Context management platforms (CMPs) address the need for provisioning context to support Context-Aware Mobile IoT applications. Such applications often require the context to be delivered under strict constraints including latency, freshness etc. With the explosion of IoT deployment and increased adoption of IoT applications, caching context at the edge becomes essential to meet these strict constraints. However, existing edge caching approaches, designed for static content or IoT data, do not account for the dynamic nature of context. In this paper, we propose Edge Context Caching Framework (ECCF), an adaptive and distributed edge-based context caching architecture. ECCF introduces three components: (i) a Context Variability Assessment Engine (CVAE) that quantifies how rapidly context changes and maps this variability to adaptive refresh intervals, (ii) an edge-based Local Caching Value Function (LCVF), which integrates multiple metrics including variability, freshness, access probability, network costs and provider SLAs to compute a utility value for improved caching decisions, and (iii) a gossip-based coordination protocol that enables edge nodes to exchange cache summaries for distributed replication and consistency among edge nodes. We implement ECCF on a Raspberry Pi 5-based real-world edge testbed and evaluate it using real-world smart city workloads with up to 85,692 queries per hour. Compared to seven state-of-the-art baselines, ECCF achieves 97% hit ratios under stable conditions and 85% during high volatility scenarios, reduces provider fetch rates by 2-3 ×, and maintains gossip overhead below 2 % of bandwidth.

Despite having billions of data sensing devices that virtually acquire data about anything, anywhere, not all smart applications are intelligent. Context-awareness being at the core of intelligence, it is a contradiction founded on the impotence of these applications to effectively leverage this plethora of big data and infer context in real-time. Hence, in this paper, we propose a conceptual architecture and instantiate a working prototype of the Distributed Contextual Intelligence System (DCIS) - a ubiquitous, mobile edge-based context management mechanism. Unlike any traditional Context Management System, DCIS is an ultra lightweight, event-driven, dynamically scaling agent-based system that overcomes the barriers to ubiquitous contextfacilitating context anywhere, anytime, and everywhere so that any application can be intelligent. Our prototype - Zoolocity, is based on edge devices that execute light-weight context inference on data that it can acquire and dynamically collaborate with nearby edges in real-time. Thus, unlike any previous works that do not provide any evidence or motivation towards distributed context management, we demonstrate how applications can use DCIS agents to become fully context-aware about any scene it is involved in, unlocking any intelligent feature. This is enabled by two key novel contributions: a method to spontaneously share interoperable context and infer geospatially sparse complex activities in near real-time using generalised entity interaction modelling.

Predicting the future location of mobile objects reinforces location-aware services with proactive intelligence and helps businesses and decision-makers with better planning and near real-time scheduling in different applications such as traffic congestion control, location-aware advertisements and monitoring public health and well-being. Recent developments in smartphone and location sensors technology and the prevalence of using location-based social networks alongside the improvements in AI and machine learning techniques provide an excellent opportunity to exploit massive amounts of historical and real-time contextual information to recognise mobility patterns and achieve more accurate and intelligent predictions. This unique survey provides a comprehensive overview of the next useful location prediction problem with context-awareness and the related studies. First, we explain the concepts of context and context-awareness and define the next location prediction problem. Then we analyse more than 30 studies in this field concerning the prediction method, the challenges addressed, the datasets and metrics used for training and evaluating the model and the types of context incorporated. Finally, we discuss the advantages and disadvantages of different approaches, focusing on the usefulness of the predicted location and identifying the open challenges and future work on this subject.

In Smart City applications, urban mobility involves complex interactions between traffic infrastructure, diverse road users, and physical environment. This paper addresses the limitations of conventional scene modeling methods that often fail to capture the varied and volatile nature of urban road scenes, particularly in representing the dynamic situations that unfold within them. Inaccurate situation representation hinders precise depiction of scene evolution, limiting our ability to understand and respond effectively to complex urban situations. This paper addresses these challenges by presenting the novel concept of the Context-Aware Scene Graph (CSG) for representing situations used in reasoning applications for enhancing safety and efficiency in urban environments, particularly for bicycle riders. CSG integrates multi-modal data, including ontological knowledge, sensor data, and images, to provide a comprehensive representation of urban road situations, enabling informed decision-making. This paper also validates the effectiveness of the proposed approach using real- world IoT datasets and camera images, with a focus on the bicycle dooring use case. The results outperform existing scene modeling methods by accurately representing situations, including those previously overlooked. The approach also ensures consistent representation, completeness, and captures transitions between situations, including causal relations. These findings highlight our approach's effectiveness in improving road safety, efficiency, and urban life quality through enhanced scene understanding.

Predicting situations in real-time applications is non-trivial. Fusing and incorporating the plethora of heterogeneous context information from many sources in the ecosystem that a user resides in to derive their situation is an expensive and time-consuming process. Yet context is useful only when a user can effectively make use of it in time and reliably. In this paper, using a proactive cyclist hazard alerting scenario, we propose a mechanism to proactively cache context information, so that cyclists are alerted of impending hazards before they might even occur. Our novel approach, which is capable of caching reliable predictive context information has significantly reduced the time to deliver context by 91% and the cost by 80%. We ensure the reliability of predictive cached context using a cross-verification routine that the false-positive rate tends to zero. The context cache is structured hierarchically such that our novel proactive context caching mechanism is capable of caching all low-level to high-level pieces of context, unlike any previous approaches.

Resilience in context-aware applications is especially important within challenged network and physical environments. This paper discusses the current methods for maintaining resilience in context management architectures. These methods largely focus on resolving contextual information loss, rather than maintaining the standard operation of the deployed system. The paper contends that this approach leads to loss of functionality and unwanted modifications to contextual information. Existing approaches for mitigating the effects of network instability have a high resource requirement and do not maintain the standard functionality of the system in real time. The paper proposes and discusses the Lightweight Context Management Architecture (LCMA) which addresses the lack of lightweight solutions for resilient context management systems. The functional requirements of the LCMA components are proposed and detailed. The proposed LCMA will be validated in mission-critical applications with potential adversarial actions.

Context-awareness (CA) has become an evolving trend, especially in the domain of Internet of Things (IoT). With the progress of IoT, the necessity for accessing real-time contextual information has become a critical factor for the advancement of IoT applications. Context management platforms (CMPs) have been proposed in the literature to support the needs of such Context-aware IoT applications. However, there are still significant gaps in terms of supporting the increasing needs of Context-aware applications, including the performance analysis of CMPs. In this paper, we propose a scene-graph based approach to generate context queries which primarily intends to support the performance analysis of CMPs and its ability to support plethora of Context-aware IoT application needs. Given the situation driven nature of IoT applications, the ability to generate relevant queries needs to be very realistic. Hence, we propose a novel Situation State Machine based approach to capture and model real-world situations. To demonstrate the potential to generate relevant context queries based on dynamic situations, a bicycle dooring use case is considered. We then present a template-based query generation approach to create realistic queries that represent real-world IoT application environment. The dooring use case is considered to validate the ability to represent complex queries, and the ability to generate complex queries in linear time.

Context-awareness is becoming more relevant for smarter modern-day applications. With billions of IoT devices able to monitor a plethora of parameters in near real-time, inferring contextual information at scale while maintaining adequate Quality of Context and delivering in time has been non-trivial for state-of-the-art centralized Context Management Systems. Further, handling complex situations and entity relations based on local awareness are areas that still need investigation. In this paper, we propose a novel edge-computing based architecture for distributed contextual intelligence systems that could address these research problems. First, we critically evaluate the current state-of-the-art in context-awareness and establish the necessity of a distributed architecture. Then, our proposed architecture is introduced along with the protocols and algorithms accompanied by real-world examples. The paper also highlights the future direction for research work in the area.

In this paper we introduce Jarvis, a context-as-a-service mobile tool, which enables context-aware data collection, service discovery, and computer-aided situational awareness through a conversational User Interface (UI). At the core of Jarvis are two main components: (i) a voice-based UI to translate speech to Context Definition and Query Language (Speech-to-CDQL), and (ii) an operational component called Context-as-a-Service (CoaaS), which enables smart things and IoT silos to discover, validate and share relevant and dependable context. The UI is based on two machine learning models: a Speech-to-Text model and a Text-to-CDQL model based on an encoder-decoder architecture. Jarvis is developed as a mobile application that allows people with different backgrounds to interact with various IoT devices. Our demo shows how easy Jarvis can be used for context-aware data collection and to interact with diverse objects in a smart home environment through voice.

Smart Cities are being designed and built for comfortable human habitation. Among services that Smart Cities will offer is the environmentally-friendly waste/garbage collection and processing. In this paper, we motivate and propose an Internet of Things (IoT) enabled system architecture to achieve dynamic waste collection and delivery to processing plants or special garbage tips. In the past, waste collection was treated in a rather static manner using classical operations research approach. As proposed in this paper, nowadays, with the proliferation of sensors and actuators, as well as reliable and ubiquitous mobile communications, the Internet of Things (IoT) enables dynamic solutions aimed at optimizing the garbage truck fleet size, collection routes and prioritized waste pick-up. We propose a top -- k query based dynamic scheduling model to address the challenges of near real-time scheduling driven by sensor data streams. An Android app along with a user-friendly GUI is developed and presented in order to prove feasibility and evaluate a waste collection scenario using experimental data. Finally, the proposed models are evaluated on synthetic and real data from the city municipality of St. Petersburg, Russia. The models demonstrate consistency and correctness.

This paper describes an advanced seminar presented at the 16th IEEE International Conference on Mobile Data Management. The advanced seminar presents the state-of-the-art in the Internet of Things, which is fast emerging as the disruptive technology for years to come. The seminar focusses on the Internet-scale sensor information management, related mobile analytics and open source IoT technologies and emerging standards.

As we are moving towards the Internet of Things (IoT), the number of sensors deployed around the world is growing at a rapid pace. Market research has shown a significant growth of sensor deployments over the past decade and has predicted a substantial acceleration of the growth rate in the future. It is also evident that the increasing number of IoT middleware solutions are developed in both research and commercial environments. However, sensor search and selection remain a critical requirement and a challenge. In this paper, we present CASSARAM, a context-aware sensor search, selection, and ranking model for Internet of Things to address the research challenges of selecting sensors when large numbers of sensors with overlapping and sometimes redundant functionality are available. CASSARAM proposes the search and selection of sensors based on user priorities. CASSARAM considers a broad range of characteristics of sensors for search such as reliability, accuracy, battery life just to name a few. Our approach utilises both semantic querying and quantitative reasoning techniques. User priority based weighted Euclidean distance comparison in multidimensional space technique is used to index and rank sensors. Our objectives are to highlight the importance of sensor search in IoT paradigm, identify important characteristics of both sensors and data acquisition processes which help to select sensors, understand how semantic and statistical reasoning can be combined together to address this problem in an efficient manner. We developed a tool called CASSARA to evaluate the proposed model in terms of resource consumption and response time.

One of the biggest challenges in the field of activity recognition is gathering training data for building activity inference models. To address this problem, we have developed an online activity toolkit for gathering activity data from online users. We use this data to build activity definitions for use in our system which is based on Context-Driven Activity Theory. We use Markov chain analysis to assign weights to activities and context attributes of a complex activity as well as to build activity signatures based on transition and path probabilities. Our demo is intended to show how complex activities and associated atomic activities and context attributes can be described using an activity toolkit. The toolkit is used to take input from users available online and the results analysis of different complex activities can be viewed online in near real-time using the graphical user interface (GUI).

In this paper, we propose and develop a platform to support data collection for mobile crowdsensing from mobile device sensors that is under-pinned by real-time mobile data stream mining. We experimentally show that mobile data mining provides an efficient and scalable approach for data collection for mobile crowdsensing. Our approach results in reducing the amount of data sent, as well as the energy usage on the mobile phone, while providing comparable levels of accuracy to traditional models of intermittent/continuous sensing and sending. We have implemented our Context-Aware Real-time Open Mobile Miner (CAROMM) to facilitate data collection from mobile users for crowdsensing applications. CAROMM also collects and correlates this real-time sensory information with social media data from both Twitter and Facebook. CAROMM supports delivering real-time information to mobile users for queries that pertain to specific locations of interest. We have evaluated our framework by collecting real-time data over a period of days from mobile users and experimentally demonstrated that mobile data mining is an effective and efficient strategy for mobile crowdsensing.

This paper proposes a situation and context-aware complex activity recognition system where we use on-body sensor information along with other context information to infer mobile user activities which are both concurrent and interleaved. We develop and validate our complex activity recognition algorithm for detecting these multiple complex activities. Activities are mapped onto situations using spatio-temporal analysis. We further build a test-bed in the social-networking domain to test and validate our approach in different scenarios using mobile devices.

Sensor networks represent an important component of distributed infrastructure supplying raw data to various applications from military to healthcare. A key challenge is cost-efficient collection of distributed data streaming from those sensor networks. In this paper we propose the use of mobile data collectors that employ K-NN queries as a cost-efficient approach to collect data within the sensor network. We investigate a 3D sensor network and propose a cost-efficient 3D-KNN algorithm that uses minimal energy and communication overheads to compute k-nearest neighbors. The 3D-KNN algorithm uses a 3 dimensional plane rotation algorithm that maps sensor nodes on a 3D plane to a reference plane identified by the mobile data collector We propose a cost-efficient KNN boundary estimation algorithm that computes KNN boundary based on network density. We also propose a neighbor prediction algorithm that uses distance, signal to noise ratio and mobile data collector' strajectory information to identify sensor nodes along the mobile data collector's path. We simulate the proposed 3D-KNN algorithm using GlomoSim and validate its cost efficiency by evaluating its energy efficiency and query latency. Lessons and results of extensive simulation conclude the paper.

Estimating the degree of traffic congestion at run time is crucially important for intelligent transportation systems (ITS) especially when selecting travel routes. One of the challenges when using a mobile sensor (e.g. mobile phone, GPS) as source of traffic data stems from its mobility. This paper proposes a novel approach to fusing mobile data and an algorithm for traffic congestion estimation to compensate intermittent available mobile sensors. Furthermore, our proposed model is also tolerant to incomplete data and it is open to any type of sensors. This approach differs from other traffic congestion estimation techniques in that it utilizes discoverable context even on sensorless road segments in real time instead of entirely relying on a sensor-based data of observed road segment. Implementation and validation through extensive experimentation with real data confirms feasibility of the proposed approach.

The emergence of context-aware systems has highlighted the requirement to reduce complexity in developing the systems by establishing a context middleware. The middleware aims to provide support for managing context independently from applications and enhance generality and reusability of context sources for multiple applications. In this paper we propose a novel framework for managing context in mobile ad hoc network environment (MANET). We model contexts and situations based on the concept of situation space in Context Spaces theory and integrate the model with Dempster-Shafer rule of combination for situation reasoning. Furthermore, we incorporate reliability of information in the fusion mechanism based on discount rule to adopt uncertainty factor that may exist due to limited and uncertain evidence in the hostpsilas environment. Finally, the implementation and experimentation issues of the prototype are also discussed.

This position paper argues that the mobile agents paradigm is a useful and important technology enabling pervasive and ubiquitous computing. Context awareness drives adaptability of pervasive computing systems. It is asserted that mobile agents capable of discovering, extracting, interpreting and validating context will make significant contribution to increasing efficiency, flexibility and feasibility of pervasive computing systems.

A priori estimation of quality of service (QoS) levels is a significant issue in e-services since service level agreements (SLAs) need to specify and adhere to such estimates. Response time is an important metric for data intensive e-services such as data mining, data analysis and querying/information retrieval from large databases where the focus is on the time taken to present results to clients. A key component of response time in such data intensive services is the time taken to perform the computation, namely, the time taken to perform either data mining, analysis or retrieval. In this paper, we present an approach for accurately estimating the computation times of data intensive e-services.

Most existing distributed file systems (DFS) implement a single consistency model to maintain one-copy equivalence. The functionality of that consistency model is based on a balance between environmental constraints and the targeted level of consistency. Such systems efficiently maintain consistency while the environmental capabilities remain constant, for example, presuming uninterrupted connectivity. However when these characteristics change, the inflexible nature of a single consistency model results in its inability, to maintain an expected balance between consistency and constraints. GLOMAR is a DFS middleware layer that allows application developers to map their specific consistency models to environmental constraints. As a result, multiple consistency models can be created, with each scoped for a particular application and environmental scenario. The paper outlines important aspects of GLOMAR, detailing its implementation and outlining a consistency model implementation based on Web services and the SOAP protocol.