Behnoosh Hariri

A Survey of First-Person Shooter Gaming Traffic on the Internet

As online games become significant contributors to Internet traffic, practitioners must look to measure and understand game traffic in order to provision for it. Despite gamings significant implications for the Internet, no general framework exists for validating various and sometimes contradicting game traffic models. Practitioners must therefore find and study dozens of reports and papers, some with conflicting measurements, to understand traffic characteristics and use this information in their systems. This survey looks at the most representative first-person shooter (FPS) game traffic to provide a one-stop shop for practitioners.

A personal assistive system for nutrient intake monitoring

According to the World Health Organization (WHO) statistics, Obesity has reached epidemic proportions, with over than 1.5 billion adults suffering from overweight in 2008. As Obesity and overweight are major risks to human health, obesity treatment has been the focus of a large number of recent studies. Obesity treatment requires constant monitoring of the patients diet. The smart technologies of todays intelligent environment can be used towards development of such monitoring system for obesity treatment. In this article we will propose a smart system that takes advantage of smartphones to build a platform for the monitoring of the amount of calorie intake for obesity patients. The patient uses the built-in camera of the smartphone to take a picture of any food that he/she wants to eat. The system will then process the images to detect the type of food and portion size, and uses the information to estimate the amount of calories.

A delaunay triangulation architecture supporting churn and user mobility in MMVEs

This article proposes a new distributed architecture for update message exchange inmassively multi-user virtual environments (MMVE). MMVE applications require delivery of updates among various locations in the virtual environment. The proposed architecture here exploits the location addressing of geometrical routing in order to alleviate the need for IP-specific queries. However, the use of geometrical routing requires careful choice of overlay to achieve high performance in terms of minimizing the delay. At the same time, the MMVE is dynamic, in sense that users are constantly moving in the 3D virtual space. As such, our architecture uses a distributed topology control scheme that aims at maintaining the requires QoS to best support the greedy geometrical routing, despite user mobility or churn. We will further prove the functionality and performance of the proposed scheme through both theory and simulations.

YawDD: a yawning detection dataset

In this paper, we present two video datasets of drivers with various facial characteristics, to be used for designing and testing algorithms and models for yawning detection. For collecting these videos, male and female candidates were asked to sit in the drivers seat of a car. The videos are taken in real and varying illumination conditions. In the first dataset, the camera is installed under the front mirror of the car. Each participant has three or four videos and each video contains different mouth conditions such as normal, talking/singing, and yawning. In the second dataset, the camera is installed on the dash in front of the driver, and each participant has one video with the above-mentioned different mouth conditions all in the same video. The car was parked for both datasets to keep the environment safe for the participants. As a benchmark, we also present the results of our own yawning detection method, and show that we can achieve a much higher accuracy in the scenario with the camera installed on the dash in front of the driver.

A Distributed Measurement Scheme for Internet Latency Estimation

Estimating latency between the hosts in the Internet can play a significant role in the improvement of the performance of many services that use latency among hosts to make routing decisions. A popular example is peer-to-peer networks that need to build an overlay between peers in a way that minimizes the message exchange delay among the peers. Acquisition of latency information requires a considerable amount of measurements to be performed at each node in order for that node to keep a record of its latency to all the other nodes. Moreover, measured latency values are frequently subject to change and need to be regularly repeated in order to be updated against network dynamics. This has motivated the use of techniques that alleviate the need for a large number of empirical measurements and try to predict the entire network latency matrix using a small set of latency measurements. Coordinate-based approaches are the most popular solutions to this problem. The basic idea behind coordinate-based schemes is to model the latency between each pair of nodes as the virtual distance among those nodes in a virtual coordinate system. This paper proposes a new decentralized coordinate-based solution to the problem of Internet delay measurement. The simulation results demonstrate that the proposed system provides relatively accurate estimations.

On the necessity of using Delaunay Triangulation substrate in greedy routing based networks

Large scale decentralized communication systems have motivated a new trend towards online routing where routing decisions are performed based on a limited and localized knowledge of the network. Geometrical greedy routing has been among the simplest and most common online routing schemes. While a geometrical online routing scheme is expected to deliver each packet to the point in the network that is closest to the destination, geometrical greedy routing, when applied over generalized substrate graphs, does not guarantee such delivery as its forwarding decision might deliver packets to a localized minimum instead. This letter investigates the necessary and sufficient conditions of greedy supporting graphs that would guarantee such delivery when used as a greedy routing substrate.

LOADER: A Location-Aware Distributed Virtual Environment Architecture

This article proposes a new architecture for distributed data management and update massage exchange in massively multi-user virtual environments. The key points in the design of such environments are scalability and QoS-aware message delivery. Therefore it requires robust distributed algorithms in a dynamic peer-to-peer system with frequent node arrivals and departures. Our proposed approach is mainly based on distributed hash tables (DHTs) in order to achieve a decentralized system where any participating node can efficiently retrieve the updates associated with a given object. Therefore, the responsibility of maintaining the mapping from names to values is distributed among the nodes in a way that a change in the set of participants causes a minimal amount of disruption. This allows the design to scale to extremely large number of nodes and to handle continual node arrivals, departures, and failures. The original form of DHT has several deficiencies for multi-user virtual environment application. This work introduces LOADER: a Location Aware Distributed Virtual Environment Architecture, which is the result of a series of efforts in alleviating a number of these problems including lack of location-aware ID assignment. Moreover, efficient multicast is addressed as an inevitable requirement for message exchange in multi-user environments and a solution based on Scribe is proposed.

A Hierarchical HMM Model for Online Gaming Traffic Patterns

Online games seem to become major contributors to Internet traffic. Therefore a great deal of effort is being devoted to the analysis and modeling of network game traffic. However the results from the previous measurements have not been unified into a single frame work while gaming traffic model is of high importance to network designers. Considering the fact that gaming traffic is generated by the players interaction during game sessions, we propose to model the players activities and later use this model to extract the traffic pattern. We will propose a model for the gaming player behavior regarding traffic generation based on hierarchical hidden Markov model. Player activities are inherently highly complicated. In order to deal with the complicated player activities, it is necessary to incorporate a hierarchical structure in representation of activities, in which a player activity is represented by a combination of multiple activity units, or actions. Any action or any combination of actions may occur during a game session. The HHMM hierarchy ends in production states that result in the transmission of update messages to the game server. HHMM Observations are considered to be client packet size and packet interarrival time.

NL-DHT: A Non-uniform Locality Sensitive DHT Architecture for Massively Multi-user Virtual Environment Applications

DHT networks offer a scalable structure for use in massively multi-user virtual environments (MMVEs). However, an issue with DHT structures is their use of uniform location-independent ID assignment. This conflicts with the locality-sensitive non-uniform ID assignment needed to achieve efficient latency-aware routing in MMVE applications. Our proposed solution is to use a modified version of the Hilbert space-filling curve in order to map users¿ locations from three dimensional VE space into single dimensional DHT ID space in the best locality preserving way. The need for such modified Hilbert curve is due to the fact that users are not typically homogeneously distributed in MMVEs while Hilbert curves fill space in a uniform manner. Generation of, and mapping to, an entire Hilbert curve can be very expensive in terms of computational complexity. Thus we present a fast dynamic mapping of MMVE locations to the modified Hilbert curve to reduce the mapping complexity.

Modeling and evaluation of overlay generation problem for peer-assisted video adaptation and streaming

In this paper, we consider the problem of overlay generation for video adaptation and streaming applications in a way to efficiently utilize the bandwidth and computing power of the participating peers. Therefore, the proposed architecture performs regular streaming functions as well as video adaptation functions, moving the video contents adaptation computation load away from dedicated media-streaming/adaptation servers to the participating peers. To verify the performance of our design, we followed an analytical approach based on 0-1 Integer Linear Programming method to model the system and to calculate the optimum overlay. The performance of our scheme is evaluated by simulations. Preliminary results demonstrate that our design performance nearly follows the optimal boundary in terms of resource utilization.