International audience

In the midst of crisis of loss of trust in the world, the Blockchain technology has emerged as a suitable solution. It is considered as a revolutionary technology that is able to solve complex problems in different fields involving trust and disintermediation. In this article, we are interested in finding optimized solutions for parking spaces' target. Hence, we present an intelligent parking system managed by a decentralized system based on the Blockchain, which brings major changes to facilitate the reservation of vehicles.

Patient medical data is the key data of an e-health application. Medical data frequently play an important role in disease analysis and are implicated in motivating values for educating diagnostic methods and modifying diagnostic results. Confidentiality of patient health documents is more important, and therefore researchers have considered many security measures, including access control, confidentiality, and integrity of these documents. To solve the security problem, this paper presents an advanced solution based on converting medical data into compact size, in the form of QR code before being deployed on our Blockchain. In this way, performing complex operations will be safely performed off-chain, such as collecting and preprocessing patient data, while submission of this data is done on an hourly basis. Chain. We recommend a data flow architecture that combines the Internet of Things (IoT) with blockchain (especially the Ethereum architecture), using smart contracts to access and store data while keeping Strong against many attacks. Our proposed system is effective in sharing and managing patient electronic health data. The tests involved implementing Ethereum chain configurations on PCs and two embedded IoT platforms, namely Raspberry Pi and PYNQ cards. Performance is evaluated in terms of security, execution time, memory, power and CPU consumption of various patient data scenarios.

Votingis a basic element of running a country. Voting will continue to take place by physically entering the voting booth. No security is guaranteed for this operation, and several cases of tampering have been noted. In order to eliminate this type of problem, the paper proposes an online voting process with blockchain technology. With encryption and hashing, the security of each vote is ensured. The votes will be stored as transactions. A peer-to-peer network is leveraged to share this distributed ledger with voting transactions. The application is designed to hide the complexities of the architecture from the user. With the QR code, each student is uniquely identified. This ensures that each voter has only one chance to vote. With the public and private key, each node will have the ability to securely encrypt, hash, and add transactions to the blockchain. Votes cannot be traced back to the voters. This paper creates a peer-to-peer network with at least three peers. This paper plans to increase voter turnout through online voting. The scalability of blockchain applications depends on the secondary storage limits of peers.

In the past few years, an increasing number of machine-learning and deep learning structures, such as Convolutional Neural Networks (CNNs), have been applied to solving a wide range of real-life problems. However, these architectures are vulnerable to adversarial attacks: inputs crafted carefully to force the system output to a wrong label. Since machine-learning is being deployed in safety-critical and security-sensitive domains, such attacks may have catastrophic security and safety consequences. In this paper, we propose for the first time to use hardware-supported approximate computing to improve the robustness of machine learning classifiers. We show that our approximate computing implementation achieves robustness across a wide range of attack scenarios. Specifically, we show that successful adversarial attacks against the exact classifier have poor transferability to the approximate implementation. The transferability is even poorer for the black-box attack scenarios, where adversarial attacks are generated using a proxy model. Surprisingly, the robustness advantages also apply to white-box attacks where the attacker has unrestricted access to the approximate classifier implementation: in this case, we show that substantially higher levels of adversarial noise are needed to produce adversarial examples. Furthermore, our approximate computing model maintains the same level in terms of classification accuracy, does not require retraining, and reduces resource utilization and energy consumption of the CNN. We conducted extensive experiments on a set of strong adversarial attacks; We empirically show that the proposed implementation increases the robustness of a LeNet-5 and an Alexnet CNNs by up to 99% and 87%, respectively for strong transferability-based attacks along with up to 50% saving in energy consumption due to the simpler nature of the approximate logic. We also show that a white-box attack requires a remarkably higher noise budget to fool the approximate classifier, causing an average of 4 dB degradation of the PSNR of the input image relative to the images that succeed in fooling the exact classifier.

The adoption of Internet of Things (IoT) technology across many applications, such as autonomous systems, communication, and healthcare, is driving the market’s growth at a positive rate. The emergence of advanced data analytics techniques such as blockchain for connected IoT devices has the potential to reduce the cost and increase in cloud platform adoption. Blockchain is a key technology for real-time IoT applications providing trust in distributed robotic systems running on embedded hardware without the need for certification authorities. There are many challenges in blockchain IoT applications such as the power consumption and the execution time. These specific constraints have to be carefully considered besides other constraints such as number of nodes and data security. In this paper, a novel approach is discussed based on hybrid HW/SW architecture and designed for Proof of Work (PoW) consensus which is the most used consensus mechanism in blockchain. The proposed architecture is validated using the Ethereum blockchain with the Keccak 256 and the field-programmable gate array (FPGA) ZedBoard development kit. This implementation shows improvement in execution time of 338% and minimizing power consumption of 255% compared to the use of Nvidia Maxwell GPUs.

Floating point arithmetic is one of the most commonly used units in nowadays computing systems and is deployed for a wide range of domains and applications. While floating point operators offer high precision calculations, a plethora of applications such as multimedia processing and machine learning tolerate errors and computation imprecision. In a context of limited power budget embedded systems, saving resources and energy with an acceptable precision loss is a challenging design task. Approximate computing is an emerging systems design paradigm that offers promising balance between accuracy on the one hand and power consumption and resource utilization on the other hand. While state of the art approximate techniques offer a wide design space at the operator level, few are the works that consider exploring different techniques to build a heterogeneous comprehensive approximate design. In this paper, we propose HEAP: a heterogeneous approximate floating point multiplier. Based on a design space exploration process, we present an approximation at the transistor level that reduces energy consumption of up to 68%. Experimental study on a set of machine learning applications shows promising results with comparable accuracy to exact multiplier based systems.

The emergency of embedded systems puts new challenges for the design of different system in many fields. One of the embedded application's fields is the medical one. The major difficulty is the embedded system's reduced energy and computational resources that must be carefully used to execute complex application often in unpredictable environments. In this paper, the used application is the detection of evoked potential with variable latency and multiple trials using consensus matching pursuit. Fitting to the noisy Evoked Potential (EP) signal persistent in all response, we use the Consensus version of the matching pursuit algorithm (CMP). EP is a resulted wave from a stimulus. The EP can be explained with a good quality of energy ratio factor (QR). If we use a noisy EP, we cannot reconstruct the original data because of the random atoms of CMP dictionary. We select the significant atoms to rebuild and EP signals. This application is embedded on a Xilinx ML 507. We used an adaptive architecture based on dynamically partial reconfiguration.

The study of the EEG inverse problem consists in active brain's source reconstruction. In this paper we review the localization error from this reconstruction using a new configuration of electrodes on the scalp. The suggested new configuration consists on studying each lobe separately. The objective is to minimize the localization error with a minimum number of electrodes. To validate this study, we use the Shrinking sLORETA-FOCUSS method as a solution of the inverse problem. The obtained results show very interesting values using a minimum number of electrodes placed on the scalp surface. In this paper we will embed the proposed approach on an FPGA platform using adaptation techniques. Due to different electrodes we used the dynamic partial reconfiguration technic. The different electrodes data are treated by IPs based on VHDL accelerators. The proposed architecture will be embedded on Xilinx Virtex 5 ML 507 platform.