Search

In MANETs, congestion typically occurs on the interconnecting nodes between two or more groups of nodes. Routing to avoid the congested nodes via alternate, perhaps longer paths, allows more throughput, e.g., 50% more in the canonical 9-node 2-ring scenario. OLSR-Q is based on the routing protocol OLSR and a reinforcement learning (RL) agent to learn the most appropriate link states or "Directional Air Time" metric to avoid the congested nodes. The challenges for the RL agent are (1) to avoid congestion before packets are dropped and (2) to minimize the number of real valued or discrete observations or states. In this paper, three simplified OLSRd2-Qx versions are presented and compared to OLSRd2 and a centralized ODRb, Omniscient Dijkstra Routing-balanced, algorithm. The proposed OLSRd2-Qload algorithm provides the expected 50% increase in throughput on the 9-node 2-ring scenario with a specific test traffic scenario. On the NATO IST-124 Anglova scenario, and using an acknowledged message application, the Q-learning agents remain to be improved. The superior results of the centralized load balancing approach taken in ODRb will be investigated to train multi-agents systems including OLSR-Q.

A Q-learning solution called OLSR-Q was proposed and tuned by McAuley et al. in 2012 to improve the performance of Optimized Link State Routing (OLSR) using a TDMA MANET simulation and a testbed with real WiFi cards. Here, the use of either IEEE802.11 or TDMA radios are compared using real time emulations based on the open source EMANE platform implemented on a cluster of virtual machines. The open source OLSRd and newer OLSRd2 are analyzed and compared to our implementation called OLSRd2-Q2 of the OLSR-Q algorithm. The benefit of our OLSRd2-Q2 is clear when using TDMA radios but not when using IEEE802.11 radios. These preliminary results confirm the potential and the practical challenges of cognitive approaches to improve the MANET performance of the open source implementation OLSRd2.

Simulation or emulation of mobile ad hoc networks (MANET) is used to predict or analyze the performance of MANETs under various scenarios. One challenge is to emulate realistically the MANET's radio performance. Running the Extendable Mobile Ad Hoc Network Emulator (EMANE) framework, we show how to reproduce measured characteristics, namely throughput and round-trip time, of real tactical radios using wideband or narrowband TDMA-based waveforms. Additionally, a solution to simulate rate adaptation is proposed. An introduction to EMANE and the EMANE radio model plugins is also provided.

A realistic troop deployment scenario for tactical MANET was developed within the NATO IST-124. The NAT0 IST-150, which relies on Anglova for benchmarking, has recognized a potential drawback in a subset of this scenario: there is a lack of long (> 4) multi-hop paths. Our contribution confirms this drawback using, for the first time, to the best of our knowledge, EMANE emulations based on realistic TDMA radio models combined with the OLSRd implementations. In addition, we propose a scenario, called 1km-chain, which is shown to be quite challenging for low power radios and might be of interest for future profiling of MANET performances. Our 1kmchain scenario simply modifies the Anglova Company 1 scenario based on stopping particular nodes and on using a dual-slope propagation model.

One of the several challenges for high capacity wireless communications is to find propagation models able to predict as easily and accurately as possible the propagation channel to assess the performance of wide-band wireless systems. An investigation of simple propagation models using detailed point cloud topographical data available in Switzerland is conducted. Our point cloud based predictions are compared to measurements in 2 ways: wide band impulse response at VHF in hilly terrain and 5GHz-WiFi outdoor picocell connectivity. The preliminary results contribute to the assessment of the potential, but also the current limitations of using precise maps but simple propagation models to predict measurements.

Accurate site-specific radio propagation simulations provide an important basis for cellular coverage analysis. The quality of these simulations relies on the accuracy of environmental description and electrical properties of constituent materials. This paper presents a novel method of on-site permittivity estimation. The method utilizes an accurate geometrical database of the environment for identifying flat and smooth surfaces producing reflections. The method exploits a limited number of on-site channel sounding to extract reflected multipaths and compare them with ray tracing based on the environmental database. The permittivity of the identified reflecting surfaces is estimated by solving an inverse reflection problem. The method was experimentally tested with limited radio channel measurements at 60 GHz in a large empty office room. The identified reflecting surfaces are classified according to their mean permittivity estimates, showing their consistency with physical material evidence and the permittivity database in the International Telecommunication Union Radiocommunication Recommendation (ITU-R P.2040-1). The estimated permittivity values are visualized as a 3-D map, giving an intuitive understanding of materials constituting the environment. This paper demonstrates on-site permittivity estimation and material classification without the need for isolated measurements of composite materials in an anechoic chamber or in situ measurements of built environments.

The Swiss Army uses tactical radios to communicate between troops on the field. Due to the operating environment, these radios experience specific characteristics such as limited bandwidth, in addition to high delay and packet losses. This paper presents the work achieved to develop, evaluate and test a novel application-layer routing algorithm specifically designed for tactical MANET networks. In order to evaluate the proposed algorithm in real conditions, two platforms and specific Quality of Experience metrics have been developed.

Describing the environment using a point cloud is a promising method for both propagation predictions and channel simulations. Recently, a specular reflection model has been proposed and verified through measurements to account for reflections from large reflecting surfaces. Here, the formulation for the specular reflections for point cloud (SRPC) is improved based on a few canonical measurements in an anechoic chamber. The specular reflection image theory for large surfaces, the Radar Cross Section equation for small surfaces, the Fresnel-Kirchhoff integral formula, and measurements are investigated. Unexpectedly to the authors, the results show that, when neglecting the diffraction effects, the contributions to the specular reflected field occur mainly from the illumination of about a third of the first Fresnel zone. A prediction of the SRPC formula is the appearance of “radio flares” due to electrically small reflecting surfaces.

The use of the intelligent packet scheduling process is absolutely necessary in order to make the radio resources usage more efficient in recent high-bit-rate demanding radio access technologies such as Long Term Evolution (LTE). Packet scheduling procedure works with various dispatching rules with different behaviors. In the literature, the scheduling disciplines are applied for the entire transmission sessions and the scheduler performance strongly depends on the exploited discipline. The method proposed in this paper aims to discuss how a straightforward schedule can be provided within the transmission time interval (TTI) sub-frame using a mixture of dispatching disciplines per TTI instead of a single rule adopted across the whole transmission. This is to maximize the system throughput while assuring the best user fairness. This requires adopting a policy of how to mix the rules and a refinement procedure to call the best rule each time. Two scheduling policies are proposed for how to mix the rules including use of Q learning algorithm for refining the policies. Simulation results indicate that the proposed methods outperform the existing scheduling techniques by maximizing the system throughput without harming the user fairness performance.

Laser scanning of physical environments allows us to construct their accurate structural database. This database is called a point cloud. Radio channel prediction employing the point cloud is a promising new method for millimeter-wave (mm-wave) frequencies. Reported works that predict the radio channel based on the point cloud utilizes a diffuse scattering model with reasonable accuracy if the environment under consideration is small and diffuse dominant. Specular reflections are also crucial propagation phenomena at mm-wave frequencies. However, a mathematical formulation to predict specular reflections is still lacking. This paper presents the formulation for simulating specular reflections based on a point cloud geometrical description of an environment. It evaluates the total specular reflected field as a coherent sum of the individual field contributions originating from the elementary surfaces over an area approximately equal to the first Fresnel reflection zone. The numerical and experimental results demonstrate the accuracy of the proposed formulation.

Mobile cellular communication has already entered the mass market, and mobile Internet services will soon become a reality. The frequent use of mobile radio technologies wherever people are has a direct impact on the deployment of base stations or radio access points, including antennas. Put simply, to serve an increasing number of users requires an increasing number of base stations. Thus, operators must carefully plan the deployment and configurations of radio base stations in order to support voice and data traffic at a level of quality expected by customers. Planning tools are used to help radio engineers in their difficult tasks of balancing requirements or radio coverage and quality with economic and other practical aspects. These planning tools make extensive use of terrain databases and of visualisation tools that lead some tool vendors to offer functionalities very similar to a geographical information system (GIS) or even to base their product on a GIS. Futhermore, because radio communication between base stations and users is crucial, all computations in a planning tool are based on the use of radio-propagation predictions. Until recently, empirical propagation prediction seemed sufficient. However, more efficient planning and the planning of nonvoice services or of a mixture of voice and nonvoice services require more accurate propagation-prediction models. These propagation models are usually based on the computation of the physical interaction of ratio waves and the environment. Thus more detailed databases are required, especially in urban environments where most users are located. Although the authors are not GIS specialists, they aim to expose some of the relationship between radio-propagation models used for mobile radio network planning and databases for terrain and buildings. For example, an accuracy of about 1 m on the coordinates of building corners is found to be acceptable in terms of accuracy in radio coverage. As another example, simulation results show that the use of conventional propagation models and rough geographical databases for the planning of future cellular systems like UMTS might cause serious difficulties.

Maintaining a desired trade-off performance between system throughput maximization and user fairness satisfaction constitutes a problem that is still far from being solved. In LTE systems, different tradeoff levels can be obtained by using a proper parameterization of the Generalized Proportional Fair (GPF) scheduling rule. Our approach is able to find the best parameterization policy that maximizes the system throughput under different fairness constraints imposed by the scheduler state. The proposed method adapts and refines the policy at each Transmission Time Interval (TTI) by using the Multi-Layer Perceptron Neural Network (MLPNN) as a non-linear function approximation between the continuous scheduler state and the optimal GPF parameter(s). The MLPNN function generalization is trained based on Continuous Actor-Critic Learning Automata Reinforcement Learning (CACLA RL). The double GPF parameterization optimization problem is addressed by using CACLA RL with two continuous actions (CACLA-2). Five reinforcement learning algorithms as simple parameterization techniques are compared against the novel technology. Simulation results indicate that CACLA-2 performs much better than any of other candidates that adjust only one scheduling parameter such as CACLA-1. CACLA-2 outperforms CACLA-1 by reducing the percentage of TTIs when the system is considered unfair. Being able to attenuate the fluctuations of the obtained policy, CACLA-2 achieves enhanced throughput gain when severe changes in the scheduling environment occur, maintaining in the same time the fairness optimality condition.

In LTE-A cellular networks there is a fundamental trade-off between the cell throughput and fairness levels for preselected users which are sharing the same amount of resources at one transmission time interval (TTI). The static parameterization of the Generalized Proportional Fair (GPF) scheduling rule is not able to maintain a satisfactory level of fairness at each TTI when a very dynamic radio environment is considered. The novelty of the current paper aims to find the optimal policy of GPF parameters in order to respect the fairness criterion. From sustainability reasons, the multi-layer perceptron neural network (MLPNN) is used to map at each TTI the continuous and multidimensional scheduler state into a desired GPF parameter. The MLPNN non-linear function is trained TTI-by-TTI based on the interaction between LTE scheduler and the proposed intelligent controller. The interaction is modeled by using the reinforcement learning (RL) principle in which the LTE scheduler behavior is modeled based on the Markov Decision Process (MDP) property. The continuous actor-critic learning automata (CACLA) RL algorithm is proposed to select at each TTI the continuous and optimal GPF parameter for a given MDP problem. The results indicate that CACLA enhances the convergence speed to the optimal fairness condition when compared with other existing methods by minimizing in the same time the number of TTIs when the scheduler is declared unfair.

Ray-tracing-based predictions in urban microcellular environments require databases for building layouts, electrical characteristics of buildings, and base stations (locations, antennas, power, etc.). The aim of this paper is to provide help in selecting the appropriate level of accuracy required in these databases in order to achieve the best tradeoff between database costs and prediction accuracy. The effects of inaccuracies in these databases are presented and analyzed by comparing predictions and measurements. The results presented here show to what extent errors, which are due to automatic vectorization of scanned maps, could lead to erroneous predictions. Furthermore, an analysis of the influence of random errors in a building vector database was performed to quantify the prediction error as a function of the accuracy in the building vector databases. Ray-tracing prediction models implementing a reflection and diffraction phenomena were found to be sensitive to the choice of the reflection coefficient attributed to building walls. This dependence can be used to fit the measurements as the complexity of real building walls does not allow one to easily derive their electrical parameters from which a reflection coefficient could be computed. It was also found that, in general and in agreement with measurements, ray-tracing-based prediction models are not sensitive to small variations on a base-station location. Finally, the sensitivity study also lead to gained insight of the propagation phenomena involved in urban microcell environments.

This paper presents a full-wave characterization of an indoor office environment at wireless local area network (WLAN) and worldwide interoperability for microwave access (WIMAX) frequencies using finite-difference time-domain (FDTD) numerical technique. The aim is to demonstrate the applicability of accurate full wave oriented approach for large scale propagation prediction in contrast to conventional approximate ray models. Numerical simulation were carried out using a 3D full-wave electromagnetic solver SEMCAD-X and simulated results including pathloss and small scale fading characteristics are validated through multiple-input multiple-output (MIMO) channel measurements. The simulated and measured results are found to be in close proximity of each other.

The tradeoff concept between system capacity and user fairness attracts a big interest in LTE-Advanced resource allocation strategies. By using static threshold values for throughput or fairness, regardless the network conditions, makes the scheduler to be inflexible when different tradeoff levels are required by the system. This paper proposes a novel dynamic neural Q-learning-based scheduling technique that achieves a flexible throughput-fairness tradeoff by offering optimal solutions according to the Channel Quality Indicator (CQI) for different classes of users. The Q-learning algorithm is used to adopt different policies of scheduling rules, at each Transmission Time Interval (TTI). The novel scheduling technique makes use of neural networks in order to estimate proper scheduling rules for different states which have not been explored yet. Simulation results indicate that the novel proposed method outperforms the existing scheduling techniques by maximizing the system throughput when different levels of fairness are required. Moreover, the system achieves a desired throughput-fairness tradeoff and an overall satisfaction for different classes of users.

In the short span of less than a decade the mobile phone has become a ubiquitous feature of life in India. Everyone from the chai-wallah to the CEO has a cell-phone and in recent years many of these are smart-phones capable of running smart-apps. Our focus is on conceiving a practical and useful app that can aid in disaster prevention and actually implementing and testing it. Towards this end we created PhoneGuard -- a smart-app that converts the phone into a remote monitoring device -- you leave the phone in a sensitive location like on the banks of a river or inside a coal-mine and it periodically takes pictures, does simple image analysis and checks for coherence over time, if warranted (e.g. in case of flooding of the river or buckling of beams in a coal-mine) it raises an alarm and follows an escalation procedure to push live images to a webpage. A complete working system was developed using standardized software and tested using realistic conditions. We believe that the image analysis and its integration into an escalation procedure is a novel aspect of our system.

International audience; In the context of heterogeneous networks, femtocells are very promising. In order to properly simulate their behavior and their impact on the macrocell layer, it is necessary to be able to simulate the radio coverage of femtocells. Hence ParFlow is a possible deterministic model that can be used for such simulation. In this paper two implementations of ParFlow are presented: time domain and frequency domain. The performance are compared and the advantages/drawbacks of each model are investigated.

International audience; Owing to its direct applicability in solving problems of the telecommunications industry, propagation prediction for a long time has been an important area of research and development. Because of the increasing complexity of wireless networks, growing number of smaller cells, and higher intercell interference, software tools that aid in network optimization are necessary. There fore, in the study of particular environments, where wireless networks are deployed, deterministic propagation models play an important role. Deterministic models can predict the path loss in a given scenario through the simulation of the main propagation phenomena such as reflections and diffractions. The most commonly used deterministic solutions for radio coverage prediction are ray models, which model electromagnetic (EM) waves through optical laws. However, since their appearance nearly 40 years ago, there has been an increasing interest in using other techniques such as finite-difference EM methods. These methods [the most well known is finite-difference time-domain (FDTD) method] solve Maxwell's equations on a discrete space time grid. They can provide highly accurate results, provided there are accurate boundary conditions, since the reflections, diffractions, and scattering effects are implicit. This contrasts with ray-based methods that must account explicitly for a limited number of rays. This also explains why Yee's FDTD methods, have been and are still widely used today, especially in confined environments such as for antenna design that involves reasonable grid sizes.

In this paper a new approach for the calibration of electromagnetic material parameters (permittivity) of 3D ray tracing prediction model at 2.4 GHz is presented. The calibration procedure is based on measurements of the channel impulse response using cross-correlation techniques and the 802.11b preamble. Further developments of a method, which has been published earlier, are presented. The analysis of the permittivity of different walls determined from a short measurement campaign is shown

This investigation outlines the importance of accurate radio propagation predictions in the planning for UMTS radio networks. The UMTS radio technology based on variable spreading DS-CDMA techniques is flexible but complex. The sensitivity of a UMTS radio network to the radio environment and traffic conditions is well publicized but rarely quantified. Theoretical analysis and commercial implementation of CDMA cellular systems also tend to show that radio network planning, simplified on one hand because of the lack of frequency planning, could be cumbersome due to difficulties linked to the so-called cell breathing effect. Since this effect is related to the traffic conditions and the radio propagation environments, the more accurate the coverage predictions are, the closer the expected performance will be from the measured performance. To quantify the advantage of accurate radio coverage predictions a case study was undertaken for a UMTS radio network planning in a 5 km/sup 2/ area of Paris (France). This study in an urban environment concludes that a conventional propagation model could lead to erroneous planning with less than expected quality of service, unacceptable interference, and more base stations than necessary. An accurate ray-tracing model integrated in an UMTS-capable planning tool allows the radio network designer to reach optimal levels for the base station deployment and configuration while meeting the expected service level requirements.

Private, corporate and public Wireless Local Area Networks (WLAN) Hot-Spots are emerging. In this rapidly evolving environment, the configuration of WLAN access points raise the classical problem of re-using limited radio resources. In this paper, the problem of dynamic frequency allocation of WLAN access point in a highly competitive multi-provider Hot-Spots environment is addressed. Our solution aims at working in locations where planned and ad-hoc deployments might be side by side. An on-line adaptive optimization process is proposed and relies on available information delivered only by the local access points in order to maintain the quality of service as high as possible. This optimization process is implemented on a scalable and highly flexible agent-based framework. The easy deployment of intelligent agents in a real WLAN network and their integration in a simulation context allows us to perform extensive tests for small and large-scale networks. The proposed approach has been tested on a limited but practical demonstrator that showed encouraging results.

This paper presents a preliminary comparison of two coverage prediction methods developed for urban microcellular environments. In this study a particular building layout is chosen as an example. The two considered methods are (1) a ray tracing method taking into account multiple specular reflections and single diffraction phenomena and (2) a time domain approach based on a transmission line matrix (TLM) method. Both methods consider infinitely high buildings and no ground reflection. The results of both methods are compared to measurements provided by KPN Research, (Netherlands). The prediction results of the ray tracing clearly show the contributions of the reflected ray which can be traced individually. The TLM method also produces results where reflections can be identified although not as clearly. Comparisons of prediction results provide an insight on measurements that can be performed to assess the existence of specular reflection in microcellular environments. Good agreement is achieved between either predictions and the measurements, even though the relevant buildings are not always above the base station antenna but only cross the radials from the base station antenna to the mobile.

The application of diffraction theory and geometrical optics for modeling the propagation in microcellular urban environment is investigated. The model simplifies the reality by considering only a two-dimensional geometry where the building walls are modelled by segments. This paper presents comparisons between several predictions and measurements in the microcell environments found in two different cities. The study shows the limitation of the model considered and points out the most important parameters affecting the predictions.

Future and current radio communications systems based on dynamic channel allocation or CDMA do not require any or very limited frequency planning. This leads to emphasize the importance of coverage predictions in radio planning tools for the future. Current radio propagation predictions are usually based on computations which must be performed separately for each cell in a radio network. Thus, the computation time is directly proportional to the number of cells considered. A scheme named ParFlow can inherently provide coverage predictions for one or several cells without any major effect on the computation time. The capability of the multiple cells simulations of the ParFlow scheme is demonstrated and analyzed for both the computation of composite coverage and the prediction of best server maps. The performance of parallel computations over a network of PC-based workstations is also investigated.

This paper presents a quasi 3D ParFlow approach for simulating EM wave propagation in the urban environment. The approach extends a 2D ParFlow to simulate the 3D propagation approximately by introducing a transparent boundary node. Originally the transparent node is for simulating the wave propagation through a building wall. However by properly choosing the parameters, the over-rooftop propagation can be simulated equivalently to the energy penetrating through the building wall. The approach has been applied to 3 kinds of typical cell in an urban area in order to evaluate the accuracy of the coverage prediction. The comparisons between the simulations and measurements show that the algorithm is applicable for coverage predictions in mobile radio networks. This work was undertaken since a pure 3D ParFlow algorithm is not a trivial upgrade of 2D ParFlow and requires tremendous computation resources.

The control of the power transmitted by the radio base stations has been proposed to improve the quality or to increase the capacity of a GSM cellular system. The performances of an downlink power control algorithm proposed by one GSM manufacturer are investigated. This algorithm is based on a linear control technique taking into account both the received signal strength and quality of the radio link. The algorithm characteristics are first analysed from a theoretical point of view. Furthermore, a stability analysis of the proposed algorithm is provided. It is pointed out that unstable states exist also in the domain of validity of the algorithm parameters. The second step of the analysis consists in simulating the nonlinear nature of the GSM radio link parameters. Measurements have been undertaken in two cities in Switzerland to assess the performance of various parameter settings on the downlink power control. Measurement results are analysed and compared to the theoretical investigation. Possible performance degradation due to the simultaneous use of downlink power control and frequency hopping over all transmitted carriers are also shown.

The ParFlow approach solves wave equations in the discrete time and space domain. ParFlow implementation has been designed for the coverage prediction for mobile networks in urban environments. In order to speed up the iteration process and save the computational memory, ParFlow uses a larger grid size than required by its constraint. This paper studies the effect of large grid size on the simulation results, in order to quantify the maximal allowed grid size for a satisfying coverage prediction. Applying a larger grid size mainly influences the simulation results in two ways: (1) it lowers the computation frequency which causes the changes of propagation attenuation rate, interfering pattern, and waveguiding effect between the building walls; (2) it distorts the building boundaries which leads to the change of scattering and artificial propagation blocking. Simple case studies and simulations enable us to propose criteria for adequate grid size which guarantees the required prediction quality. Furthermore the proposed criteria have been applied to the coverage prediction for part of the city of Munich in Germany. Compared to the measurements, the behavior of predictions to the grid size confirmed the proposed grid size criteria in this paper.

Based on measurements in five urban environments in three different cities, the accuracy of a ray tracing based model is characterized. First, the conventional characterization using the mean and standard deviation of the prediction error is used. Then, a figure of merit defined as the percentage of correct prediction with respect to a coverage threshold is computed. Finally, an example is given to illustrate how prediction results could be displayed to provide some insights into the accuracy of the model. For comparison purpose, results from power law curve fits are also analyzed. It is concluded that the standard deviation is a fair characterization of the accuracy of a model as long as its value is small (

This paper investigates the wave propagation in the small and micro-cell of urban environments. The goal was to identify the main wave propagation contribution as a function of the base station (BS) antenna height, the distance from the base station and the environment. Three typical cases were studied where the BS antenna is: (1) above the rooftop, (2) at the rooftop height and (3) much below the surrounding buildings (microcell case). A 2D ray tracing in the horizontal plane and a 2D ray tracing in the vertical plane have been combined to perform the coverage predictions in the three cases. The prediction results are compared to measurements and the main propagation contributions are identified.

This paper investigates the effects of the building heights on the results of our MCOR prediction tool in the area of Munich (Germany). The goal is to compare predictions performed assuming buildings having a constant height and either (1) predictions considering the buildings with their real heights or (2) measurements. The three dimensional building layout (contour and height) and the topographical data are the geometrical inputs of the prediction tool. The study shows that predictions performed with the real building heights can be well approximated by predictions made with buildings having a constant height. The errors between the two predictions are Gaussian with a standard deviation of about 4 dB. A reasonable approximation of the constant value for the building height is given by the mean building height. Another good approximation of the constant building height value is obtained from a statistical analysis based on a set of measurement points.

The databases required for ray-tracing-based-prediction in urban microcellular environments include those for the building layout, the electrical characteristics of the buildings, and for the base station (locations, antennas, power, etc.). The effects of inaccuracies in these databases are presented and analyzed by comparing the computed results to measurements. The results presented show that acceptable predictions can be computed but care must be taken (a) to use an accurate vector database, (b) to precisely place the base station location with respect to the building corners, and (c) to adapt the reflection coefficient for the given area.

We consider a new approach to modeling wave propagation in urban environments, based on the Transmission Line Matrix (TLM) method. Two-dimensional simulations are performed using a map of a city A renormalization technique is proposed to convert the results to the three-dimensional space. Our approach provides good predictions for the intensity of a wave when compared with in-situ measurements and is appropriate to very fast massively parallel computations. In order to provide a performance analysis, the algorithm has been used as a benchmark on different parallel architecture (CM200, CM5, IBM SP2 and Cray T3D).

A lattice Boltzmann model which simulates and predicts, on a massively parallel computer, wave propagation in urban environments is presented. This technique takes into account complicated boundary conditions. Two-dimensional simulations are performed starting from a city map and a renormalisation scheme is proposed to extend the results to three dimensions and adjust the wave length. The method, which is simple and easy to implement, provides good path loss predictions when compared with in situ measurements.

A numerical study of waves propagating in a turbulent, stratified medium, such as the ionosphere, is performed by using the phase-screen-diffraction-layer method. In this method, phase fluctuations due to irregularities and diffraction effects are computed sequentially along the ray path. Stepping in altitude, phase changes are imbedded into phase screens, while diffraction between phase screens is computed using fast Fourier transform techniques. In the reflection region, a careful examination of the propagation is required. The derivation presented in this paper leads to improvements in the computational scheme and gives some physical insights. The resulting scheme is used to compute the angle of arrival and some of its statistics as an example of practical applications.

2D and 3D ray-tracing-predictions based on UTD are compared to measurements (at 945 MHz) in urban microcellular environments characterized by an irregular mixture of building heights due to the presence of relatively high rise buildings. It is shown that the 2D ray-tracing underestimates the measurement in an area far from the transmitter. The theoretical study and the preliminary comparisons with measurements in Rotterdam (NL) showed that the 3D backward diffraction by high rise buildings might account for the propagation in an area far from the transmitter and overcome the limitation of the 2D model in these areas. Trees were found to influence heavily the 2D prediction. Taking into account the same 3D contributions becomes even more important when there are obstructions such as trees in the 2D plane. A reasonable agreement with measurement can be obtained with a combined 3D and 2D (considering the absorption effects of trees) predictions.

International audience