Precise positioning of cyclists is critical for road safety applications, yet there is no pervasive approach for it. We present SignAlign, a pervasive system that enables sub-meter positioning using only a smartphone, which many road users already carry. SignAlign uses a smartphone camera to detect road signs and estimates the distance to them. These signs act as reference points to determine a cyclist’s global position by matching them to known map data, such as a city’s digital twin. We evaluate SignAlign through real-world experiments in different urban environments. Results show that our approach achieves a median positioning accuracy of 0.64m. We show that in road safety applications (i.e., cooperative collision avoidance systems), SignAlign can reduce the probability of missing a collision from 0.61 to 0.005, compared to smartphone GNSS positioning. We demonstrate the real-life feasibility of SignAlign by showing that it maintains robust results even during nighttime. Furthermore, our proposed system can already cover large areas of cities around the world without requiring extra infrastructure. We show that SignAlign can operate on battery-constrained hardware, such as smartphones, in real time. By using only a smartphone, SignAlign offers a scalable, low-cost solution for enhancing pervasive road safety in urban mobility.

Every year, around 300,000 vulnerable road users (VRUs), such as pedestrians and cyclists, die from car accidents. Using cellular communication, such as 4G, 5G or in future 6G, in combination with cooperative collision avoidance systems is a promising approach to reduce these accidents. Integrating contextual information about the current movement of the VRU has been demonstrated to improve these systems. To achieve an accurate context prediction, sensors need to sample data for a relatively long period of time compared to the communication delay. These requirements for the context recognition time leave little room for communication delays in collision avoidance. The influence of the communication delay in 4G, 5G and future 6G is therefore an important research question. In this paper, we try to close this research gap by analysing how communication delays of 4G, 5G and future 6G influences collision detection performance in context-aware systems. We show how the system with a client-server architecture performs for different communication delays after the context has been created on the pedestrian side. For our experiments, we use our software VRU-CARE to simulate collision scenarios based on the Euro NCAP test protocol.

Everyday life scenarios where non-expert users (e.g., customers) are confronted with frontline service robots will become more and more likely. In particular, misunderstandings and incidents may occur during these interactions because of wrong expectations of the robot’s capabilities. Current applicable laws are based on technological assumptions from prior decades unsuitable to modern robotics and AI. The new AI Act as a part of the solution to this is still in development. In addition to the pure legal view, a technological viewpoint may be beneficial for establishing a fitting, trustful, and, thus, acceptable technology liability law. This work contributes to this by empirically evaluating the service robot non-expert user’s liability expectations, the use of robots, and well-being. The results in a DIY store environment significantly show that the store deploying the robot should be liable if an incident happens. Further, we examined that even a minor simulated incident affected the participants’ emotions and moods. Consequently, this influences their perception of liability while not mitigating their acceptance of frontline service robots.

A promising approach to further increase the safety of Vulnerable Road Users (VRUs) are cooperative collision avoidance systems. Cooperative collision avoidance systems actively integrate the VRUs in collision detection by using movement data from a VRU’s mobile device. While in recent years great attention was payed to solve technical challenges, e.g., regarding communication and sensor accuracy, little attention was payed to threats of privacy. However, the use and collection of such data poses certain privacy risks to the VRU. These privacy risks cannot be addressed by encryption alone. While some pseudonymisation approaches are used to protect the identity and location of VRUs, in this paper, we analyse to which extent the perturbation of movement data, specifically the speed data, can prevent the linkage of this data to a particular VRU, thus reducing the probability that this specific VRU can be identified. At the same time, we evaluate the trade-off between the probability of User Identification and the probability of collision detection. The evaluation is based on a standardised urban collision scenario between pedestrians and vehicles from the European New Car Assessment Programme. Our results show that privacy and "survival" are not mutually exclusive.

Every year, approximately 350,000 vulnerable road users (VRU), like pedestrians and bicyclists, die due to collisions with vehicles. One approach for reducing this number of fatalities is using VRU collision avoidance systems, which aim at detecting and preventing possible collisions. Existing products for VRU collision avoidance systems using sensors mounted on vehicles may already decrease the number of accidents. However, these products require a line-of-sight between the vehicle and a VRU. Especially in urban environments, the line-of-sight between a vehicle and a VRU is often obstructed, e.g., by parked cars. Cooperative VRU collision avoidance systems promise to be a solution to this limitation. In such cooperative collision avoidance systems, the VRU is equipped with a portable device like a smartphone or a smartwatch. This allows exchanging the movement information between the vehicle and the VRU. However, one open question is how accurately such a system needs to recognize the movement of the VRU to achieve the necessary system performance. In this article, we provide a comprehensive analysis of how the accuracy of recognizing a pedestrian' s movement influences the reliability of detecting collisions between the car and the pedestrian in cooperative VRU collision avoidance systems. We conduct this analysis for the “Car-to-Pedestrian Nearside” scenarios from the “Test protocol for Autonomous Emergency Braking VRU systems” of the European New Car Assessment Programme (Euro NCAP).

Around a third (29.7%) of all accident scenarios between pedestrians and cars occur in “non-line-of-sight” (NLOS) situations, i.e., when a pedestrian crosses the street with visual obstruction, like a parked car. On the other hand, current products for active collision avoidance for pedestrians and other vulnerable road users (VRUs) require a “line-of-sight” (LOS) between the vehicle and VRUs. Therefore, these systems might fail to detect a VRU early enough in NLOS situations. Cooperative collision avoidance systems are a promising approach to also handle NLOS scenarios. In a cooperative collision avoidance system, VRUs are equipped with a mobile device, like a smartphone, that recognizes the VRU's movement and exchanges information with the car via wireless communication, like 5G or DSRC. An open question is for which accident scenarios such systems show an improvement over systems that require LOS. In this article, we perform a simulation-based performance comparison between a cooperative and a camera-based collision avoidance system in a NLOS test scenario from the European New Car Assessment Programme (Euro NCAP) for different VRU types and car speeds. We compare the performance of both collision avoidance systems in terms of the collision detection probability and the time that a possible collision can be detected in advance.

Every year, approximately 350 000 vulnerable road users (VRUs) still lose their lives in road traffic accidents world-wide. To reduce this number, cooperative VRU collision avoidance systems aim at complementing current vehicle or infrastructure-based systems for VRU protection. The cooperative approach assumes user equipment (UE) like a smartphone or a smartwatch on the VRU side, which allows to obtain and exchange movement and contextual information with other road users for determining the current risk level for a collision. In recent publications, the usage of additional contextual information, such as the pedestrian's current activity (e.g., standing, walking or crossing a curb), has been shown to improve the collision detection accuracy. However, those approaches focused on the detection accuracy, but did not investigate how time delays for both detection and communication of VRU contexts affect the ability to detect collisions. We fill this gap by investigating the influence of both activity recognition and communication delays on the collision detection performance. As a baseline, we use the standardized Euro NCAP test protocol. For the evaluation, we exemplarily consider the usage of a curb detection module for position correction, which improves the collision detection accuracy at the moment the pedestrian crosses the curb.

Accurate and low-cost indoor localization continues to be a research challenge. The emerging communication technologies in 5G and 802.11 standards (e.g., 802.11n and 802.11 ac), such as massive MIMO and mmWave, provide opportunities for accurate indoor localization. Meanwhile, location information is also critical for the development of new technologies, e.g., directional communication using beamforming. It leads to a synergistic relationship between localization and communication. In this paper, we review the potential benefits of the emerging communication technologies (i.e., smart antennas, Massive MIMO, mmWave, beamforming, CSI) for indoor localization. It may provide a useful guide for researchers who are interested in developing indoor localization systems based on emerging communication technologies.

Feature selection algorithms are an essential part in many areas of pervasive computing such as Human Activity Recognition (HAR). Due to the increasing amount of information available and the large size of the HAR datasets, it has become important to reduce the dimensions of the data to minimize the computational costs of the recognition systems. This is particularly important for devices with limited computational and battery capacities such as smartphones. In this paper, we present FeSNOC, a novel feature selection algorithm based on the ecological measure of niche overlap. The niche overlapping coefficient is used as a measure of the similarity between classes and hence as an indicator of the classification accuracy. FeSNOC is a hybrid algorithm that combines the accuracy (obtained by using the K-Nearest Neighbor algorithm) and the overlapping coefficient to find the best features in a dataset. This algorithm consists of four simple steps, which do not require large computational times. We evaluate FeSNOC by using one publicly available dataset containing 561 features. We show that FeSNOC successfully selects the best features for the HAR problem outperforming other traditional feature selection approaches, such as the Information Gain and the Relief-F.

As Wi-Fi fingerprinting is a promising approach for indoor localization, one of the major challenges is understanding how fast fading and device heterogeneity problems influence the performance of the Wi-Fi fingerprinting systems. In this paper, we experimentally investigate the influence of fast fading and the device heterogeneity problems on the Wi-Fi RSS values, as well as on the performance (i.e., the recognition rate) of a Wi-Fi fingerprinting system WHERE. Besides, we also compare the performance of Wi-Fi fingerprinting in the experimental scenarios and in the real scenarios with the consideration of the fast fading problem. It concludes that both fast fading and device heterogeneity heavily influences the RSS values. However, the influence of fast fading on the performance of the Wi-Fi fingerprinting system can be mitigated when the Wi-Fi data is collected in the real scenarios.

Wi-Fi fingerprinting is a promising technique for indoor localization. The use of both 2.4 GHz and 5 GHz for Wi-Fi fingerprinting has been preliminarily studied by some research groups. However, most of the previous studies only consider the reflection and multi-paths caused by walls around but do not consider the damping and scattering through walls between them. In this paper, we particularly consider the situation that the radio propagation covers several rooms divided by walls, and compare the localization performance of using 2.4 GHz and 5 GHz networks. Moreover, we also compare the power consumption of scanning the surrounding wireless channels used for 2.4 GHz and 5 GHz networks. The results show that the performance of accuracy of Wi-Fi fingerprinting is similar when using 2.4 GHz and 5 GHz respectively. However, the use of 5 GHz results in higher response rate, while the use of 2.4 GHz consumes less power.

Place learning is a process of analyzing various sensor data (e.g., continuously collected Wi-Fi scans) to discover the places, which are related to a user's routines or activities in his daily life. While the geometry-based approaches can correctly distinguish places in different buildings, most of the fingerprint-based approaches can correctly distinguish places in different rooms. However, if two neighbouring places are very near to each other, they are likely to be learned as a single place. Previous studies do not specifically indicate the minimum distance to distinguish neighbouring places. In this paper, we deal with the question of "how near is near" for two neighbouring places, and experimentally investigate the minimum distance required to distinguish two neighbouring places. Results of the case studies show that places in adjacent rooms or with a distance of at least 4-5 meters can be correctly distinguished, while locations with a distance of fewer than 4 meters are likely to be learned as one place in the indoor environments.

Sensors embedded in smartphones are an essential component for activity recognition. Even though the accelerometer is the most widely used sensor, the highest recognition accuracies are obtained when using data collected from multiple sensors. However, the use of multiple sensors has an adverse impact on the energy consumption of power-limited devices such as smartphones. In this paper, we present a new method to improve the recognition accuracy of physical activities by using only the accelerometer. We utilize a low-pass filter to split the acceleration data into a low- and a high-frequency component. These components provide a new set of features, which can be used as a complement to the raw acceleration to reduce the number of sensors needed to recognize physical activities. After evaluating our method for a public dataset, we found that our approach represents an average of up to 16% increase in the recognition accuracy over the raw acceleration data, outperforming even widely used combinations such as the raw acceleration plus the gyroscope. The highest accuracies are obtained when using a cut-off frequency in the interval [0:001--0:05] Hz as well as a combination of the acceleration with its low-frequency component.

Unfortunately, every year many vulnerable road users (VRUs), such as pedestrians or bicyclists are killed or seriously injured in traffic accidents. Several research groups are working on different solutions including radar or laser based approaches to reduce the number of traffic accidents VRUs are involved. The approach presented in this paper is based on wireless communications such as WLAN and 5G/LTE and a "context filter". The "context filter" identifies vulnerable road users in potentially dangerous situations based on several contexts (VRU position, movement direction, accelerations). An identified dangerous situation is communicated between the VRU and the cars using wireless communication such as ad hoc or cellular systems. As one specifically interesting context for identifying dangerous situations this paper investigates on detecting pedestrians stepping onto the road. We examine smartphone sensor data and several reasoning classifiers to investigate whether "stepping onto the road" by detecting a pedestrian "crossing a curb" is possible. To the best of our knowledge detecting pedestrians crossing a curb by using smartphone sensors, has not been investigated so far.

A person's context data can be used for a multitude of applications, such as energy management or health care. Common context recognition approaches rely on several factors, such as the sensor set, features, or the context modeling algorithm. Discovering the recognition performances of different parameter setting combinations is a complex, time-consuming, and error-prone task. To support the context recognition research, we present the Context Recognition Assistance Tool (CRAT). The Context Recognition Assistance Tool assist by automatically conducting the evaluation for a multitude combination of parameter settings and clearly presents the findings. Using the CRAT, we investigate to what degree five parameters influence the recognition accuracy. To support the research in the field of context recognition, the CRAT is publicly available.

Smartphone sensors deliver useful information for applications such as indoor and outdoor navigation. An integral part of such applications is the detection of the orientation and movement direction of a smartphone user. Until now, movement direction detection using smartphones typically relies on GPS, which is often not available indoors. Alternatively, other approaches use sensors such as accelerometer and compass instead. These approaches rely on carrying the smartphone in a predefined orientation, or knowing the orientation of the smartphone in relation to the orientation of the user. In this paper, we present an approach to detecting the orientation and movement direction of users carrying smartphones inside the trouser pocket. This approach first determines the orientation of the smartphone's top using compass and orientation sensor. Second, this approach determines the orientation of the smartphone's screen, and the user's movement direction by observing compass and accelerometer during at least two steps the user takes. After these two steps, the approach is capable of continuously aligning smartphone orientation and user orientation. With our approach, the user is free to change direction, movement speed, or to stop moving at all. The smartphone can be placed in the trouser pocket arbitrarily. And the smartphone is free to wobble in the trouser pocket. How well our approach works, is investigated based on experimental measurements.

The built-in magnetometer sensor of a smartphone is an attractive sensor suitable for detecting the direction of pedestrians' movements. This detection can be used to enable collision avoidance between cars and pedestrians, particularly if the pedestrian is not visible from inside the moving car. However, surrounding metallic materials have undesired effects and this influences the accuracy of the magnetometer. This paper investigates the amount of magnetic deviation caused by surrounding metallic material, such as a car, when the magnetometer sensor of a smartphone is used to determine the orientation. A set of experiments were conducted to investigate the effects of magnetic deviation. The experimental setup includes both parked and moving cars. Based on the results of the experiments, we show that the magnetic deviation can influence the efficiency of distinguishing the endangered or secure pedestrians in a car to pedestrian collision avoidance scenario. If we intend to use the magnetometer to detect the pedestrians' movement directions, the magnetic deviation needs to be compensated in order to detect the accurate movement direction of pedestrians. Therefore, in this paper we present a compensation algorithm, which successfully compensates the possible amount of magnetic deviation in the magnetometer of a smartphone.

Sensors built into current smartphones or in modern facilities are becoming ubiquitous and influence our daily lives more and more. Hence, services that proactively adapt their behaviour to the user’s needs is no longer a vision of the future. The proactiveness of such services is achieved by using context prediction algorithms. These algorithms are primarily applied to personal context data of a user, for example to user’s motion activities or even to complete daily routines in order to predict a user’s next context. Due to this fact the paper presents the most significant legal criteria that have to be considered by the prediction approaches while using personal context data. Following, the criteria are used to legally assess the context prediction approaches. Finally, the resulting consequences are discussed.

Tragically, traffic accidents involving pedestrians or cyclists cause thousands of fatalities and serious injuries world- wide every year. Therefore, improving the safety of vulnerable road users is an international priority. One key challenge in designing an "ideal" protection system is to filter the endangered pedestrians out of potentially many. In this paper, we present a novel approach to proactively filter those pedestrians whose very next step would bring them (dangerously) closer to the street so as to provide an extra and crucial time advantage for a collision avoidance system. To predict a pedestrian's next step, we use the Collaborative Context Predictor. It takes advantage of collaborative behaviour patterns, in this case the movement patterns of the pedestrians. As well, a comparison with two state of the art context prediction approaches is carried out. The comparison is performed using real measured movement data, which is received from a smartphone the pedestrians carry in their trouser pocket.

Context prediction is the task of inferring information about the progression of an observed context time series based on its previous behaviour. Prediction methods can be applied at several abstraction levels in the context processing chain. In a theoretical analysis as well as by means of experiments we show that the nature of the input data, the quality of the output, and finally the flow of processing operations used to make a prediction, are correlated. A comprehensive discussion of basic concepts in context prediction domains and a study on the effects of the context abstraction level on the context prediction accuracy in context prediction scenarios is provided. We develop a set of formulae that link scenario-dependent parameters to a probability for the context prediction accuracy. It is demonstrated that the results achieved in our theoretical analysis can also be confirmed in simulations as well as in experimental studies.

In today's increasingly mobile society, car manufacturers and suppliers worldwide face the imperative requirement and have already been working to reduce the number of people killed in traffic accidents. In this context pedestrian protection respectively collision avoidance between cars and pedestrians is of vital importance. Several approaches exist to detect pedestrians but only a few of them are able to work without line of sight and/ or only have a very brief view of the movement history of the pedestrian. However, radio communication based systems which exchange GPS based position information of the pedestrian and the car might have a detailed view to the movement history but the update frequency of GPS positions provided by the mobile phone reduces the speed of the system and might induce false alarms. In this paper, the additional use of pedestrian movement recognition using an acceleration sensor implemented in a mobile phone in the area of collision avoidance between cars and pedestrian is presented and a performance analysis is carried out using the results of an experiment. Finally, the benefits of this approach to the reduction of false alarms and detection of endangered pedestrians are discussed.

The prediction of contexts plays an important part in the field of context aware systems and environments for adapting services proactively to users' needs. To the best of our knowledge, most research literature on context prediction focused on forecasting a user's contexts only using his available context history. In the case of a user suddenly changing his behaviour in an unexpected way, the context history of the user does not provide future context information for the observed pattern. Hence context prediction algorithms will fail to forecast the appropriate future context. To overcome the gap of missing context information in the user's context history, we propose the Collaborative Context Prediction (CCP) approach. Our results show that the proposed CCP approach is able to give accurate predictions in the absence of needed context information and outperforms the Active LeZi method.

Context-aware computing refers to a general class of computing systems that can sense their physical environment and adapt their behavior accordingly. In order to behave properly, such systems rely on valid and reliable context information. Therefore concepts for validating the correctness of context data are required. In our previous research we have introduced the Context Pattern Method (CPM) that calculates the probability of correctness measure, which quantifies the belief that the context information is correct and reliable. In this paper we use the CPM method and two well known statistical classification methods, Linear Discriminant Analysis (LDA) and the Support Vector Machines (SVM), to validate the context information and compare their performances under different conditions.

Improving the safety of vulnerable road users is an international demand because still thousands of pedestrians and cyclists are injured or killed in accidents with cars, trucks and busses. Car manufacturers and several research groups work on different solutions to avoid accidents with the help of sensor based systems. Moreover, research is done in radio communication based collision avoidance systems which uses exchanged information to distinguish between dangerous and non dangerous situations. In this paper we present various filters differing by the input information used to find dangerous situations implemented in a radio based collision avoidance system simulator. The results of the filters applied to a scenario are also presented. Furthermore, we analyse the influence of different conditional parameters to transmission delay times by taking further measurements for cellular networks to determine if cellular networks can be used as a communication technology in a collision avoidance system.

Context-awareness refers to computing systems that are able to sense and to comprehend their environment in order to adapt themselves in dependence to the available and relevant contextual information they depend upon. Gathering such contextual information involves real world entities such as sensors, which, for various reasons, are often prone to certain degrees of uncertainty and inaccuracy. Nevertheless, high quality context information plays a vital role in ensuring correct system behavior as well as dynamic system and service adaptation. Thus, a set of indicators is required which allows determining the quality of contextual information, which is commonly known as quality of context (QoC). One of the most relevant parameter of the QoC is the probability of correctness (PoC), which expresses the level of confidence, that the contextual information sensed, are in fact correct or not. In this paper, we propose an approach for measuring the PoC of context information by firstly analyzing the nature of context information and, secondly, revisiting the concept of quality of context also discussing other QoC parameters. Finally, we present a novel approach for quantifying the PoC for specific context information and evaluate the proposed method on a concrete case study.

Context-aware adaptive systems aim at automatically personalizing the user's environment depending on the user's situation, and hence, minimizing user interaction with the system. We present a novel user profile ontology that is dedicated to describe situation-dependent sub-profiles. This ontology can be used by context-aware adaptive service platforms for mobile communication and information services to automatically trigger the situation-dependent personalization of services. The design of this novel ontology also takes into consideration recommendations from the human factors research area. In particular, the ontology enables the easy specification of situational conditions and situation-dependent user sub-profiles.

Multi-standard radio resource management (MxRRM), based on a user data rate optimised algorithm, is studied for non-real-time (NRT) services in a heterogeneous WiMAX/UMTS/GSM scenario. This algorithm utilises a parameter, namely NRT load information, which is mapped from the average user throughput in a respective network. The performance level of MxRRM is assessed through a network simulator. A WiMAX cell covers a hotspot area in the co-located GSM/EDGE and UMTS/HSDPA networks. The load redistribution, the mean data rate and the intersystem handover activity are investigated in three scenarios: a) completely separate systems, b) separate WiMAX and MxRRM for GSM and UMTS, and c) MxRRM for these three radio access networks. An additional preliminary default access network selection utilises the network's knowledge about the capacity variety and the load status of the heterogeneous networks. Simulation results have shown a promising gain of mean data rate and a significant reduction of intersystem handover rate.

With the successful deployment of various radio access networks, multi-radio access has become an important research topic for heterogeneous wireless networks. This paper presents several multi-standard radio resource management strategies for integrated voice and data services. The input parameters as well as the availability of overflow at call setup stage and intersystem handover during call connection stage differentiate these strategies from each other. The heterogeneous networks capacity of integrated services is investigated via simulations modelling the stochastic traffic, mobility, and technology related air interface features. Performance comparison reveals that a strategy merely maximising the capacity might not consider the performance of individual service in some cases.

Context prediction mechanisms proactively provide information on future contexts. Due to this knowledge novel applications become possible that provide services with proactive knowledge to users. The most serious problem of context prediction mechanisms lies in a basic property of prediction itself. A prediction is always a guess. Since erroneous predictions may cause the application to behave insufficiently, prediction errors have to be minimised. The accuracy of prediction is seriously affected by the reliability of the context data that is utilised by the method. We study two paradigms for context prediction and compare their potential prediction accuracy. We show that the designer of context prediction architectures has to choose wisely as to which prediction paradigm to follow in order to maximise the accuracy of the whole architecture. We also introduce a simulation environment and present simulation results that support the gained insights regarding context prediction.

We present a user profile management approach that takes into account basic requirements for its execution on service platforms for ubiquitous computing environments. In particular, our approach provides means for managing and delivering context-dependent user profiles for several applications and decouples the application development from context processing. In addition, it takes into account the re-use and sharing of application-related user data between different applications. Hence, on the one hand, our approach considers the increasing trend to research and develop context-aware and personalized applications for the wireless world. On the other hand it also takes into account the fact that, in general, application and service developers do not agree on common semantics, which complicates the re-use and sharing of user data.

We study the impact of the context interpretation error on the context prediction accuracy. Benefits and drawbacks of current context prediction schemes are analyzed and opposed to a contemporary alternative. We propose a novel context prediction scheme that has the potential to significantly improve the context prediction accuracy. The impact of the context interpretation error on the context prediction accuracy is further analyzed in simulations inspired by our analytical considerations

Recent simulation studies of heterogeneous wireless networks have shown promising system performance improvements by using multi standard radio resource management (MxRRM) with a centralised MxRRM architecture. In the present paper, the distributed MxRRM architecture is studied, including the modelling of signalling, where intersystem cell load information can only be exchanged at certain events. Consequently, the distributed MxRRM entities have to base their decisions on potentially inaccurate load information. The system performance is investigated by simulation studies, where the results of distributed MxRRM scenarios are compared to two reference cases, 1) assuming ideal system information for MxRRM and 2) separated systems without any MxRRM functionality. Furthermore, an additional dedicated mechanism to exchange the load information aiming to improve the load information accuracy is investigated

The ability to predict future contexts significantly expands the possibilities of context-aware computing applications. However, an incorrect prediction may also mislead the application and may result in inappropriate application behaviour. We study influences on the prediction accuracy and propose a novel approach to context prediction in ubiquitous computing environments. In our paper we introduce a context time series prediction algorithm based on local alignment techniques. Our approach has the potential to improve the prediction accuracy since it explores the observed context history in more detail than current algorithms. In conclusion, we present simulation results that support our studies

The vision of pervasive computing and communication will shape individual life in the future. Following this vision, the goal for new business sectors will be to provide an easy way of creating and deploying different kinds of service environments. This paper introduces briefly the problem of overall architectures for service execution environments and identifies an architectural approach, the requirements and design guidelines to built service execution environments and related platforms that are fully compatible with each other.

Personalisation provides a convenient means to ease the use of pervasive computing systems and related services. Profiles are one of the major building blocks for making real personalisation happen. In the existing heterogeneous and upcoming pervasive computing environment typically numerous different profiles are used and maintained by the user or different services, devices or operators. In order to avoid multiple copies of the same data profile synchronisation and linking should be used and applied between the different locations and administrative domains, in which the different types of profiles reside. But keeping in mind the different contexts', in which these profiles have been constructed, it is mostly unknown how to match the meaning of one profile to another or how to harmonize the different kind of values in the different profiles. In this paper, mechanisms for efficient synchronisation of profiles based on semantic distinctions are introduced. By using ontology's to distinguish between semantic entities profiles can be kept synchronous even using low-bandwidth radio links and mobile devices like smart-phones and PDAs.

Developing system software for ubiquitous computing makes consideration of numerous requirements necessary. This is primarily due to the dynamic, mobile, unpredictable and heterogeneous environment of ubiquitous computing systems. Middleware provide solutions to common tasks in ubiquitous computing, e.g. communication, discovery, deployment, personalisation, group management etc. Middleware are developed using frameworks. Existing frameworks such as CORBA component model, Webservices, and Jini support middleware in certain aspects, e.g. communication protocols, but restrict them to certain assumptions, e.g. no disconnected operation or the client/server model. In this paper a framework for middleware in ubiquitous computing is presented. The framework is used to develop middleware with the service oriented architecture paradigm. The framework presented is applied to a middleware to allow for group based communication. The middleware, and especially the management entity of the middleware, is described in greater detail in this paper.

This paper outlines how the main- and subsidiary goals of the RACE R2084 Advanced TDMA Mobile Access (ATDMA) project being relevant for system performance evaluation will be put into action. Therefore, a four level top-down approach, the ATDMA system performance evaluation strategy, has been developed together with common evaluation requirements for comparisons of TDMA and CDMA radio access schemes.