This paper is presented in the context of the First Indoor Pathloss Radio Map Prediction Challenge at IEEE ICASSP 2025. We propose a deep learning approach using a customized ResUNet architecture with physics-informed features for predicting radio maps in indoor environments. Our architecture progressively adapts to handle increasing task complexity, incorporating dual-stream processing for frequency generalization and specialized antenna gain processing with dilated convolutions. Experimental results demonstrate effective generalization across unknown indoor scenes, frequencies, and antenna patterns while maintaining computational efficiency.

The massive machine-type communication will be at the core of ambient connectivity, requiring for energy-efficient systems. Earlier studies highlighted the efficiency of positioning approaches based on channel-state information (CSI) in different environments. Many works limited the solution assessment to a single room in a fully indoor testbed. This article extends the application of CSI for indoor–outdoor detection on an unprecedented large area and considers mMTC-oriented long-term evolution and fifth-generation Internet of Things in the sub-GHz frequency band. Hinged on a novel long-term evolution protocol dedicated for machine-type communications, the results focus on a unique packet exchange with a single access point to save battery life and simplify deployment. The study evaluates different input features and investigates the target positioning accuracy for multiple unsupervised and supervised dimensionality reduction methods. We present a new dimension reduction scheme consisting of an unsupervised funnel on top of a supervised dimension reduction approach. Results show that the introduced Funnel Fukunaga–Koontz transform outperforms other dimension reduction approaches, regardless of the input features and the number of locations.

A location-based service in the massive machine-type wireless communications (mMTC) must respect different requirements such as a minimal energy consumption at the target device or estimating the location in ambient connectivity. The solutions in mMTC must then consider localization approaches that provide target locations with few transmitted signals and with the support of only one anchor gateway. It is also major to use a relevant input data that manages the complex radio propagation mediums. In indoor environments, a solution builds on fingerprinting approach based on the channel state information (CSI) between the target device and a single anchor gateway. This paper presents a novel CSI fingerprinting localization method, E-Loc for mMTC dedicated to indoor systems in the Wi-Fi ambient connectivity context. E-Loc architecture is based on a convolutional neural network implementing inception models with an innovative design. CSI has been collected in a complex indoor environment, post-processed to handle the transmit power diversity, phase and timing offsets and fed to E-Loc. In various spatial distributions of training locations, E-Loc outperforms other tested solutions with a 99% confidence level for localization errors around 5 meters.

Vehicular Ad hoc NETworks (VANETs) are characterized by the rapidly changing topology and then a frequent network disruption. Hence, connectivity of moving vehicles presents an important challenge that critically influences the data transmission. Furthermore, data delivery ratio depends on routing protocols, applications type as well as environment characteristics. As a matter of fact, real experimentation in vehicular networks are costly and hard to deploy especially on large scale. Consequently, a vehicular mobility simulator is a good compromise to study how efficient are the data transmission mechanisms. In this paper, we comprehensively study the impact of the radio connectivity on data communication in vehicular networks. The analysis were realized based on a vehicular mobility simulator which runs a realistic scenario of mobility traffic in a real urban environment. A simple scenario of a safety application was implemented to examine the behavior of three well-known topology-based routing protocols. For the purpose of the analysis, we varied the simulation setup such as the density and the data traffic rate to determine the impact of the connectivity. The simulation results show that a realistic modelling of radio propagation has an important role in data transmission.

In this study, we proposed a deep learning model to infer the daily individual mobility pattern from static census data. Our work was inspired by Google Brain team work on machine learning system to automatically produce captions that accurately describe images using recurrent encoder-decoder model. They also use a convolutional neural network to exploit the strong spatially local correlation present in their structured data i.e., images. Unfortunately, survey data are generally heterogeneous with unknown local structure. Thus we have adapted their model using instead an appropriate mixed-variate version of restricted Boltzmann machine (MVRBM). This leads to estimation of daily regular mobility displacements in the form of variable length sequence given input individual attributes. The prime strength of our approach is that the resulting mobility flows inherit all the attributes contained in the input census which are typically missing in portable digital media data. Moreover, the model makes use of land use and point of interest data. optimized in this way, the model is scalable to apply to other places conditioned with censuses availability. Finally, it has been validated and showed its efficiency in a real context.

Many applications in Internet of Things (IoT) require an ubiquitous localization to provide their services. Whereas the global navigation satellite systems are mainly used in outdoor environment, multiple solutions based on mobile sensors or wireless communication infrastructures exist for indoor localization. One of them is the fingerprinting approach which consists in collecting the signals at known locations in a studied area and estimating the locations of new incoming signals thanks to the collected database. This approach interests many researches due to its connection with machine learning concepts. In this paper we propose to implement a deep learning architecture for a fingerprinting localization based on Wi-Fi channel frequency responses in IoT context. Our solution, DelFin reduces the median and 90-th percentile localization errors up to 50% and 47% respectively compared to other fingerprinting methods. DelFin has been tested with different spatial distributions of training locations in the studied area and still performed the best results.

With the rapid growth of cell phone networks during the last decades, call detail records (CDR) have been used as approximate indicators for large scale studies on human and urban mobility. Although coarse and limited, CDR are a real marker of human presence. In this paper, we use more than 800 million CDR to identify weekly patterns of human mobility through mobile phone data. Our methodology is based on the classification of individuals into six distinct presence profiles where we focus on the inherent temporal and geographical characteristics of each profile within a territory. Then, we use an event-based algorithm to cluster individuals and we identify 12 weekly patterns. We leverage these results to analyze population estimates adjustment processes and as a result, we propose new indicators to characterize the dynamics of a territory. Our model has been applied to real data coming from more than 1.6 million individuals and demonstrates its relevance. The product of our work can be used by local authorities for human mobility analysis and urban planning.

the human mobility is always organized nowadays in a multimodal context. However, the transport system has become more complex. For the sake of helping passengers, building Advanced Travelers Information Systems (ATIS) has therefore become a certain need. Since passengers tend to consider several other criteria than the travel time, an efficient routing system should incorporate a multi-objective analysis. Besides, the transport system may behave in an uncertain manner. Integrating uncertainty into routing algorithms may thus provide more robust itineraries. The main objective of this paper is to propose a Memetic Algorithm (MA) in which a Genetic Algorithm (GA) is combined with a Hill Climbing (HC) local search in order to solve the multicriteria shortest path problem in stochastic multimodal networks. As transport modes, railway, bus, tram and metro are considered. As optimization criteria, stochastic travel time, number of changes and walking time are taken into account. Experimental results have been assessed by solving real life itinerary problems defined on the transport network of the city of Paris and its suburbs. Results indicate that unlike classical deterministic algorithms and pure GA and HC, the proposed MA is efficient enough to be integrated within real world journey-planning systems.

Device-to-Device (D2D) is a promising technique for the future mobile networks and the resource allocation is one of the most crucial problems for its application. In order to efficiently allocate radio resource, the D2D network is organized with clusters such that the adjacent devices are assembled as an one-hop cluster. The cluster resource assignment problem is formulated as a dynamic graph coloring problem, and a graph coloring algorithm is designed from the graph structure point of view. This algorithm is able to allocate radio resource to clusters while they are dynamically generated and deleted. The numerical analysis results show that our algorithm has good performance in resource utilization, runtime and scalability.

Spatial and temporal decomposition of aggregated mobility flows is nowadays a commonly addressed issue, but a trip-purpose decomposition of mobility flows is a more challenging topic, which requires more sensitive analysis such as heterogeneous data fusion. In this paper, we study the relation between land use and mobility purposes. We propose a model that dynamically decomposes mobility flows into six mobility purposes. To this end, we use a national transportation database that surveyed more than 35,000 individuals and a national ground description database that identifies six distinct ground types. Based on these two types of data, we dynamically solve several overdetermined systems of linear equations from a training set and we infer the travel purposes. Our experimental results demonstrate that our model effectively predicts the purposes of mobility from the land use. Furthermore, our model shows great results compared with a reference supervised learning decomposition.

Multimedia Broadcast Multicast Service system supports efficient diffusion of multicast multimedia services in cellular networks. Our previous work shows that the radio resource management problem for MBMS can be modeled as a combinatorial optimization problem which tries to find optimal assignment of power and channel codes [1]. In this paper, we propose to solve such problem by using metaheuristic algorithm: Tabu Search (TS). In our work, we modify the general TS algorithm and map it onto our model. We also extend the classic TS procedure by proposing a tabu repair mechanism, which helps to explore new candidate solutions. The proposed algorithm is compared with two other metaheuristics: Greedy Local Search (GLS) and Simulated Annealing (SA). Simulations show that, within acceptable amount of time, TS can find better solution than GLS and SA.

Multimedia Broadcast Multicast Service (MBMS) system supports efficient diffusion of multicast multimedia services in cellular networks. Our previous work shows that the radio resource management (RRM) problem for MBMS can be modeled as an optimization problem which tries to find optimum assignment solution of power and channel codes in a given search space [1]. In this paper, based on the proposed model, we design a resource assignment approach by using the tabu search (TS) algorithm. Based on the model characteristics, we define three tabu memory structures and evaluate their search performance. We also extend the classic TS by proposing a tabu repair mechanism, which helps to avoid local optimum and improve the search efficiency. Simulation results show that the proposed TS algorithm outperforms the existing algorithms.

In this paper we present a mathematical modeling of Radio Resource Management (RRM) for multicast service diffusion based on Multimedia Broadcast Multicast Service (MBMS) standard. In this model, a flexible allocation approach named F2R2M is proposed, combining three candidate transport channels with scalable video transmission technology. The allocation procedure is implemented based on simulated annealing algorithm with a two- dimensional optimization objective and lexicographic order evaluation criteria. Experiments prove that, comparing with existing channel allocation approaches, F2R2M obtains allocation solution with equal QoS and lower transmission power consumption. Moreover, it reduces the possibility of achieving saturation of power or channelization codes when simulation scenarios have more users and heavy traffic load.

The VANET are mainly characterized by the high nodes velocity, high nodes density at road intersections and traffic jams and severe radio signals degradation caused by obstacles present in the external environment. This make the direct application of routing protocols defined for MANET not suitable for VANET. In this paper we consider a simulation environment representing a real city map with its terrain characteristics and urban infrastructures and analyze the most common protocols developed for MANETs. The objective is to identify the appropriate and inappropriate routing strategies for vehicular networks. So, we examine the behavior of each protocol varying the vehicular density and the data traffic rate to determine the mechanisms that enable them to have a good efficiency and those that cause their performance degradation. The results show that when the effect of obstacles on the radio signals is ignored the reactive protocols outperform the proactive protocols while when the impact of the obstacles is taken into account, the results are almost similar.

Nowadays, Indoor Positioning Systems capitalize on the existing wireless local area network infrastructure and are very popular and attractive. However, most systems only focus on the network deployment for positioning but overlook that the original purpose of these WLAN infrastructures is providing the required connectivity. In this paper, we propose an innovative approach where WLAN planning and positioning error reduction are modeled as an optimization problem and tackled together during the WLAN planning process. A Mono-objective algorithm called Variable Neighborhood Search (VNS) is implemented. The simulations results demonstrate that this approach is highly efficient in solving the indoor positioning optimization problem.

In recent years, the indoor positioning systems using the existing wireless local area network and Location fingerprinting schemes are the most popular system. The accuracy of the system is the most important indicator. In this paper we present experimental studies to emphasize on location error. Two experimentation stages are realized. The first one is based on selected reference points. The second one is based on precision indicators. The obtained results give more insights for environment parameters and their impact on location error. Finally, we propose an optimization algorithm to effectively increase the location accuracy.

Frequency assignment is one of the main issues in radio networks planning. The multiple interferences are seldom taken into account in literature. There is not a framework with their modeling. A hypergraph modeling of the network gives a more realistic representation of this phenomenon. We generalize theT-coloring problem for graphs to hypergraphs. We apply this new modeling to IEEE 802.11b/g wireless networks and study its interest.

In this article we address the frequency management during WLAN planning. Frequency management refers to channels interference and SINR computation. We propose a new approach where location selection and frequency assignment are tackled together during WLAN planning process. Two steps characterize this approach. Firstly we use all the available channels for frequency assignment. Secondly multiple signals are taken into account to compute the SINR. Several experimental results show the benefits of this new approach.

Standard tabu search methods are based on the complete exploration of current solution neighborhood. However, for some problems with very large neighborhood or time-consuming evaluation, the total exploration of the neighborhood is impractical. In this paper, we present an adaptive exploration of neighborhood using extension and restriction mechanisms represented by a loop detection mechanism and a tabu list structure. This approach is applied to the K-coloring problem and evaluated on standard benchmarks like DIMACS in comparison with more powerful recently published algorithms

Networks based on cellular concept suffer of bad performance when calls are done during handoff, i.e. during the transfers from one cell to another one. A part of these bad performances are due to the lack of management of overlapping area between cells. These imperfections find expression in cells discontinuity and distortion, in irregularity on inter-antenna distances, and in cells size variation. In this paper, we present several new criteria for cellular networks optimization based on geometrical computation on handoff and interference zones. The relevance of the model is studied on the basis of numerical and visual appreciation accorded to different networks. The huge impact of cell geometry improvement on radio quality of the network will be also discussed.

We focus on the dimensioning process of cellular networks that addresses the evaluation of equipment global costs to cover a city. To deal with frequency assignment, that constitutes the most critical resource in mobile systems, the network is usually modeled as a pattern of regular hexagonal cells. Each cell represents the area covered by the signal of a transmitter or base station (BS). Our work emphasizes on the design of irregular hexagonal cells in an adaptive way. Hexagons transform themselves and adapt their shapes according to a traffic density map and to geometrical constraints. This process, called adaptive meshing (AM), may be seen as a solution to minimize the required number of BS to cover a region and to propose a basis for transmitter positioning. The solution we present to the mesh generation problem for mobile network dimensioning is based on the use of an evolutionary algorithm. This algorithm, called hybrid island evolutionary strategy (HIES), performs distributed computation. It allows the user to tackle problem instances with large traffic density map requiring several hundreds of cells. HIES combines local search fast computation on individuals, incorporated into a global island-like strategy. Experiments are done on one real case representing the mobile traffic load of the second French city of Lyon and on several other traffic maps from urban fictive data sets.

In this paper, a semi-empirical model of field strength prediction combining theoretical results of propagation loss algorithms and artificial neural networks is considered. This approach expects to overcome some limitations inherent in existing semi-empirical models: linear behaviour of the statistical analysis used in the construction of the models, unfitness for learning new situations. The good results obtained in a dense urban area show that neural networks are a very efficient empirical method to compute new kinds of models which integrate theoretical and experimental data.