With the rapid proliferation of the Internet of Things, video analytics has become a cornerstone application in wireless multimedia sensor networks. To support such applications under bandwidth constraints, learning-based adaptive quantization for video compression has demonstrated strong potential in reducing bitrate while maintaining analytical accuracy. However, existing frameworks often fail to fully exploit the fine-grained quality control enabled by modern blockbased video codecs, leaving significant compression efficiency untapped.

Edge Video Analytics (EVA) has become a major application of pervasive computing, enabling real-time visual processing. EVA pipelines, composed of deep neural networks (DNNs), typically demand efficient inference serving under stringent latency requirements, which is challenging due to the dynamic Edge environments (e.g., workload variability and network instability). Moreover, EVA pipelines face significant resource contention due to resource (e.g., GPU) constraints at the Edge. In this paper, we introduce OctopInf, a novel resource-efficient and workload-aware inference serving system designed for real-time EVA. OctopInf tackles the unique challenges of dynamic edge environments through fine-grained resource allocation, adaptive batching, and workload balancing between edge devices and servers. Furthermore, we propose a spatiotemporal scheduling algorithm that optimizes the co-location of inference tasks on GPUs, improving performance and ensuring service-level objectives (SLOs) compliance. Extensive evaluations on a real-world testbed demonstrate the effectiveness of our approach. It achieves an effective throughput increase of up to 10× compared to the baselines and shows better robustness in challenging scenarios. OctopInf can be used for any DNN-based EVA inference task with minimal adaptation and is available at https://github.com/tungngreen/PipelineScheduler.

Test-time adaptation (TTA) allows a model to be adapted to an unseen domain without accessing the source data. Due to the nature of practical environments, TTA has a limited amount of data for adaptation. Recent TTA methods further restrict this by filtering input data for reliability, making the effective data size even smaller and limiting adaptation potential. To address this issue, We propose Feature Augmentation based Test-time Adaptation (FATA), a simple method that fully utilizes the limited amount of input data through feature augmentation. FATA employs Normalization Perturbation to augment features and adapts the model using the FATA loss, which makes the outputs of the augmented and original features similar. FATA is model-agnostic and can be seamlessly integrated into existing models without altering the model architecture. We demonstrate the effectiveness of FATA on various models and scenarios on ImageNet-C and Office-Home, validating its superiority in diverse real-world conditions. Code is available at https://github.com/RangeWING/FATA.

Human activity recognition is a key challenge in pervasive computing and its solutions have been presented based on various disciplines. Specifically, for human activity recognition in smart spaces without privacy and accessibility issues, data streams generated by deployed ambient sensors are leveraged. In this paper, we focus on group activities by which a group of users perform a collaborative task without user identification and propose an efficient group activity recognition scheme that extracts causality patterns from ambient sensor event sequences, to support as good recognition accuracy as the state-of-the-art models with missing or false data tolerance. To filter out irrelevant noise events from a given data stream, a set of rules is leveraged to highlight causally related events. Then, a pattern-tree algorithm extracts frequent causal patterns by means of a growing tree structure. Based on the extracted patterns, a weighted sum-based pattern-matching algorithm computes the likelihood of stored group activities to the given test event sequence using matched event pattern counts for group activity recognition. We evaluate the proposed scheme using the data collected from real-world testbed and open datasets where users perform their tasks on a daily basis. Experiment results show that the proposed scheme performs higher recognition accuracy and is tolerant to missing or false data with a smaller amount of runtime overhead than the existing schemes.

Traffic prediction is one of the key elements to ensure the safety and convenience of citizens. Existing traffic prediction models primarily focus on deep learning architectures to capture spatial and temporal correlation. They often overlook the underlying nature of traffic. Specifically, the sensor networks in most traffic datasets do not accurately represent the actual road network exploited by vehicles, failing to provide insights into the traffic patterns in urban activities. To overcome these limitations, we propose an improved traffic prediction method based on graph convolution deep learning algorithms. We leverage human activity frequency data from National Household Travel Survey to enhance the inference capability of a causal relationship between activity and traffic patterns. Despite making minimal modifications to the conventional graph convolutional recurrent networks and graph convolutional transformer architectures, our approach achieves state-of-the-art performance without introducing excessive computational overhead.

Cross-camera real-time object tracking is one of the important, yet challenging applications of video analytics in edge computing environments. To provide accurate and efficient real-time tracking, a tracking target's future movements need to be predicted. Particularly, the destination camera and travel time of the target object are to be identified so that tracking duties can be handover-ed seamlessly. In this paper, we propose a collaborative cross-camera tracking system, called PreActo, with two key features: (1) ResNet-based trajectory learning to exploit the rich spatio-temporal information embedded within objects' moving patterns, which has not been utilized by the existing literature, and (2) collaboration between the edge server and the edge device for real-time trajectory prediction and tracking handover. To prove the validity of our proposed system, we evaluate PreActo on a video dataset leveraging real-world trajectories. Evaluation results show that the proposed system reduces up to$7\times$the number of processed frames for handover, with$2\times$lower latency while providing$1.5\times$tracking precision improvement compared to the state-of-the-art.

Reference-based video super-resolution (RefVSR) is a promising domain of super-resolution that recovers high-frequency textures of a video using reference video. The multiple cameras with different focal lengths in mobile devices aid recent works in RefVSR, which aim to super-resolve a low-resolution ultra-wide video by utilizing wide-angle videos. Previous works in RefVSR used all reference frames of a Ref video at each time step for the super-resolution of low-resolution videos. However, computation on higher-resolution images increases the runtime and memory consumption, hence hinders the practical application of RefVSR. To solve this problem, we propose an Efficient Reference-based Video Super-Resolution (ERVSR) that exploits a single reference frame to super-resolve whole low-resolution video frames. We introduce an attention-based feature align module and an aggregation upsampling module that attends LR features using the correlation between the reference and LR frames. The proposed ERVSR achieves 12× faster speed, 1/4 memory consumption than previous state-of-the-art RefVSR networks, and competitive performance on the RealMCVSR dataset while using a single reference image.

Activity Segmentation, dividing a continuous sensor stream into a set of activity segments, is a crucial pre-process in Human Activity Recognition (HAR) and it is required to be done in real-time for real-world smart services. Existing single-user activity segmentation schemes fail to correctly detect transition points due to concurrent and overlapping events from multiple users in case of Multi-user Collaborative Activity Recognition (MCAR). In this paper, we propose a novel scheme for activity segmentation for MCAR that expresses complex events and the correlations between them. For this, the proposed scheme first creates an event stream from a sensor stream and defines event sets in terms of time windows. For each time window, two types of correlations for every event pair are calculated: duration correlation and history correlation. After calculating event correlation, the change score of a time window is measured by comparing the calculated correlation values with those of the preceding windows. Then, the proposed scheme elects as an activity transition point a time window whose change score exceeds the transition threshold. We evaluate the proposed method on two multi-user collaborative activity datasets and experiment results show that the proposed scheme achieves better segmentation performance than existing approaches.

The growing number of smart vehicles makes it possible to envision a crowdsensing service where vehicles can share video data of their surroundings for seeking out traffic conditions and car accidents ahead. However, the service may need to deal with situations like malicious vehicles propagating false information to divert other vehicles to arrive at destinations earlier or lead them to dangerous locations. This article proposes a context-aware trust estimation scheme that can allow roadside units in a vehicular edge network to provide real-time crowdsensing services in a reliable manner by selectively using information from trustworthy sources. Our proposed scheme is novel in that its trust estimation does not require any prior knowledge of vehicles on roads but quickly obtains the accurate trust value of each vehicle by leveraging transfer learning. and its Q-learning-based dynamic adjustment scheme autonomously estimates trust levels of oncoming vehicles with the aim of detecting malicious vehicles and accordingly filtering out untrustworthy input from them. Based on an extensive simulation study, we prove that the proposed scheme outperforms existing ones in terms of malicious vehicle detection accuracy.

Incompatible updates of IoT systems and protocols give rise to interoperability problems. Even though various protocol adaptation and unknown protocol inference schemes have been proposed, they either do not work where the updated protocol specifications are not given or suffer from inefficiency issues. In this work, we present an efficient protocol inference scheme for incompatible updates in IoT environments. The scheme refines an active automata learning algorithm, L*, by incorporating a knowledge base of the legacy protocol behavior into its membership query selection procedure for updated protocol behavior inference. It also infers protocol syntax based on our previous work that computes the most probable message field updates and adapts the legacy protocol message accordingly. We evaluate the proposed scheme with two case studies with the most popular IoT protocols and prove that it infers updated protocols efficiently while improving the L* algorithm’s performance for resolving the incompatibility.

Coverage maximization is an important issue of Mobile wireless sensor networks (M-WSN). Especially for visual sensors like video camera which have a specific sensing direction and range, obstacles and the position of the sensors should also be considered. In this paper, we propose an efficient coverage maximization method for mobile video camera networks by leveraging DQN, a deep reinforcement learning algorithm. Evaluation results show that the proposed method can cover up to 4.93% better than existing ones.

Commercial gentrification refers to the replacement of low-value businesses like small local stores into high-value businesses like boutiques and chain stores. A handful of research efforts have been made to identify gentrification and their change by leveraging social media. However, their approaches lack in inferring how much commercial gentrification is developed in a target area and how long it has taken for the area to get to that phase. In this paper, we propose a novel scheme to estimate the commercial gentrification status of a target area and its development in terms of time and geographic dispersion using Instagram data. For this, we define our commercial gentrification phase criteria based on the conceptual model from the urban study. Then, we extract social features from both images and texts of Instagram posts, and leverage regression models to infer the commercial gentrification phase of a target area at the monthly timestamp. We also measure how geographical dispersion of geo-tagged Instagram posts matches the boutiques, which is the physical variable that has the strongest correlation with the commercial gentrification. Evaluation results show that our method yields a good quality of estimation compared to the ground truth. This assures that our method could be a meaningful tool for urban planners and policymakers to investigate and manage commercial gentrification.

This work proposes a context-aware trust estimation scheme that can allow roadside units in a vehicular edge network to provide real-time crowdsensing services in a reliable manner by selectively using information from trustworthy sources. Our proposed scheme is novel in that its trust estimation does not require any prior knowledge towards vehicles on roads but quickly obtains an accurate trust value of each vehicle. To that end, we particularly leverage the concept of I-sharing which removes a cold-start problem during the system bootstrapping period. Based on an extensive simulation study, we prove that the proposed scheme outperforms its competitive counterpart and baseline models in terms of trust bias and malicious vehicle detection accuracy.

There have been several efforts on preference learning in a smart space by means of multi-agent collaborations. Each agent captures a user action or handles part of learning but decision makings are done in a centralized manner. This makes it difficult for a smart space to deal with learning complexity due to the increase and reconfiguration of smart devices. While the complexity is relieved by articulating the learning space, it is not flexible because the articulation procedure needs to be resumed whenever a smart space reconfiguration occurs. In this paper, we propose a distributed multi-agent preference learning architecture which allows a group of physically separate agents to collaborate with each other for learning a user's task preference efficiently in an IoT enriched smart space. For this, the proposed scheme provides four key features: ontology-based knowledge structure for task-driven agent collaboration, knowledge exchange protocol for task-aware causality among agents, Q-learners for observing and learning from user behaviors, and negotiation and acknowledgement protocol for preventing agents from performing disorganized actions. Evaluation results show that the proposed scheme allows smart device agents to learn user preferences in a fully distributed way and outperforms existing approaches in terms of learning speed and system overhead.

Multi‐constrained routing is a key driver to support quality‐of‐service (QoS) for real‐time multimedia applications in wireless mesh networks (WMNs). Due to the difficulty of applying strict admission control into a public WMN, it is inevitable to accommodate multiple application flows with different QoS requirements exceeding the capacity of a certain link shared by multiple flows. However, existing multi‐constrained routing protocols under such an environment find the QoS degradation based on end‐to‐end path quality probing and trigger flooding‐based route discovery from a scratch for resolving the QoS degradation, which incurs a longer recovery time and much routing overhead. In this paper, we propose a novel multi‐constrained routing protocol for WMNs that finds problematic links that may affect QoS degradation to end‐to‐end paths and replaces them with a detour path using a local repair principle. We model congestion threshold estimation for finding problematic links and design algorithms for quickly finding detour paths and selecting an optimal path by minimizing the negative effect on existing flows nearby the detour path. Simulation results show that the proposed routing protocol achieves up to 19.6% more goodput of live video streaming applications with up to 33% reduced routing overhead compared with an existing work.

As IoT devices become prevalent in our daily lives, estimation of their trustworthiness plays an important role for privacy protection complementary to security solutions. Existing trust estimation solutions are based on SIoT whose full social network is not likely to be observable in a public space or do not reflect situation-dependent dynamism of trust. In this paper, we propose a new trust estimation scheme that computes a user's trust value of an IoT device combining both the personal trust from the interaction history and non-personal stereotypical reputation from the general public.

Smartphone notifications provide application-specific information in real-time, but could distract users from in-person social interactions when delivered at inopportune moments. We explore breakpoint-based notification management, in which the smartphone defers notifications until an opportune moment. With a video survey where participants selected appropriate moments for notifications from a video-recorded social interaction, we identify four breakpoint types: long silence, a user leaving the table, others using smartphones, and a user left alone. We introduce a Social Context-Aware smartphone Notification system, SCAN, that uses build-in sensors to detect social context and identifies breakpoints to defer smartphone notifications until a breakpoint. We conducted a controlled study with ten friend groups who had SCAN installed on their smartphones while dining at a restaurant. Results show that SCAN accurately detects breakpoints (precision=92.0%, recall=82.5%), and reduces notification interruptions by 54.1%. Most participants reported that SCAN helped them to focus better on in-person social interaction and found selected breakpoints appropriate.

The proliferation of the Internet into every household has provided more opportunities for residents to become closer to each other than before. However, solid structural barrier is raised and social relationships within such neighborhoods are weak compared to those in traditional towns. Accordingly, activating communities and ultimately enhancing a sense of community through constructive participation and communal sharing of labor among residents has currently emerged as a challenging issue in a contemporary housing complex. In an effort to activate those communities, a notion of smart community is presented in which multiple smart homes are equipped with Internet of Things and interconnected with each other. Beyond the unadorned smart community composed by physical proximity, it is essential to discover a human-centric community that achieves communal benefits and enables residents to maximize individual economic gain by leveraging collective intelligence. In this article, we present a multi-dimensional smart community discovery scheme that enables householders to find human-centric community considering multi-dimensional factors in terms of physical, social, and economical aspects. We conduct experiments with 30 real households by applying a community-based energy saving scenario. Experiment results show that the proposed scheme performs better when compared to the physical proximity-based one in energy consumption and user satisfaction.

We present a generic framework for group activity recognition using simple non-obtrusive sensors. The proposed scheme is based on that group activity patterns can be derived from mining interval-based relationships between users' temporally overlapped actions. We leverage a hybrid architecture of probabilistic and logic knowledge that can capture the essence of the temporal dependencies, represented as a set of weighed rules. It can also learn different weights for common rules between similar group activities, which share most of sensor events and events order. The evaluation results show that our scheme outperforms the sequential baseline model, a mixture of Gaussian Hidden Markov Models.

Power management of Wi-Fi interfaces can greatly impact the battery life time of a smart device. Thus, many commercial devices utilize the Power Save Mode - Adaptive (PSM-A) mechanism for energy saving by switching wireless radio between high and low power. Tail time is introduced in PSM-A to put the radio in high power state until the time expires, allowing near future packets to arrive without much delay. However, this fixed tail time may result in a considerable amount of energy drainage under various types of traffic. This paper proposes an adaptive tail time adjustment scheme, a simple yet efficient way to save energy wastage by adaptively resizing the tail time according to prediction of data packet arrival times. The simulation results show up to 28.4% energy savings and reduces packet delivery delay compared to existing schemes.

This paper proposes a novel priority-aware packet fragmentation extension to high throughput WLANs such as IEEE 802.11n for streaming of H264/AVC encoded videos. Unlike existing fragmentation schemes, the proposed scheme fragments IP packets based on the priority of video packets and the characteristics of MPEG-2 TS, where the original IP packet is fragmented into smaller IP packets containing fewer TS packets and prioritizes individual TS packets, allocated to an appropriate priority queue. The proposed scheme is evaluated on a testbed with various network congestion levels and channel conditions. The results show that the proposed scheme achieves higher quality of the streaming video in terms of PSNR than existing schemes as the network congestion level and the bit error rate increase.

The accuracy of wireless network packet simulation critically depends on the quality of wireless channel models. Path loss is the stationary component of the channel model affected by the shadowing in the environment. Existing path loss models are inaccurate, require excessive measurement or computational overhead, and/or often cannot be made to represent a given environment. This paper contributes a flexible path loss model that uses a novel approach for spatially coherent interpolation from available nearby channels to allow accurate and efficient modeling of path loss. We show that the proposed model, called Double Regression (DR), generates a correlated space, allowing both the sender and the receiver to move without abrupt change in path loss. Combining DR with a traditional temporal fading model, such as Rayleigh fading, provides an accurate and efficient channel model that we integrate with the NS-2 simulator. We use measurements to validate the accuracy of the model for a number of scenarios. We also show that there is substantial impact on simulation behavior when path loss is modeled accurately. Finally, we show that unlike statistical models, DR can make a simulation representative of a given environment by using a small number of seeding measurements. Thus, DR provides a cost-effective alternative to ray tracing or detailed site surveys.

Since the Web of Things (WoT) term was first proposed, there has a big trend in IT vendors providing users with various services through their smart products. To enable users to discover and leverage these services, SDPs play an important role. However, so many variations of SDPs have been introduced it has caused a heterogeneity issue. Including standardization, many solutions have been proposed, but they require too much overhead or have practicality issues. In this paper, we propose a system composed of fundamental building blocks, including a knowledge base and probing packets to address this issue. We evaluate the performance of our system by conducting real world experiments using smart object services in a local network. The experiment results show that Sherlock-SD identifies up to 92% of target services correctly only with 3 probing packets out of 6 in the best case, without any of the overhead that existing solutions impose. In terms of resource consumption overhead, compared to 4 SDPs enumeration, Sherlock-SD requires 67.6% of the memory and consumes 30.7% of power.

This paper presents a distributed smart home middleware where each appliance is able to learn user behavior and customize their actions by themselves as well as cooperate with other appliances through a more light-weight smart home gateway. As the key components, we present a knowledge base which describes common- and appliance-specific concepts in a smart home domain, and design libraries for smart appliances and a smart home gateway. We implement the proposed middleware on our testbed and conduct evaluations. The result shows that our scheme reduces the interaction time and the runtime memory allocation.

As many-to-one traffic patterns prevail in data center networks, TCP flows often suffer from severe unfairness in sharing bottleneck bandwidth, which is known as the TCP outcast problem. The cause of the TCP outcast problem is the bursty packet losses by a drop-tail queue that triggers TCP timeouts and leads to decreasing the congestion window. This paper proposes TCPRand, a transport layer solution to TCP outcast. The main idea of TCPRand is the randomization of TCP payload size, which breaks synchronized packet arrivals between flows from different input ports. We investigate how TCPRand reduces consecutive packet drops and demonstrate various benefits of TCPRand with extensive experiments and ns-3 simulation. Our evaluation results show that TCPRand guarantees the superior enhancement of TCP fairness with negligible overheads in all of our test cases.

The aim of this research is to develop a quantitative usability evaluation method (UEM) for elderly drivers, which has different weight values on each factor concerning physical and cognitive context of elderly drivers. An analysis of the relationship between universal design guidelines for elderly drivers and usability principles was conducted by using the quality function deployment method. In addition, developmental priorities are derived from analysis results of difficulty in achieving performance improvement, max relationship values, and relative weight. Furthermore, n and positive relationships among the universal design guidelines are defined by means of relationship analysis. Combining these results, a quantitative evaluation guideline for elderly drivers is derived, and based on the context and developmental goals of the developer, selective design is facilitated. The proposed UEM is compared with existing UEM in terms of thoroughness, validity, and effectiveness.

In this paper, we present an opportunistic service recommendation scheme in a place with smart objects to provide users with appropriate composite services to accomplish their task goals. The proposed scheme infers possible tasks from interactions that people have experienced with smart objects including other places that are similar to a given place. The proposed scheme then evaluates the candidate tasks by measuring quality satisfaction of each task against given QoS constraints. We test the proposed scheme on our three smart testbeds where various smart objects equipped. Experiment results show that the proposed scheme finds potential tasks in the testbeds in a reasonable time.

In mobile ad hoc networks (MANETs), a layered approach to networks makes a routing protocol to be responsible on discovery of a path between two nodes. However, this approach has fundamental limitations in diversifying the strategies of routing in a network when various types of applications require different routing support. In this paper, we propose an application-driven path discovery mechanism. In the proposed architecture, a routing packet controller resides in each MANET node and each application can control path discovery process by installing rules in the routing packet controller. Several examples are described to show the feasibility of our approach.

Cross layer approach is a promising technology to satisfy applications' fastidious requirements and maximize the wireless channel usage in wireless ad hoc networks. Existing cross layer architecture either adopts a specific MAC for an application type or does not change the MAC behavior because the MAC is commonly used for all types of applications in a node. MAC virtualization is an appropriate method to associate the MAC behavior with a specific application's requirement without impacting on other types of application. In this paper, we propose a partial MAC virtualization to support cross layer design in wireless ad hoc networks. In the proposed scheme, a MAC scheme is divided into a set of controllable condition variables and common MAC executor. A MAC control manager generates a virtual MAC with a set of controllable condition variables and related state variables. This paper also proposes an interface through which an application or an agent of the application can control the MAC for its own purpose, which allows a cross-layer mechanism for each application in a node. We show the feasibility of the proposed scheme by an example and experiments.

The accuracy of wireless network packet simulation critically depends on the quality of the wireless channel models. These models directly affect the fundamental network characteristics, such as link quality, transmission range, and capture effect, as well as their dynamic variation in time and space. Path loss is the stationary component of the channel model affected by the shadowing in the environment. Existing path loss models are inaccurate, require very high measurement or computational overhead, and/or often cannot be made to represent a given environment. The paper contributes a flexible path loss model that uses a novel approach for spatially coherent interpolation from available nearby channels to allow accurate and efficient modeling of path loss. We show that the proposed model, called Double Regression (DR), generates a correlated space, allowing both the sender and the receiver to move without abrupt change in path loss. Combining DR with a traditional temporal fading model, such as Rayleigh fading, provides an accurate and efficient channel model that we integrate with the NS-2 simulator. We use measurements to validate the accuracy of the model for a number of scenarios. We also show that there is substantial impact on simulation behavior (e.g., up to 600% difference in throughput for simple scenarios) when path loss is modeled accurately.

Wireless mesh networks (WMNs) are promising networks in the future Internet environments due to their self-configuration and ease of deployment. Although streaming and real time applications are possible using current mobile devices, these applications still suffer from low quality. Because link failures are not reported in congested situations, the existing routing protocols do not capture these deteriorations. An extension of the AODV-MR that is adaptive to changes in link quality is proposed in order to detect deteriorated situations and to recover the path in WMNs. The path is recovered proactively by route discovery if the quality difference of the current link and other available links exceeds the threshold which is determined by the network status. The simulation results demonstrate that the proposed method is effective in multi-radio WMNs and achieve higher throughput, lower average jitter, and lower end-to-end delay.

By leveraging location data in online social networks, Location-based Social Networks (LBSNs) can support diverse human activities such as tourism. Different applications aim to aid tourists and provide better experience in their travels by matching co-located users based on what they have in common. However, users with little in common but with potential to help each other given the context and place could not be matched. In this paper we introduce traMSNet, a LBSN that implements a matching algorithm considering homophily, as well as users complementary skills in a touristic location. Our idea is validated with a survey that asked potential travelers about their needs when looking for a travel partner. Moreover, we present a matching algorithm that is evaluated it with real tourists. The evaluation shows that considering complementarity when matching individuals is preferred by users. Therefore, by only considering similarities, important issues are left aside.

In this paper, we propose a new class of message routing scheme for a distributed network which can scale up to the needs of a very large number of mobile users. We leverage the work in the GeoKad paper which uses the Distributed Geographic Table, allowing node retrieval from resources that are close to any given region. The paper addresses the problem regarding efficient message passing and peer misses. We propose a new routing scheme that addresses this kind of limitation. In fact, the problem can be solved in two ways, either by preserving the state of peer and further use a prediction algorithm or by better message routing. The proposed scheme is based on swarm intelligence (SI) and its attribute of exploit and explore. We leverage the statistical properties of time-varying network connectivity for opportunistic message forwarding. Considering the present 3G or Wi-Fi environment this approach can also help in optimizing energy as it can prevent query over flooding which is mostly encountered in such network scenarios.

Uncovering user interest plays an important role to develop personalized systems in various fields including the Web and pervasive computing. In particular, online social networks (OSNs) are being spotlighted as the means to understand users' social behavior out of abundant online social information. In this paper, we explore a computational method of inferring user interest in Facebook by combining the degree of familiarity and topic similarity with social neighbors based on social correlation phenomenon. By conducting a question-naire survey, we demonstrate that our proposed method increases the accuracy of inference by 12.4% compared to existing methods which do not consider the latent topic structure implied in social contents.

Recent advancement in smart phones enables users to collaborate in mobile environments. Their diverse communication interfaces allow them to interact with nearby devices, enabling the sharing of computing capabilities and resources among collocated users. Given the personal nature of smart phones, the selection of trustworthy users must be incorporated to enable this sharing process. In this work, a context-aware trust decision scheme for mobile ad-hoc environments is presented. This scheme is decentralized as uses information acquired by mobile phones, while leveraging social theories to reflect real-life aspects properly. The proposed scheme is validated against required properties of trust decision schemes for mobile ad-hoc environments.

Map services help people find places they wish to visit. Users input the target address and the system returns that location on a map, with additional information such as nearby points of interest, directions to the location, etc. However, these applications mainly focus on general knowledge, failing to provide personalized information that may help users find/identify a place faster. In this paper, we present a map application based on previous user knowledge. We represent two types of knowledge - spatial and social - in the form of a social-location ontology. This ontology is then used to present a destination based on places familiar to the user. By showing an unknown location in terms of a familiar place, we allow the user to find the target location with much briefer instructions. This is especially useful on mobile devices, in which the limited screen size restricts the amount of information that can be displayed. In order to adapt to new knowledge, the ontology evolves as the user visits new places and meets new people. Our evaluation shows that users prefer to have concrete information related to a destination for easier understanding.

Context data are updated frequently due to the dynamic changes of the various sensor values and the situations of application entities. Without a proper management, the stored contexts will become different from those of the real-world. Those invalid contexts will cause context inconsistency problems and thus should be eliminated at the right time and in an appropriate manner. In this paper, we propose a context inconsistency management scheme based on context elimination rules that describe the semantics of context invalidity to solve context inconsistency problems. The proposed rule-based scheme will enable users to easily specify elimination conditions for inconsistent contexts. Our performance evaluation shows that the rule processing overhead is compensated for by virtue of the well-maintained repository of the stored contexts.

A social-based routing protocol for opportunistic networks considers the direct delivery as forwarding metrics. By ignoring the indirect delivery through intermediate nodes, it misses chances to find paths that are better in terms of delivery ratio and time. To overcome this limitation, we propose to incorporate transitivity, which considers the indirect delivery through intermediate nodes, as one of the forwarding metrics. We also found that some message forwards do not improve the delivery performance. To reduce the number of these useless forwards, the proposed scheme forwards messages to an encountered node when the increase of total utility value is greater than a threshold. Using a simulator with real world trace data sets, we compare the proposed scheme with the existing protocols, epidemic routing and SimBetTS. Compared with SimBetTS, the proposed scheme increases delivery ratio by 1.5 percent and decreases delay time by 2 percent while reducing overhead by 30 percent.

In this paper, we propose QoS-constrained peer selection and nondisruptive rate adaptation schemes with considering shared bottleneck problem to provide a stable peer-to-peer live video streaming service during vertical handoff. In the proposed scheme, a QoS-constrained peer is selected by estimating bandwidth with active probing, and a reception rate converges to a stable state without any QoS degradation with help of multi-path handoff. The shared bottleneck problem is addressed by adjusting the estimated bandwidth and yielding the redundant bandwidth usage.

Users and developers require some way to identify by human readable names the many resources that comprise a ubiquitous computing environment. In addition, naming in such environments should work well despite ad hoc situations, and it should also provide context-based naming. This paper describes the Non- anchored Unified Naming (NUN) system, which satisfies these requirements. It is based on relative naming among resources and provides the ability to name arbitrary types of resources. By having resources themselves take part in naming, resources are able to contribute their specialized knowledge into the name resolution process, making context-dependent mapping of names to resources possible.

A key feature of ubiquitous computing is service continuity which allows a user to transparently continue his task regardless of his movement. For service continuity, the underlying system needs to not only discover a service satisfying a user's request, but also provide an interface differences resolution scheme if the interface of the service found is not the same as that of the service requested. For resolving interface mismatches, one of solutions is to use an interface adapter. The most serious problem in the interface adapter-based approach is the overhead of adapter generation. There are many research efforts about adapter generation load reduction and this paper focuses on an adapter chaining scheme to reduce the number of necessary adapters among different service interfaces. We propose a construction-time adaptation loss evaluation scheme and an adapter chain construction algorithm, which finds an adapter chain with minimal adaptation loss.

One of the most evident characteristics of an ad hoc ubiquitous computing environment is the dynamism of connections. Due to the mobility of nodes, paths between communicating peers change over time and hence should be discovered continually. Applications also dynamically discover available services and replace connections to provide contextually more appropriate services. In such an environment, the reduction of discovery cost and the reflection of context are main design factors. In order to avoid discovery cost being paid by both application and routing layer and to delay a service binding until forwarding time, we propose a shim layer called shared generic routing (or SGR) that discovers paths by accommodating application layer labels as well as IP addresses as routing targets and manages forwarding states. On top of SGR, IP and application layer routing can be efficiently implemented. Through simulation, we show that SGR can effectively remove duplicated discovery efforts.

Advancement in mobile devices makes possible to add more features to them while reducing the size. If we also consider the fact that they posses advance communications features, we can convert these devices into the central manager of users’ activities. One feature in this line is the ability to control the devices surrounding the user. Problems arise when the devices belong to a shared space. In this case we also need to consider interactions and conflicts between users. In this research we present a multi-user ad-hoc resource manager for smart spaces, which allow us to control devices present in the area, and avoid conflicts between users.

Abstract With the expansion of the internet and its bandwidth, distributed virtual environment (DVE) applications have become more prevalent. In DVE applications, users frequently crowd in a specific place, and a key aspect to consider is how to provide interactive performance for users. However, existing approaches using multicast require users to receive uninteresting messages. Even though recent works have addressed fine‐grained filtering, they still incur other drawbacks in terms of assigning lots of multicast addresses or handling overhead of multicast groups. This makes the system less scalable as the number of users increases. In this paper, we propose a new scalable filtering scheme that reduces not only the number of messages during interaction in a region and among neighboring regions, but also the number of multicast addresses without significant computational overhead. Interest management in a region dynamically creates groups of users with the same interests. While members communicate with each other with high fidelity, a representative sends information to non‐members with low frequency. For interaction among neighboring regions, we propose a sub‐region concept to select only a subset of users from the neighboring regions based on proximity, the distribution of the users' locations, and the viewing direction of a user. Copyright © 2007 John Wiley & Sons, Ltd.

In this paper, we suggest a efficient multiple-forwarder based file distribution scheme working over a tree-based overlay multicast. The proposed scheme takes advantage of a fact that each member joins the session with time interval. In a tree, the set of data delivered from a parent to each member since a member has joined until the next member joins is not overlapped. By using this property, in the proposed scheme, each forwarder distributes data to its succeeding members regardless of the underlying tree. The effectiveness of the proposed scheme is measured using numerical analysis and the performance evaluation on planet-lab shows that the proposed scheme enhances the throughput while maintaining data duplication ratio and control overhead significantly less compared to an existing scheme, Bullet.

In this paper, we propose a context-conflict management scheme for group-aware ubiquitous computing environments that allows context-aware applications to serve multiple users without distracting and sacrificing each other. The proposed scheme can detect semantic conflicts without explicit descriptions of the conflicts between different applications. Conflicts are resolved by a dynamically generated adaptation policy based on the weight value of user preference on each service. Individual context-aware applications can be reused without modification since middleware components for the context-conflict management take charge of detecting and resolving conflicts. Thus, the complicated context-conflict situation becomes more transparent to application programmers.

There have been many research efforts to enhance the throughput of file distribution using overlay multicast. The tree-based approach has limitations in enhancing throughput due to its single forwarder architecture. Recent approaches introduce multiple packet forwarders other than the parent in the delivery tree for throughput enhancement. However, the overhead of control packet exchanges is required to appoint proper forwarders and to avoid duplicate packet delivery. In this paper, we suggest a throughput enhancement scheme for a tree-based overlay multicast scheme with multiple forwarders. It exploits a forward error correction scheme and the fact that the members join the session with time interval. Thus, each set of packets that are delivered from a parent to a member after the member joins and before a succeeding member joins is not overlapped. By deploying a dedicated sender to manage the membership information and packet reception state, the overhead of control packet exchanges is alleviated significantly. The performance evaluation shows that the proposed scheme reduces file delivery time with low control overheads and small packet duplication ratio, compared with existing scheme, Bullet.

We present active surroundings, a group-aware middleware infrastructure for embedded application systems where entities (devices or services) actively respond to user actions and help users to perform their jobs with no or minimal involvement of users. Our system focuses on two key issues: transparent application reconfiguration and group-context awareness.

Abstract Even though tree‐based reliable multicast protocols are known to be most scalable for one‐to‐many sessions, there is still an open question as to whether these protocols are also scalable for many‐to‐many sessions. In this paper, we analyze and compare two promising multicast protocols—the receiver‐initiated protocol with NACK suppression and the tree‐based protocol—using a new spatial loss model. The proposed model considers the correlation of packet loss events for more realistic analysis unlike the previous work. The analysis results show that the tree‐based protocol achieves much higher throughput than the receiver‐initiated protocol for a many‐to‐many session as the number of participants in the session becomes larger. Copyright © 2004 John Wiley & Sons, Ltd.

Congestion control is a key task in reliable multicast along with error control. However, existing tree-based congestion control schemes such as MTCP[11] and TRAMCC[3] are designed for one-to-many reliable multicast and have some drawbacks when they are used for many-to-many reliable multicast. We propose an efficient congestion control mechanism, TMRCC, for tree-based many-to-many reliable multicast protocols. The proposed scheme is based on the congestion windowing mechanism and a rate controller is used in addition. The feedback for error recovery is exploited for congestion control as well to minimize the overhead at the receivers. The ACK timer and the NACK timers are set dynamically reflecting the network condition changes. The rate regulation algorithm in the proposed scheme is designed to help the flows sharing the same link to achieve the fair share quickly. The performance of the proposed scheme is evaluated using ns-2 [2]. The simulation results show that the proposed scheme outperforms TRAMCC in terms of intra-session fairness and shows good level of responsiveness, TCP-friendliness, and scalability.

This paper proposes and evaluates a scalable dynamic load distribution scheme for multi-server distributed virtual environment systems, where users are highly skewed rather than uniformly distributed over a virtual environment. In the proposed scheme, an overloaded server initiating load distribution selects a set of servers to be involved in load distribution by dynamically adapting to the workload status of other servers, unlike the existing approaches. Upon completion of server selection, the intiating server repartitions the regions dedicated to the involved servers using a graph partitioning algorithm so that all the involved servers have the roughly equal workload. The involved servers then migrate their workloads with each other in a peer-to-peer manner according to the result of repartitioning. We present and analyze the simulation results that compare the performance of the proposed scheme with that of the existing schemes.

Group communication is a useful mechanism guaranteeing consistency among replicated objects. The existing approaches do not allow transparent plug-in of group communication protocols into CORBA. They either require modification of CORBA or OS, or provide no room for incorporating group communication transport protocols into CORBA. We thus propose a generic group communication framework that allows transparent plug-in of various group communication protocols with no modification of existing CORBA. We extend the open communications interface (OCI) to support interoperability, reusability of existing group communication, and independency on ORB and OS. We also define the group communication inter-ORB protocol (GCIOP) as a group communication instantiation of the general inter-ORB protocol (GIOP) that encapsulates underlying group communication protocols. The proposed scheme can be exploited for fault-tolerant CORBA (FT CORBA).

Tree-based reliable multicast protocols are known to be most scalable for one-to-many sessions. However, there is still an open question whether these protocols are also scalable for many-to-many sessions. We categorize existing tree-based protocols into four classes according to the feedback/retransmission policies employed, and analyze and compare them on the basis of a proposed spatial loss model. This model considers the correlation of packet loss events for more realistic analysis, unlike previous work. The analysis reveals that, in many-to-many cases, a tree-based protocol with unicast NACK, unicast retransmission and periodic polling is most scalable with respect to throughput for practical session sizes. The protocol is also simple to implement.

We present the design, implementation, and performance analysis of group-aided multicast (GAM), a scalable many-to-many reliable multicast transport protocol. GAM achieves high quality ACK trees while keeping the tree maintenance overhead reasonably low in the presence of dynamic group membership and route changes. It is supported by a group configuration mechanism organizing the members in a multicast session into multiple small groups and a tree configuration mechanism maintaining logical trees according to the underlying multicast routing trees. With the two mechanisms, GAM builds a two-layer hierarchy of multi-level logical trees from which high-quality per-source ACK trees are generated. Simulation results show that the GAM protocol is more scalable than a NACK suppression protocol in terms of processing time for request/repair messages and recovery latency.

Object replication is a technique to enhance fault tolerance and high availability. Group communication is a useful mechanism guaranteeing the consistency among replicated objects. We propose a generic group communication framework that allows transparent plug-in of various group communication protocols with no modification of existing CORBA. For this, we extend the Open Communications Interface (OCI) to support interoperability, reusability of existing group communication, and independency on ORB and OS. The proposed scheme is composed of group membership, group IOR, and group multicast. Experimental results show that group object invocation using the proposed scheme performs substantially better than group object invocations using IIOP as the number of members in a group increases.

Shared sense of a virtual world is often enhanced by replicating the information at each user's site since replication provides acceptable interactive performance, especially when users are geographically distributed over large networks like the Internet. However, multiple concurrent updates may lead to inconsistent views among replicas. Therefore concurrency control is a key factor to maintaining a consistent state among replicas. We proposed a scalable prediction-based scheme in which an ownership request is multicasted to only the users surrounding a target entity. In our previous work, we assumed that all the users navigate a virtual world with a single speed. It, however, is quite common in a networked virtual environment like a network game that users are allowed to change their navigation speed as they interact with a virtual world for adding more realism. This paper proposes an enhancement to support users with various speeds. The enhanced scheme allows as many Entity Radii as the number of different speed and allocates a separate queue for users of each speed. Each queue is examined in parallel to predict the next owner candidate and among the selected candidates is chosen the final candidate, which has a minimum predicted collision time. It contributes to the timely advanced transfer of ownership by using appropriate Entity Radius based on a user's speed, fair granting of ownership by reducing the interference between users with different speed and latency, and high prediction accuracy by reducing the redundant ownership transfer.

For supporting real-time interaction in distributed virtual environments (DVEs), it is common to replicate virtual world data at clients from the server. For efficient replication, two schemes are used together in general - prioritized transfer of objects and a caching and prefetching technique. Existing caching and prefetching approaches for DVEs exploit spatial relationship based on distances between a user and objects. However, spatial relationship fails to determine which types of objects are more important to an individual user, not reflecting user's interests. We propose a scalable data management scheme using user-based caching and prefetching exploiting the object's access priority generated from spatial distance and individual user's interest in objects in DVEs. We also further improve the cache hit rate by incorporating user's navigation behavior into the spatial relationship between a user and the objects in the cache. By combining the interest score and popularity score of an object with the spatial relationship, we improve the performance of caching and prefetching since the interaction locality between the user and objects are reflected in addition to spatial locality. The simulation results show that the proposed scheme outperforms the hit rate of existing caching and prefetching by 10% on average when the cache size is set to basic cache size, the size of expected number of objects included in the user's viewing range.

This paper proposes a Barrier Tree for Meshes (BTM) to minimize the barrier synchronization latency for two-dimensional (2D) meshes. The proposed BTM scheme has two distinguishing features. First, the synchronization tree is 4-ary. The synchronization latency of the BTM scheme is asymptotically /spl theta/(log/sub 4/ n), while that of the fastest scheme reported in the literature is bounded between /spl Omega/(log/sub 3/ n) and /spl theta/(n/sup 1/2/), where n is the number of member nodes. Second, nonmember nodes are neither involved in the construction of a BTM nor actively participate in the synchronization operations, which avoids interference among different process groups during synchronization. This not only results in low setup overhead, but also reduces the synchronization latency. The low setup overhead is particularly effective for the dynamic process model provided in MPI-2. Extensive simulation study shows that, for up to 64/spl times/64 meshes, the BTM scheme results in about 40/spl sim/70 percent shorter synchronization latency and is more scalable than conventional schemes.

VLSI-based processor arrays have been widely used for computation intensive applications such as matrix and graph algorithms. Algorithm-based fault tolerance designs employing various encoding/decoding schemes have been proposed for such systems to effectively tolerate operation time fault. In this paper, we propose an efficient algorithm-based fault tolerance design using the weighted data-check relationship, where the checks are obtained from the weighted data. The relationship is systematically defined as a new (n, k, N/sub w/) Hamming checksum code, where n is the size of the code word, k is the number of information elements in the code word, and N/sub w/ is the number of weights employed, respectively. The proposed design with various weights is evaluated in terms of time and hardware overhead as well as overflow probability and round-off error. Two different schemes employing the (n, k, 2) and (n, k, 3) Hamming checksum code are illustrated using important matrix computations. Comparison with other schemes reveals that the (n, k, 3) Hamming checksum scheme is very efficient, while the hardware overhead is small.

This paper proposes a solution to delivering multicast data reliably not only to/from fired hosts but also to/from mobile hosts in combined fixed/mobile IP networks, particularly with regard to the per-source shortest path multicast routing protocol and remote subscription option in the IETF mobile IP multicast standard. We propose exploiting tree-based error recovery, well-established concept for reliable multicast in fixed networks, where a hierarchical logical tree of a source and receivers is constructed over which the responsibility of error recovery is distributed. We introduce a new entity named ARMM (Agent for Reliable Mobile Multicast) located at each cell of the wireless/mobile networks. Each ARMM participates in a logical tree as a parent of mobile hosts in the corresponding cell. It is responsible for retransmission and requesting for retransmission on behalf of the mobile hosts. With aid of ARMM, the proposed scheme features (1) adaptive reconstruction of logical trees due to host mobility, (2) retransmission to mobile hosts via optimal routes, (3) abstracting a group of mobile hosts into a single ARMM (and thus reducing logical tree maintenance overhead), (4) efficient use of wireless bandwidth for retransmission, and (5) low processing burden on mobile hosts themselves. The experimental results show that the proposed scheme provides higher scalability than a scheme applying a generalized tree-based reliable multicast in mobile environments.

In multicasting multimedia data, effectively adapting to heterogeneous receivers is very difficult. In this paper, we propose a new scheme called the "active traffic control mechanism for layered multimedia multicasting" (ATLM) in an active network environment. The proposed scheme controls traffic at each router using the traffic condition of immediate children nodes. Also, the traffic adjustment is much finer than existing multicast protocols of single-layer granularity. Computer simulation reveals that the proposed scheme significantly improves the amount of delivered traffic and the end-to-end delay by as much as 10%, as compared to layer-wise adjustment. Moreover, it does not require one to maintain multiple sessions for transferring multi-layer data but only one session, and thus reduces the overheads on session management and network bandwidth.

Group Communication System (GCS) is an important building block for reliable, fault-tolerant distributed applications. Due to highly variable and unpredictable latency in WANs, GCS needs to cope with frequent membership changes. It results in diminishing the scalability of GCS. In this paper we propose a new scheme that improves the scalability of group membership service for WANs by exploiting dedicated membership servers configured into multiple layers and reaching global agreement on the view through layered servers. It improves the scalability of a process group of N member processes from O(N/sup 2/) to O(N/sup 3/) with three layers, where N is the number of processes in the group. Computer simulation verifies that the proposed scheme requires much smaller number of message exchanges for reaching an agreement on a next view and does that much quicker. In addition, the proposed scheme prevents applications from being blocked until reaching the final view agreement if the applications permit temporary inconsistency.

Replication is often used to provide users of distributed virtual environments with high-performance interactions. Concurrency control is required to avoid inconsistent views among replicas due to multiple concurrent updates. G. Lann has developed a prediction-based concurrency control scheme to allow real-time interactions for users and to eliminate the need for repairs. The existing scheme does not scale in terms of delivering ownership on time as the number of users increases. In this paper, we propose a scalable prediction-based concurrency control scheme with entity-centric multicasting: only the users surrounding a target entity multicast the ownership requests, by using the multicast address assigned to the entity. The experimental results and analysis reported in this paper show that the proposed scheme achieves the benefits of prediction-based concurrency control with efficiency and scalability for large distributed virtual environments.

As distributed virtual environment (DVE) scales in terms of users and network latency, a key aspect to consider is scalability for interactive performance because a large number of objects likely impose heavy burden especially on the network and computational resources. To improve the scalability, various relevance-filtering mechanisms and aggregation mechanisms have been proposed. However the existing filtering mechanisms do not scale well in terms of interactive performance as the number of users increases and crowds in a specific place.In this paper, we propose a new scalable filtering scheme that reduces the number of messages by dynamically grouping users based on their interests and distance. Within a group, members communicate with each other with high fidelity. However, a representative sends up-to-dated group information of members with low transmission frequency when they are not of immediate interest but are still within the interest area. The representative is elected from members of the group in distributed manner. The proposed scheme enhances the interactive performance scalability of large-scale DVE systems as much as 18% compared with the existing approach.

We propose an enhanced prediction-based concurrency control scheme that supports the scalability of concurrency control for large distributed virtual environments especially where entities are highly populated and tend to gather closely. The prediction scheme is based on an entity-centric multicast group. Only the users surrounding a target entity multicast the ownership requests via an entity multicast group and become owner candidates. The current owner predicts the next owner among the owner candidates and sends an ownership to the next owner in advance. However, if entities are assigned their own multicast address when they are close to each other, users have to continuously issue join messages as moving by the entities. To reduce the network and message exchange overhead, we use the location proximity of entities in virtual environments. By grouping closely gathered entities into one entity group and sharing a multicast address among group member entities, we reduce the number of frequent join and leave operations and join messages, therefore maintain enough interactive performance. The experiment results show that the proposed mechanism improves scalability especially when entities are closely gathered.

As the scale of a network becomes larger in terms of both the number of users and geographic span, reliable multicast protocols suffer more severely from acknowledgement implosion and exposure to retransmissions. In this paper, we propose a scalable, efficient recovery scheme for a tree-based reliable multicast transport protocol. The scheme first constructs a logical tree of receivers as close to a multicast routing tree as possible by using error bitmap information. This ensures that the receivers residing at a higher level in a multicast routing tree than ones requesting retransmissions be appointed as parents in a corresponding logical tree. In our scheme, the logical tree is adaptively reconfigured as the session membership or the multicast route changes. Our scheme also forms local groups with separate multicast addresses in the tree for efficient recovery. We compare our scheme with a static tree-based reliable multicast protocol. Results of the simulation show that our adaptive tree-based recovery scheme outperforms the compared protocol in terms of the implosion and exposure as the session size increases.

Among the various systems developed for parallel and distributed computing, networks of workstations (NOWs) based on the Message Passing Interface (MPI) have been recognized as an efficient platform. In this paper, we implement and compare two important message logging protocols, pessimistic and optimistic, for a NOW employing MPI. An experiment reveals that the total execution time is not significantly affected by the number of failures, while the performance of the optimistic protocol is more influenced by the number of failures than the pessimistic protocol is. Also, the former is more efficient than the latter for a reasonable number of failure points.

Most of interprocess communications in transaction oriented distributed systems are based on client-server model, in which one process communicates with several processes simultaneously. Datagram service fits this type of communication rather than virtual circuit service. Transaction oriented distributed systems sometimes need to move large data such as a file or a part of database from a machine to other. The reliable delivery is fairly important since an error might cause retransmission of a whole data. In this paper, we propose the efficient mechanism for reliable datagram service, i) the concept of semi-virtual circuit and ii) the simple error detection and recovery algorithm based on our existing transport protocol, UDSTP. We describe implementation details supporting those facilities and performance results.