The Emeraude team of the CITI laboratory are organizing the 2025 “Journées de l’Informatique Musicale” (JIM) and Linux Audio Conference (LAC) jointly on June 23-28, 2025 in Lyon : https://jimlac25.inria.fr/
The calls for papers for these 2 events are published here :
– https://jimlac25.inria.fr/lac/#call-for-papers-presentations-demos-workshops
Josiane Kouam will give a talk in the meeting room at the 4th floor of the Inria building.
Title
Mobile Human Behavior: Availability, Leveraging, and Addressing Privacy Attacks
Abstract
In an era of ubiquitous mobile connectivity, the security and privacy of users have become paramount concerns. My research delves into the intricate balance between harnessing mobile human behavior data for network optimization and security while safeguarding individual privacy from emerging threats. This talk will cover three key facets of my work in mobile security and privacy, offering insights into the opportunities and challenges presented by the pervasive nature of mobile technologies.
First, I will overview the importance and challenge of ensuring privacy-preserving access to such datasets. Next, I will discuss how behavioral data can be leveraged to enhance network security, including the development of a game-theoretic approach to mitigate international bypass frauds in cellular networks. Finally, I will address privacy risks linked to mobile behavioral data, focusing on vulnerabilities in permission-less sensors like accelerometers when combined with publicly available distinctive network properties.
This discussion will provide insights into balancing the benefits and risks of mobile data in an increasingly connected world.
Biography
Anne Josiane Kouam is a post-doctoral researcher at TU Berlin and member of the Machine Learning and Security team. She obtained her PhD from INRIA Saclay and Ecole Polytechnique, France, in May 2023. With a focus on privacy and security in Mobile and Cellular networks, her research investigates the evolving landscape of threats, particularly those arising from the intersection of Machine Learning and network security.
The RoboCup, an international robotics competition, will be held from July 17 to 21 in the Netherlands. The LyonTech team (CPE, INSA Lyon, Inria, PaloIT), some members of which are part of the Chroma project team, will participate in the @Home category (in the Open Platform league). Their goal: to design a companion robot capable of assisting humans in the daily challenges of the home. It is one of the world’s most important technological events for robotics research and education.
Good luck to the team!
More information : https://robocup-lyontech.github.io/opl/
The defense will take place on Tuesday 9th June (morning) in the Heidi Lamarr building (Amphi Chappe), Insa-Lyon, Villeurbanne.
Exact and anytime heuristic search for the Time Dependent Traveling Salesman Problem with Time Windows
The Time Dependent (TD) Traveling Salesman Problem (TSP) is a generalization of the TSP which allows one to take traffic conditions into account when planning tours in an urban context: travel times between points to visit depend on departure times instead of being constant. The TD-TSPTW further generalizes this problem by adding Time Window constraints, i.e., constraints on visit times. Existing exact approaches such as Integer Linear Programming and Dynamic Programming usually do not scale well; heuristic approaches scale better but provide no guarantees on solution quality.
In this thesis, we introduce a new exact and anytime solving approach for the TD-TSPTW which aims at quickly providing approximate solutions and gradually improving them until proving optimality. We first show how to reduce the TD-TSPTW to the search for a best path in a state-transition graph. We provide an overview of existing search algorithms, with a focus on exact and anytime extensions of A*, and introduce a new one by hybridizing two of them. We show how to combine these exact and anytime search algorithms with local search – in order to faster find solutions of higher quality – and with bounding and time window constraint propagation – in order to filter the search space. Finally, we provide extensive experimental results to (i) validate our main design choices, (ii) compare our approach to state-of-the-art solving approaches on various TD benchmarks with different degrees of realism and different temporal granularities and (iii) compare TD solving approaches to recent TSPTW solvers on constant benchmarks. These experimental results show us that our approach offers a good compromise between the time needed to find good solutions and the time needed to find optimal solutions and prove their optimality for both TD and constant TSPTW instances.
The defense will take place on Friday 24th May at 10AM in the Heidi Lamarr building, Insa-Lyon, Villeurbanne.
Contributions to Wireless Sensor Networks for Air Quality Monitoring
In this talk, I will present a summary of my research, which revolves around the design and evaluation of novel solutions for Wireless Sensor Networks to efficiently monitor physical phenomena. I have addressed several scientific and technical issues by adopting a global methodology combining theoretical solutions and experimental developments. Although our solutions can be easily adapted to different applications, the focus was on air quality monitoring, a major societal challenge where new low-cost sensing technologies offer a significant advantage over traditional solutions.
This talk focuses on our main contributions in this area of low-cost sensor networks for environmental monitoring. It is structured around three axes: i) static sensor networks for air quality monitoring in cities and on industrial sites; ii) participatory sensing of air quality and Urban Heat Islands; and iii) UAV fleets for monitoring highly dynamic phenomena. The common thread running through all our solutions is that they take into account both the physical domain knowledge and the characteristics of low-cost sensors such as the limited and heterogeneous measurement accuracy. I will conclude this talk by discussing some personal feedback and setting out some future perspectives.
The defense will take place on Tuesday 22th May at 10 AM in the East amphitheater of the humanities building, Insa-Lyon, Villeurbanne.
Programming language abstractions for the Internet of Things era
The Programmable Audio Workshop (PAW) is a yearly one day FREE event gathering members of the programmable audio community around scientific talks and hands-on workshops. The 2023 edition of PAW is hosted by the INRIA/INSA/GRAME-CNCM Emeraude Team at the Marie Curie Library of INSA Lyon (France) on December 2nd, 2023. The theme of this year’s PAW is “Artificial Intelligence and Audio Programming Languages” with a strong focus on computer music languages (i.e., Faust, ChucK, and PureData). The main aim of PAW-23 is to give an overview of the various ways artificial intelligence is used and approached in the context of Domain Specific Languages (DSL) for real-time audio Digital Signal Processing (DSP).
More information and registration here
The defense will take place on Tuesday 19th December at 2 PM in the Heidi Lamarr building (Amphi Chappe), Insa-Lyon, Villeurbanne.
Navigation Among Movable Obstacles (NAMO) Extended to Social and Multi-Robot Constraints
As robots become ever more commonplace in human environments, taking care of ever more tasks such as cleaning, security or food service, their current limitations only become more apparent. One such limitation is of their navigation capability in the presence of obstacles: they always avoid them, and freeze in place when avoidance is impossible.
This is what brought about the creation of Navigation Among Movable Obstacles (NAMO) algorithms, expected to allow robots to manipulate obstacles as to facilitate their own movement. However, these algorithms were designed under the hypothesis of a single robot per environment, biasing NAMO algorithms into only optimizing the single robot’s displacement cost – without any consideration for humans or other robots. While it is desirable to endow robots with the human capability of moving obstacles, they must however do so while respecting social norms and rules of humans.
We have thus extended the NAMO problem as to take into account these new social and multi-robots aspects. By relying on the concept of affordance spaces, we have developed a social occupation cost model allowing the evaluation of the impact of moved objects on the environment’s navigability. We implemented (and improved) reference NAMO algorithms, in our open source simulation tool, and modified them so that they may plan compromises between robot displacement cost and social occupation cost of moved obstacles – resulting in improved navigability. We also developed an implicit coordination strategy allowing the concurrent execution of these same algorithms by multiple robots as is, without any explicit communication requirements, while preserving the no-collision guarantee; verifying the relevance of our social occupation cost model in the actual presence of other robots. As such, this work constitutes the first steps towards a Social and Multi-Robot NAMO.
The defense will take place on Friday 29th December at 2 PM in the Heidi Lamarr building (Amphi Chappe), Insa-Lyon, Villeurbanne.
Human and Network Mobility Management using Mobile Phone Data
Over the past decade, the increasing use of smartphones has led to a significant rise in the volume of data exchanged through mobile networks of telecommunications operators. Each new generation of mobile network generates more data than its predecessor. By 2027, it is estimated that 289 EB of data will be exchanged per month, with 62% originating from the 5G mobile network. This vast availability of data has opened up new research perspectives, particularly in the study of mobility. Mobile data enables studies on larger populations and extended geographical areas.
In this thesis, we demonstrate that the events described in mobile data can also be found in other data sources. Through comparisons between mobile data and sensors detecting human presence, we observe a reasonable correlation. However, certain events, such as synchronization of peak presence or end-of-day activity, exhibit less similarity. We also utilize mobile data to examine the impact of the COVID-19 lockdowns imposed by the French government on land usage in Paris. Our findings indicate that the first lockdown had a profound impact on mobility patterns and land utilization, while the second and third lockdowns had a lesser impact. Lastly, we leverage this data for the reconfiguration of the mobile network in managing user micro-mobility, known as handover. The eNodeBs, which constitute the access network, can have different profiles and categories. By distinguishing between mobile and stationary users, we can optimize resource allocation through network reconfiguration. Dynamic network reconfiguration, employing various eNodeB profiles, also enables resource savings for mobile users.