Abdelsalam Helal

Drishti: an integrated navigation system for visually impaired and disabled

Drishti is a wireless pedestrian navigation system. It integrates several technologies including wearable computers, voice recognition and synthesis, wireless networks, Geographic Information System (GIS) and Global positioning system (GPS). Drishti augments contextual information to the visually impaired and computes optimized routes based on user preference, temporal constraints (e.g. traffic congestion), and dynamic obstacles (e.g. ongoing ground work, road blockade for special events). The system constantly guides the blind user to navigate based on static and dynamic data. Environmental conditions and landmark information queried from a spatial database along their route are provided on the fly through detailed explanatory voice cues. The system also provides capability for the user to add intelligence, as perceived by, the blind user, to the central server hosting the spatial database. Our system is supplementary to other navigational aids such as canes, blind guide dogs and wheel chairs.

Smart home-based health platform for behavioral monitoring and alteration of diabetes patients.

Background: Researchers and medical practitioners have long sought the ability to continuously and automatically monitor patients beyond the confines of a doctors office. We describe a smart home monitoring and analysis platform that facilitates the automatic gathering of rich databases of behavioral information in a manner that is transparent to the patient. Collected information will be automatically or manually analyzed and reported to the caregivers and may be interpreted for behavioral modification in the patient. Method: Our health platform consists of five technology layers. The architecture is designed to be flexible, extensible, and transparent, to support plug-and-play operation of new devices and components, and to provide remote monitoring and programming opportunities. Results: The smart home-based health platform technologies have been tested in two physical smart environments. Data that are collected in these implemented physical layers are processed and analyzed by our activity recognition and chewing classification algorithms. All of these components have yielded accurate analyses for subjects in the smart environment test beds. Conclusions: This work represents an important first step in the field of smart environment-based health monitoring and assistance. The architecture can be used to monitor the activity, diet, and exercise compliance of diabetes patients and evaluate the effects of alternative medicine and behavior regimens. We believe these technologies are essential for providing accessible, low-cost health assistance in an individuals own home and for providing the best possible quality of life for individuals with diabetes.

Decentralized ad-hoc groupware API and framework for mobile collaboration

We describe a mobile collaborative system designed for wireless, ad-hoc collaboration. In recent years, mobile computing has emerged as a new discipline in the field of computer science. Due to advances in technology, portable computing devices have become more pervasive. From smart phones, and personal digital assistants (PDAs) running embedded operating systems, to portable computers running conventional desktop operating systems, these devices have increasingly provided communication capabilities that utilize wireless connections. With those communication capabilities firmly established, the next logical step is in the direction of greater interactions between mobile users equipped with such devices. However, conventional collaborative tools are ill suited for the demands of portable computers and mobile networks, especially in situations in which no fixed-network infrastructure is present. With these considerations in mind, we designed and implemented a collaborative environment and a framework API suited towards ad-hoc networks of small mobile devices. By creating such a framework, developers can easily take advantage of a decentralized and fault-tolerant collaborative environment, and rapidly develop custom collaboration spaces suited towards their specific need.

ILC-TCP: an interlayer collaboration protocol for TCP performance improvement in mobile and wireless environments

The growth of the wireless Internet has led to the optimization of network protocols to provide for a better performance. Most of the Internet traffic uses TCP, the de facto transport layer protocol. Unfortunately, TCP performance degrades in the mobile and wireless environments. A good amount of research has been attempted to improve its performance in the unpredictable mobile and wireless environments where link disconnections, packets losses and delays are common. Most of these proposed solutions target the case where the mobile host acts as a TCP receiver. For various reasons, almost none of the solutions have reached the stage of deployment. In this paper, we propose an interlayer collaboration model for TCP performance improvement in mobile and wireless environments. We specifically target the case where a mobile acts as a TCP sender. ILC-TCP is an end-to-end approach and does not need any special support from the base station infrastructure. ILC-TCP is evaluated against the normal TCP in various scenarios. Performance results suggest that ILC-TCP performs better than the normal TCP in many scenarios involving long disconnections, frequent disconnections and in the scenarios where a mobile host moves at considerable speeds.

Incremental hoarding and reintegration in mobile environments

Disconnection is one of the popular techniques for operating in mobile environments and is here to stay, until long-range wireless connectivity becomes a reality. However, disconnection requires periodic hoarding and reintegration of data, which raises performance issues especially during weak connection. A common hoarding and reintegration mechanism involves complete transfer of contents. In order to hoard and reintegrate efficiently, an incremental approach is being introduced to do data transfers based on the delta between changes. Analysis also shows that incremental reintegration is particularly beneficial in the weakly connected mode of operation. The incremental hoarding and reintegration setup is built within the Coda File System of the Carnegie Mellon University to replace the full file transfer mechanism with the incremental approach, based on the Revision Control System.

CAD-HOC: a CAD-Like tool for generating mobility benchmarks in ad-hoc networks

The paper addresses the problem of using unrealistic mobility scenarios for simulation of ad-hoc routing protocols. We present CAD-HOC, a cousin tool to Network Simulator (ns), which allows ad-hoc networking experimentation to be performed under visually realistic scenarios such as an airport or a bus terminal, buildings, highways and other facilities. CAD-HOC focuses on capturing the visual scenario and transforming it into mobility and connection benchmarks, which are subsequently fed to ns to drive simulation experiments. Generation of mobility and connection benchmarks is facilitated by allowing the users to choose from Brownian, Column, Pursue, and Nomadic models, or to specify user-defined movements. The tool also provides online analyses on ad-hoc mobility, ad-hoc connectivity and message complexity, which give an estimate of the overall ns simulation complexity, as the mobility scenario is built. Comparative simulation based on CAD-HOC generated scenarios raise several questions about the reliability of ns simulation results of ad-hoc routing protocols based on random, nonrealistic scenarios.

Exploiting visual quasi-periodicity for real-time chewing event detection using active appearance models and support vector machines

Steady increases in healthcare costs and obesity have inspired recent studies into cost-effective, assistive systems capable of monitoring dietary habits. Few researchers, though, have investigated the use of video as a means of monitoring dietary activities. Video possesses several inherent qualities, such as passive acquisition, that merits its analysis as an input modality for such an application. To this end, we propose a method to automatically detect chewing events in surveillance video of a subject. Firstly, an Active Appearance Model (AAM) is used to track a subject’s face across the video sequence. It is observed that the variations in the AAM parameters across chewing events demonstrate a distinct periodicity. We utilize this property to discriminate between chewing and non-chewing facial actions such as talking. A feature representation is constructed by applying spectral analysis to a temporal window of model parameter values. The estimated power spectra subsequently undergo non-linear dimensionality reduction. The low-dimensional embedding of the power spectra are employed to train a binary Support Vector Machine classifier to detect chewing events. To emulate the gradual onset and offset of chewing, smoothness is imposed over the class predictions of neighboring video frames in order to deter abrupt changes in the class labels. Experiments are conducted on a dataset consisting of 37 subjects performing each of five actions, namely, open- and closed-mouth chewing, clutter faces, talking, and still face. Experimental results yielded a cross-validated percentage agreement of 93.0%, indicating that the proposed system provides an efficient approach to automated chewing detection.

Assistive environments for successful aging

With nearly 80 million baby boomers in the United States just reaching their sixties, the demand for senior-oriented devices and services will explode in the coming years. Managing the increasing health-care costs for such a population requires developing technologies that will allow seniors to maintain active, independent lifestyles. Pervasive computing pervasive computing environments, such as smart homes , bundle assistive technologies assistive technology and specially designed architectural and home furnishing elements provide health-care and well-being services to its residents. However, for such environments to be commercially viable, we require a system that allows technology to be easily utilized and included as it enters the market place. Also we require new technology to be introduced in a plug-and-play fashion, and applications that are developed by programmers, not system integrators. The Gator Tech Smart House, a full-size, free-standing residential home located in the Oak Hammock Retirement Community in Gainesville, Florida, is an example of this kind of assistive environment . It uses the Atlas sensor network platform, an enabling technology that combines a hardware platform and software middleware , making the Gator Tech Smart House a truly programmable pervasive computing space.

A resilient application-level failure detection system for distributed computing environments

A methodology for detecting failures that occur in distributed computer systems connected by a communications network is described. The methodology utilizes active polling of monitored systems. The entities polled must be service entities that function at the application layers of service providing machines. A prototype system has been implemented to test this methodology.

Algorithms for the detection of chewing behavior in dietary monitoring applications

The detection of food consumption is key to the implementation of successful behavior modification in support of dietary monitoring and therapy, for example, during the course of controlling obesity, diabetes, or cardiovascular disease. Since the vast majority of humans consume food via mastication (chewing), we have designed an algorithm that automatically detects chewing behaviors in surveillance video of a person eating. Our algorithm first detects the mouth region, then computes the spatiotemporal frequency spectrum of a small perioral region (including the mouth). Spectral data are analyzed to determine the presence of periodic motion that characterizes chewing. A classifier is then applied to discriminate different types of chewing behaviors. Our algorithm was tested on seven volunteers, whose behaviors included chewing with mouth open, chewing with mouth closed, talking, static face presentation (control case), and moving face presentation. Early test results show that the chewing behaviors induce a temporal frequency peak at 0.5Hz to 2.5Hz, which is readily detected using a distance-based classifier. Computational cost is analyzed for implementation on embedded processing nodes, for example, in a healthcare sensor network. Complexity analysis emphasizes the relationship between the work and space estimates of the algorithm, and its estimated error. It is shown that chewing detection is possible within a computationally efficient, accurate, and subject-independent framework.