive aims to develop low-density parity-check (LDPC) codes and decoder circuits that together provide a 10x reduction in the energy consumption of decoders, while preserving equivalent communication performance. This will be achieved through contributions to both communication theory and VLSI system design, with the ultimate objective of demonstrating a decoder ASIC that can tolerate circuit faults while operating in the energy-efficient near-threshold regime. First, an approach to model the energy consumption of LDPC decoders combining an analysis of the decoding algorithm with simulations of circuit models will be developed. Based on these energy models, theoretical tools will be created to design LDPC codes and decoder circuits that minimize the decoding energy. Then, decoder implementations operated in the near-threshold regime will be studied. In this regime, extremely energy-efficient operation can be achieved, but maintaining fast processing performance requires the system to tolerate faulty computations. By developing accurate models of the effect of faults on the decoder and of its energy consumption, methods will be proposed to jointly optimize the LDPC code construction, the implementation parameters, and the amount of circuit faults allowed to drastically reduce the decoding energy. Finally, an ASIC prototype of the decoder will be designed, fabricated and tested to demonstrate the energy gains in practice.

Deep Neural Networks (DNN) have become the state of the art in the fields of Image Classification, Object Detection and Machine Translation, among others. However, this comes at the cost of increased complexity: more parameters, computations, energy consumption. DNN pruning is an effective way to reduce this complexity and provide high performance, low energy DNN implementations for embedded systems. ProPruNN proposes to precisely study the impact of structured pruning. This exploration will be done by co-designing hardware architectures capable of taking advantage of this pruning. The first objective is to clearly identify the real impact of structured pruning on the performance of networks implemented on FPGA. Indeed, in the literature, this impact is underestimated, because only a fraction of the prunable parameters are actually pruned. The second is to design predictive models of this impact, to incorporate it in the training of networks in order to optimize their throughput, latency, and energy efficiency during the training itself.

Nanoporous materials derived from clays feature textural properties and volumetric gas adsorption performances comparable to those of traditional industrial adsorbents for gas treatment process, such as bio-methane upgrading by CO2 capture. In this specific context, we have recently shown that smectites with residual moisture in the expanded interlayer space, exhibit not only higher gas adsorption capacity, but also the exceptional selectivity to CO2 over CH4. The combination of these specific properties with the availability of the natural precursors (smectites) makes clay adsorbents attractive for industrial applications. The present project aims at the design of novel high-surface-area clay-based porous materials (combing the strategies of smectite cation exchange, acidic activation etc.) supported by thorough characterization of their CO2/CH4 adsorption properties in a large range of P and T conditions and biogas upgrading process design relying on measured adsorption data.

To reduce the operational, automation, management and production cost and to simplify the production chain, the IoT technology is now considered in the industrial domain, also called the Industry 4.0, i.e., the 4th Industrial Revolution. The Industry 4.0 ambition is to make the factory more flexible and adaptable. The main goal is to replace the existing cables with a wireless medium, while guaranteeing network reliability above 99.999%. Furthermore, the Industry 4.0 requires robust communications, messages need to be sent securely and the communication framework must guarantee message delivery in a given delay with jitter close to 0. In addition, some environments may require a dense network, with thousand of nodes sending a large amount of messages. Scalability is thus another constraint imposed by the Industry 4.0. To reach this ambition, the wireless industrial networks must not only be reliable but also deterministic and predictable. A deterministic network guarantees that the transported information will be carry out in a pre-defined and in a tight window of time, whatever the link quality and the network congestion. Moreover, a periodic process will be repeated identically every time. So the most important characteristic of the network is to exhibit a jitter (different on the consecutive packet inter-arrival time) close to 0. Determinism is a required property in the power grid, to ensure that high tension lines breakers can be activated within milliseconds, in public transportation to make sure that automated vehicles are operated safely for their passengers, and in industrial automation for control loops. However, the current technologies deployed for the IoT are based on best-effort packet switched network. Data are encapsulated within packets that are subject to variable delays in the network, due to retransmissions and enqueuing in intermediate nodes. In 2016, the IEEE 802.15.4 standard was published to offer QoS for deterministic industrial-type applications. Time-Slotted Channel Hopping (TSCH) allows for competing the industrial standards. However, it does not avoid retransmissions when a data packet is lost, due to collision, or outage of one node. Moreover, the potential interferences, which lead to packet losses, with technologies operating on 2.4GHz, may decrease the reliability performance [4]. Since TSCH seems to be a good candidate solution, we propose to focus on implementing and extending the standard to bridge the gap between academia and Industry 4.0. Moreover, we propose to use distributed scheduling, multi-path routing and hybrid network to build the Industry 4.0 as presented in the following. In this project, firstly, we plan to define algorithm and protocols for nodes to set up their network: select the “good” channel (i.e., less interfering channels) and discover their neighbor and the routes within the network. Then we will investigate in depth both the distributed and centralized scheduling solutions of 6TiSCH to identify the pros and cons, as well as their appropriateness for ultra-deterministic industrial networks. We will propose new dynamic scheduling, tightly coupled with routing algorithms over multiple interfaces. Moreover, we plan to propose isolated (dedicated) tracks to provide flow isolation, and to make the transmissions reliable and independent: each application has dedicated (i.e., reserved) bandwidth for its packets transmissions. Associating innovative and original forwarding techniques (such as duplication, overhearing, opportunism and glossy networking) and a dynamic and efficient scheduling will allow us to have a jitter close to 0, thus to fully control delays in the network. In addition to those new algorithms and protocols, we will provide a security threat analysis and provide countermeasure within our protocol to secure the network. Both simulation and experimental set up will be used to validate our work toward the Industrial IoT.