Showing papers by "Orange S.A. published in 2021"

Predictor Antenna: A Technique to Boost the Performance of Moving Relays

Abstract: In future wireless systems, a large number of users may access the networks via moving relays (MRs) installed on top of vehicles. One of the main challenges of MRs is rapid channel variation, which may make channel estimation and its following procedures difficult. To address these issues, various schemes are designed, among which the predictor antenna (PA) is a candidate. The PA setup refers to a system with two (sets of) antennas on top of a vehicle, where the PA(s) positioned in front of the vehicle is (are) utilized to predict the channel state information required for data transmission to the receive antennas aligned behind the PA. In this article, we introduce the concept and the potential of PA systems. Moreover, summarizing the field trials for PAs and the 3GPP attempts on (moving) relays, we compare the performance of different PA and non-PA methods for vehicle communications in both urban and rural areas with the PA setup backhauled through terrestrial or satellite technology, respectively. As we show, with typical parameter settings and vehicle speeds, a single-antenna PA-assisted setup can boost the backhaul throughput of MRs, compared to state-of-the-art open-loop schemes, by up to 50 percent.

Cross-Collaboration Processes based on Blockchain and IoT: a survey.

Abstract: Cross-collaboration processes are decentralized by nature and their centralized monitoring can trigger mistrust. Nevertheless, a decentralized monitoring facility such as a blockchain-based and Internet-of-Things-aware (IoT-aware) business process management system can reduce this pitfall. However, concerns related to usability, privacy, and performance, hamper the wide adoption of these systems. To better understand the challenges at stake, this paper reviews the use of blockchain and IoT devices in cross-collaboration processes. This survey sheds some light on standard uses such as model engineering or permissioned blockchains which help adopt cross-collaboration business process management systems. Moreover, with respect to process design, two schools of thought coexist, addressing both constrained and loosely processes. Furthermore, a focus on data-centric processes appears to get some momentum, as many industries go digital. Finally, this survey underlines the need to orient future research towards a more flexible, scalable, and data-aware blockchain-based business process management system.

Neural-Driven Search-Based Paraphrase Generation

Abstract: We study a search-based paraphrase generation scheme where candidate paraphrases are generated by iterated transformations from the original sentence and evaluated in terms of syntax quality, semantic distance, and lexical distance. The semantic distance is derived from BERT, and the lexical quality is based on GPT2 perplexity. To solve this multi-objective search problem, we propose two algorithms: Monte-Carlo Tree Search For Paraphrase Generation (MCPG) and Pareto Tree Search (PTS). We provide an extensive set of experiments on 5 datasets with a rigorous reproduction and validation for several state-of-the-art paraphrase generation algorithms. These experiments show that, although being non explicitly supervised, our algorithms perform well against these baselines.