Embedded Files
Papers
Papers
CrossCount: A Deep Learning System for Device-free Human Counting using WiFi
CrossCount: A Deep Learning System for Device-free Human Counting using WiFi
Osama T. Ibrahim, Walid Gomaa and Moustafa Youssef
IEEE Sensor Journal (Q1 - Impact factor: 3.076)
Abstract:
Counting humans is an essential part of many people-centric applications. In this paper, we propose CrossCount: an accurate deep-learning-based human count estimator that uses a single WiFi link to estimate the human count in an area of interest. The main idea is to depend on the temporal link-blockage pattern as a discriminant feature that is more robust to wireless channel noise than the signal strength, hence delivering a ubiquitous and accurate human counting system. As part of its design, CrossCount addresses a number of deep learning challenges such as class imbalance and training data augmentation for enhancing the model generalizability. Implementation and evaluation of CrossCount in multiple testbeds show that it can achieve a human counting accuracy to within a maximum of 2 persons 100% of the time. This highlights the promise of CrossCount as a ubiquitous crowd estimator with non-labour-intensive data collection from off-the-shelf devices.
Zero-Calibration Device-Free Localization for the IoT Based on Participatory Sensing
Zero-Calibration Device-Free Localization for the IoT Based on Participatory Sensing
Osama T. Ibrahim, Walid Gomaa and Moustafa Youssef
IEEE Global Communications Conference (GlobeCom) 2018, Abu Dhabi, UAE
Abstract:
Device-free localization (DFL) is an emerging technology for estimating the position of a human or object that is not equipped with any electronic tag, nor participate actively in the localization process. Similar to device-based localization, the initial phase in DFL is to build the fingerprint database which is usually done manually using site surveying. This process is tedious, time-consuming, and vulnerable to environmental dynamics. Motivated by the recent advances in the Internet of Things (IoT), this paper introduces RadioGrapher; a system that automates the process of device-free fingerprint calibration in IoT environments. RadioGrapher leverages the device-based locations of entities in the area of interest in a crowd-sensing manner, aided with Fresnel zones of the wirelessly connected IoT devices to automatically construct a device-free fingerprint. Experimental evaluation of RadioGrapher in an IoT testbed using multiple entities shows that it can construct DFL fingerprints with high accuracy. Moreover, its median localization accuracy is comparable to that of manual fingerprinting. This comes with no calibration overhead, highlighting the promise of RadioGrapher as a crowdsourcing device-free fingerprint constructor in IoT environments.
RadioGrapher System Architecture
An Efficient Load Balancing Method for Tree Algorithms
An Efficient Load Balancing Method for Tree Algorithms
Osama T. Ibrahim, Ahmed El-Mahdy
IEEE International Conference on Scalable Computing and Communications (ScalCom) 2016, Toulouse, France.
Download paper - Conference Presentation - Poster.
Abstract:
Nowadays, multiprocessing is mainstream with exponentially increasing number of processors. Load balancing is, therefore, a critical operation for the efficient execution of parallel algorithms. In this paper we consider the fundamental class of tree-based algorithms that are notoriously irregular,, hard to load-balance with existing static techniques. We propose a hybrid load balancing method using the utility of statistical random sampling in estimating the tree depth, node count distributions to uniformly partition an input tree. To conduct an initial performance study, we implemented the method on an Intel Xeon Phi accelerator system. We considered the tree traversal operation on both regular, irregular unbalanced trees manifested by Fibonacci, unbalanced (biased) randomly generated trees, respectively. The results show scalable performance for up to the 60 physical processors of the accelerator, as well as an extrapolated 128 processors case.
Proposed Framework:
Patent
Patent
Efficient Load-Balancing Method and System for Tree-Based Applications
Efficient Load-Balancing Method and System for Tree-Based Applications
US Patent - Publication No.: US 2018/0095794A1.
Patent Link - Application Publication.
Abstract:
In a system having multiple parallel processors, a process to enhance performance of tree based applications by balancing the processing load amongst all available parallel processors when processing the tree structure. Tree nodes and leaves are uniformly sampled at random to estimate the corresponding work (such as node counts and leave work). This is done through novel uniform node sample and weighted random depth probing. A linear workload mapping then maps subtrees into sub-intervals of a one-dimensional interval. Such mapping facilities inverse mapping of the estimated workload to achieve efficient partitioning of the tree. The process further adaptively decides upon subtrees to sample allowing for matching with the characteristics of input trees, decreasing the number of probes, while resulting in accurate load-balancing. The process provides for fast load balancing for complex tree-based applications by exploiting statistical random sampling and requires only modest memory resources for such process making it suitable and applicable to even modest embedded devices. A significant speedup in processing is achieved which increases with the number of available processors.
Thesis
Thesis
MSc Thesis
MSc Thesis
Efficient Load Balancing For Tree Algorithms
Egypt-Japan University of Science and Technology (E-JUST)
Google Sites
Report abuse