Hatim M. Behairy

Innovative Schemes for Resource Allocation in the Cloud for Media Streaming Applications

Media streaming applications have recently attracted a large number of users in the Internet. With the advent of these bandwidth-intensive applications, it is economically inefficient to provide streaming distribution with guaranteed QoS relying only on central resources at a media content provider. Cloud computing offers an elastic infrastructure that media content providers (e.g., Video on Demand (VoD) providers) can use to obtain streaming resources that match the demand. Media content providers are charged for the amount of resources allocated (reserved) in the cloud. Most of the existing cloud providers employ a pricing model for the reserved resources that is based on non-linear time-discount tariffs (e.g., Amazon CloudFront and Amazon EC2). Such a pricing scheme offers discount rates depending non-linearly on the period of time during which the resources are reserved in the cloud. In this case, an open problem is to decide on both the right amount of resources reserved in the cloud, and their reservation time such that the financial cost on the media content provider is minimized. We propose a simple-easy to implement-algorithm for resource reservation that maximally exploits discounted rates offered in the tariffs, while ensuring that sufficient resources are reserved in the cloud. Based on the prediction of demand for streaming capacity, our algorithm is carefully designed to reduce the risk of making wrong resource allocation decisions. The results of our numerical evaluations and simulations show that the proposed algorithm significantly reduces the monetary cost of resource allocations in the cloud as compared to other conventional schemes.

Statistical Modeling of Ultrawideband MIMO Propagation Channel in a Warehouse Environment

This paper describes an extensive propagation channel measurement campaign in a warehouse environment for line-of-sight (LOS) and nonline-of-sight (NLOS) scenarios. The measurement setup employs a vector network analyzer operating in the 2-8-GHz frequency band combined with an 8 × 8 virtual multiple-input multiple-output (MIMO) antenna array. We develop a comprehensive statistical propagation channel model based on high-resolution extraction of multipath components and subsequent spatiotemporal clustering analysis. The intracluster direction of departure (DoD), direction of arrival (DoA), and the time of arrival (ToA) are independent, both for the LOS and NLOS scenarios. The intracluster DoD and DoA can be approximated by the Laplace distribution, and the intracluster ToA can be approximated by an exponential mixture distribution. The intercluster analysis, however, shows a dependency between the cluster DoD, DoA, and ToA. To capture this dependency, we separately model the clusters caused by single and multiple bounce scattering along the aisles in the warehouse. The intercluster DoD distribution follows a Laplace distribution, while the cluster DoA conditioned on the DoD is approximated by a Gaussian mixture distribution. The model was validated using the capacity and delay-spread values.

Bayesian multi-target localization using blocking statistics in multipath environments

Passive localization based on distributed MIMO radar has a wide variety of applications ranging from security/military to consumer support. This paper considers the problem of localizing a (passive) target through time-of-arrival measurements from a number of (distributed) single-antenna transmitters to distributed receivers in the presence of objects that can block the line of sight between transmitter/receiver and target. The key effect investigated in this paper is the correlation of the blocking between an object and different transmitters/receivers. We formulate the problem in a Bayesian framework and suggest a sub-optimum but efficient algorithm to solve this problem; simulations show considerably better performance than can be achieved by ignoring the blocking statistics.

Multiple Parallel Concatenated Gallager Codes: Code Design and Decoding Techniques

Abstract A novel technique to concatenate multiple component low density parity check (LDPC) codes in parallel is presented. Multiple Parallel Concatenated Gallager Codes leverage the advantage of breaking an equivalently long LDPC code into multiple small (lower-complexity) LDPC codes to offer scalability and scope for improving performance in practical implementation for delay-sensitive and resource-constrained applications. The construction as well as two proposed decoding techniques for MPCGCs is evaluated. It is shown that by optimizing the construction of the component LDPC codes and the number of decoding iterations at various channel conditions, the performance of MPCGCs can be significantly improved and their complexity reduced.

Dual Mode PCGCs For Advanced Wireless Communications Networks

Parallel concatenated Gallager codes (PCGCs) are presented in two different modes of decoding. The turbo decoding mode is proposed for long, delay sensitive applications to reduce the decoding complexity, while the single matrix decoding may be used for short data oriented applications in advanced wireless communications networks. Techniques for the design, analysis, and convergence predictions of PCGCs are discussed.

Local linear embedded regression in the quantitative analysis of glucose in near infrared spectra

This paper investigates the use of Local Linear Embedded Regression (LLER) for the quantitative analysis of glucose from near infrared spectra. The performance of the LLER model is evaluated and compared with the regression techniques Principal Component Regression (PCR), Partial Least Squares Regression (PLSR) and Support Vector Regression (SVR) both with and without pre-processing. The prediction capability of the proposed model has been validated to predict the glucose concentration in an aqueous solution composed of three components (urea, triacetin and glucose). The results show that the LLER method offers improvements in comparison to PCR, PLSR and SVR.

A Novel L-Shape Ultra Wideband Chipless Radio-Frequency Identification Tag

A novel compact dual-polarized-spectral-signature-based chipless radio-frequency identification (RFID) tag is presented. Specifically, an L-shape resonator-based structure is optimized to have different spectral signatures in both horizontal and vertical polarizations, in order to double the encoding capacity. Resonators’ slot width and the space between closely placed resonators are also optimized to enhance the mutual coupling, thereby helping in achieving high-data encoding density. The proposed RFID tag operates over 5 GHz to 10 GHz frequency band. As a proof of concept, three different 18-bit dual-polarized RFID tags are simulated, fabricated, and tested in an anechoic chamber environment. The measurement data show reasonable agreement with the simulation results, with respect to resonators’ frequency positions, null depth, and their bandwidth over the operational spectrum.

Asymptotic Blind-Spot Analysis of Localization Networks Under Correlated Blocking Using a Poisson Line Process

In a localization network, the line-of-sight between anchors (transceivers) and targets may be blocked due to the presence of obstacles in the environment. Due to the non-zero size of the obstacles, the blocking is typically correlated across both anchor and target locations, with the extent of correlation increasing with obstacle size. If a target does not have line-of-sight to a minimum number of anchors, then its position cannot be estimated unambiguously and is, therefore, said to be in a blind-spot. However, the analysis of the blind-spot probability of a given target is challenging due to the inherent randomness in the obstacle locations and sizes. In this letter, we develop a new framework to analyze the worst-case impact of correlated blocking on the blind-spot probability of a typical target; in particular, we model the obstacles by a Poisson line process and the anchor locations by a Poisson point process. For this setup, we define the notion of the asymptotic blind-spot probability of the typical target and derive a closed-form expression for it as a function of the area distribution of a typical Poisson-Voronoi cell. As an upper bound for the more realistic case when obstacles have finite dimensions, the asymptotic blind-spot probability is useful as a design tool to ensure that the blind-spot probability of a typical target does not exceed a desired threshold,

Design of UWB chipless RFID tags using 8-bit open circuit stub resonators

\boldsymbol {\epsilon }

Blind-spot analysis of localization networks using second-order blocking statistics

.