Showing papers by "Ali H. Sayed published in 2008"

Adaptive Filters

Abstract: Adaptive Filters Adaptive filtering is a topic of immense practical and theoretical value, having applications in areas ranging from digital and wireless communications to biomedical systems. Now, preserving the style and main features of the earlier award-winning publication, Fundamentals of Adaptive Filtering (2005 Terman Award), the author offers readers and instructors a concentrated, systematic, and up-to-date treatment of the subject in this valuable new book. Adaptive Filters allows readers to gain a gradual and solid introduction to the subject, its applications to a variety of topical problems, existing limitations, and extensions of current theories. The book consists of eleven partseach part containing a series of focused lectures and ending with bibliographic comments, problems, and computer projects with MATLAB solutions available to all readers. Additional features include: Numerous tables, figures, and projects Special focus on geometric constructions, physical intuition, linear-algebraic concepts, and vector notation Background material on random variables, linear algebra, and complex gradients collected in three introductory chapters Complete solutions manual available for instructors MATLAB solutions available for all computer projects Adaptive Filters offers a fresh, focused look at the subject in a manner that will entice students, challenge experts, and appeal to practitioners and instructors.

Optimal Linear Cooperation for Spectrum Sensing in Cognitive Radio Networks

Abstract: Cognitive radio technology has been proposed to improve spectrum efficiency by having the cognitive radios act as secondary users to opportunistically access under-utilized frequency bands. Spectrum sensing, as a key enabling functionality in cognitive radio networks, needs to reliably detect signals from licensed primary radios to avoid harmful interference. However, due to the effects of channel fading/shadowing, individual cognitive radios may not be able to reliably detect the existence of a primary radio. In this paper, we propose an optimal linear cooperation framework for spectrum sensing in order to accurately detect the weak primary signal. Within this framework, spectrum sensing is based on the linear combination of local statistics from individual cognitive radios. Our objective is to minimize the interference to the primary radio while meeting the requirement of opportunistic spectrum utilization. We formulate the sensing problem as a nonlinear optimization problem. By exploiting the inherent structures in the problem formulation, we develop efficient algorithms to solve for the optimal solutions. To further reduce the computational complexity and obtain solutions for more general cases, we finally propose a heuristic approach, where we instead optimize a modified deflection coefficient that characterizes the probability distribution function of the global test statistics at the fusion center. Simulation results illustrate significant cooperative gain achieved by the proposed strategies. The insights obtained in this paper are useful for the design of optimal spectrum sensing in cognitive radio networks.

Diffusion Least-Mean Squares Over Adaptive Networks: Formulation and Performance Analysis

Abstract: We formulate and study distributed estimation algorithms based on diffusion protocols to implement cooperation among individual adaptive nodes. The individual nodes are equipped with local learning abilities. They derive local estimates for the parameter of interest and share information with their neighbors only, giving rise to peer-to-peer protocols. The resulting algorithm is distributed, cooperative and able to respond in real time to changes in the environment. It improves performance in terms of transient and steady-state mean-square error, as compared with traditional noncooperative schemes. Closed-form expressions that describe the network performance in terms of mean-square error quantities are derived, presenting a very good match with simulations.

Diffusion recursive least-squares for distributed estimation over adaptive networks

Abstract: We study the problem of distributed estimation over adaptive networks where a collection of nodes are required to estimate in a collaborative manner some parameter of interest from their measurements. The centralized solution to the problem uses a fusion center, thus, requiring a large amount of energy for communication. Incremental strategies that obtain the global solution have been proposed, but they require the definition of a cycle through the network. We propose a diffusion recursive least-squares algorithm where nodes need to communicate only with their closest neighbors. The algorithm has no topology constraints, and requires no transmission or inversion of matrices, therefore saving in communications and complexity. We show that the algorithm is stable and analyze its performance comparing it to the centralized global solution. We also show how to select the combination weights optimally.

Collaborative wideband sensing for cognitive radios

Abstract: Cognitive radio (CR) has recently emerged as a promising technology to revolutionize spectrum utilization in wireless communications. In a CR network, secondary users continuously sense the spectral environment and adapt transmission parameters to opportunistically use the available spectrum. A fundamental problem for CRs is spectrum sensing; secondary users need to reliably detect weak primary signals of possibly different types over a targeted wide frequency band in order to identify spectral holes for opportunistic communications. Conceptually and practically, there is growing awareness that collaboration among several CRs can achieve considerable performance gains. This article provides an overview of the challenges and possible solutions for the design of collaborative wideband sensing in CR networks. It is argued that collaborative spectrum sensing can make use of signal processing gains at the physical layer to mitigate strict requirements on the radio frequency front-end and to exploit spatial diversity through network cooperation to significantly improve sensing reliability.

Wideband Spectrum Sensing in Cognitive Radio Networks

Abstract: Spectrum sensing is an essential enabling functionality for cognitive radio networks to detect spectrum holes and opportunistically use the under-utilized frequency bands without causing harmful interference to legacy networks. This paper introduces a novel wideband spectrum sensing technique, called multiband joint detection, which jointly detects the signal energy levels over multiple frequency bands rather than consider one band at a time. The proposed strategy is efficient in improving the dynamic spectrum utilization and reducing interference to the primary users. The spectrum sensing problem is formulated as a class of optimization problems in interference limited cognitive radio networks. By exploiting the hidden convexity in the seemingly non-convex problem formulations, optimal solutions for multiband joint detection are obtained under practical conditions. Simulation results show that the proposed spectrum sensing schemes can considerably improve the system performance. This paper establishes important principles for the design of wideband spectrum sensing algorithms in cognitive radio networks.

Wideband Spectrum Sensing in Cognitive Radio Networks

Abstract: Spectrum sensing is an essential enabling functionality for cognitive radio networks to detect spectrum holes and opportunistically use the under-utilized frequency bands without causing harmful interference to legacy networks. This paper introduces a novel wideband spectrum sensing technique, called multiband joint detection, which jointly detects the signal energy levels over multiple frequency bands rather than consider one band at a time. The proposed strategy is efficient in improving the dynamic spectrum utilization and reducing interference to the primary users. The spectrum sensing problem is formulated as a class of optimization problems in interference limited cognitive radio networks. By exploiting the hidden convexity in the seemingly non-convex problem formulations, optimal solutions for multiband joint detection are obtained under practical conditions. Simulation results show that the proposed spectrum sensing schemes can considerably improve the system performance. This paper establishes important principles for the design of wideband spectrum sensing algorithms in cognitive radio networks.

Diffusion strategies for distributed Kalman filtering: formulation and performance analysis

Abstract: We consider the problem of distributed Kalman filtering, where a set of nodes are required to collectively estimate the state of a linear dynamic system from their individual measurements. Our focus is on diffusion strategies, where nodes communicate with their direct neighbors only, and the information is diffused across the network. We derive and analyze the mean and mean-square performance of the proposed algorithms and show by simulation that they outperform previous solutions.

Diffusion adaptive networks with changing topologies

Abstract: Adaptive networks (AN) have been recently proposed to address distributed estimation problems [1]-[4]. Here we extend prior work to changing topologies and data-normalized algorithms. The resulting framework may also treat signals with general distributions, rather than Gaussian, provided that certain data statistical moments are known. A byproduct of this formulation is a probabilistic diffusion adaptive network: a simpler yet robust variant of the standard diffusion algorithm [2].

Steady-State Performance Analysis of a Variable Tap-Length LMS Algorithm

Abstract: A steady-state performance analysis of the fractional tap-length (FT) variable tap-length least mean square (LMS) algorithm is presented in this correspondence. Based on the analysis, a mathematical formulation for the steady-state tap length is obtained. Some general criteria for parameter selection are also given. The analysis and the associated discussions give insight into the performance of the FT algorithm, which may potentially extend its practical applicability. Simulation results support the theoretical analysis and discussions.

Ieee signal processing society

Abstract: Poster Sessions and Show & Tell demonstrations are aimed at strengthening the interactions between researchers and practitioners. They offer an opportunity for participants to demonstrate their state-of-the-art results to a wide audience of professionals in the area of signal processing. POSTER PAPERS In the IEEE Signal Processing Society, no differentiation is made between the paper quality of poster and oral sessions. Papers are assigned primarily to ensure consistency among the sessions, at the discretion of the conference organizers. Some organizers elect to use poster sessions to provide the opportunity for attendees to meet authors personally and to discuss their papers in depth. The poster session papers must be vetted together with the oral session papers, to ensure the same standard of quality. Oral and Poster sessions, and those papers will be submitted to Xplore with the same conditions as oral and poster papers. We expect papers in Show & Tell Sessions to have novelty coming from practical realization techniques, interesting/new applications and advanced system structures, especially suitable for industrial applications. The paper vetting process should be the same as those in the Oral and Lecture sessions to ensure the same high quality. Note: Show & Tell papers, poster papers and oral papers are all considered equal for review, selection, and posting procedures in Xplore. SHOW & TELL DEMONSTRATIONS without a regular paper There may be cases where Show & Tell Demonstrations are made without an accompanying paper, or with a paper that has not undergone the same review process as the above papers. Such papers or other documentation of the demonstration will not be included in Xplore. It is the discretion of the conference organizers to determine whether these materials are to be included as supplements of the proceedings. Please note: all papers included in the conference proceedings and/or Xplore, must submit a copyright form. All of these conditions/criteria should be disclosed to authors during the submission stage of the process.

Diffusion LMS algorithms with information exchange

Abstract: We consider the problem of distributed estimation, where a set of nodes are required to collectively estimate some parameter of interest. We motivate and propose new versions of the diffusion LMS algorithm, including a version that outperforms previous solutions without increasing the complexity or communications, and others that obtain even better performance by allowing additional communications. We analyze their performance and compare with simulation results.

Spatial-spectral joint detection for wideband spectrum sensing in cognitive radio networks

Abstract: Spectrum sensing is an essential functionality that enables cognitive radios to detect spectral holes and opportunistically use under-utilized frequency bands without causing harmful interference to primary networks. Since individual cognitive radios might not be able to reliably detect weak primary signals due to channel fading/shadowing, this paper proposes a cooperative wideband spectrum sensing scheme, referred to as spatial-spectral joint detection, which is based on a linear combination of the local statistics from spatially distributed multiple cognitive radios. The cooperative sensing problem is formulated into an optimization problem, for which suboptimal but efficient solutions can be obtained through mathematical transformation under practical conditions.

Diffusion mechanisms for fixed-point distributed Kalman smoothing

Abstract: We consider the problem of fixed-point distributed Kalman smoothing, where a set of nodes are required to estimate the initial condition of a certain process based on their measurements of the evolution of the process. Specifically, we consider linear state-space models where the Kalman smoother gives us the MMSE estimate of the initial state of the system. We propose distributed diffusion solutions where nodes communicate with their neighbors and information is propagated through the network via a diffusion process. Hierarchical cooperation schemes are also described.

An Efficient Carrier Phase Synchronization Technique for High-Order M-QAM–OFDM

Abstract: In this correspondence, we propose a blind carrier phase synchronization algorithm for high-order M-ary quadrature amplitude modulation-orthogonal frequency division multiplexing (M-QAM-OFDM) systems, which can effectively recover residual frequency offset (RFO) in the presence of intercarrier interference (ICI). The proposed algorithm performs frequency and phase synchronization by using post-fast Fourier transform (FFT) demodulated signals without the aid of reference signals (e.g., pilots, guard intervals, and virtual carriers), and is simple to implement. By analyzing its open-loop characteristics, we show that the proposed algorithm is superior to the conventional decision-directed (DD) scheme for high-order M-QAM-OFDM systems.

Digital compensation of RF nonlinearities in software-defined radios

Abstract: The wideband RF receiver in a software-defined radio (SDR) system suffers from the nonlinear effects caused by the front-end analog processing. In the presence of strong blocker (interference) signals, nonlinearities introduce severe cross modulation over the desired signals. This paper investigates how the nonlinear distortions can be compensated for by using digital signal processing techniques. In the proposed solution, the SDR scans the wide spectrum and locates the desired signal and strong blocker signals. After down-converting these signals separately into the baseband, the baseband processor processes them jointly to mitigate the cross-modulation interferences. As a result, the sensitivity of the wideband RF receiver to the non-linearity impairment can be significantly lowered, simplifying the RF and analog circuitry design in terms of implementation cost and power consumption.

Mathematical models of cerebral hemodynamics for detection of vasospasm in major cerebral arteries.

Abstract: BACKGROUND Vasospasm is a common complication of aneurismal subarachnoid hemorrhage (SAH) that may lead to cerebral ischemia and death. The standard method for detection of vasospasm is conventional cerebral angiography, which is invasive and does not allow continuous monitoring of arterial radius. Monitoring of vasospasm is typically performed by measuring Cerebral Blood Flow Velocity (CBFV) in the major cerebral arteries and calculating the Lindegaard ratio. We describe an alternative approach to estimate intracranial arterial radius, which is based on modeling and state-estimation techniques. The objective is to obtain a better estimation than that offered by the Lindegaard ratio, that might allow for continuous monitoring and possibly vasospam prediction without the need for angiography. METHODS We propose two new models of cerebral hemodynamics. Model 1 is a more general version of Ursino's 1991 model that includes the effects of vasospasm, and Model 2 is a simplified version of Model 1. We use Model 1 to generate Intracranial Pressure (ICP) and CBFV signals for different vasospasm conditions, where CBFV is measured at the middle cerebral artery (MCA). Then we use Model 2 to estimate the arterial radii from these signals. FINDINGS Simulations show that Model 2 is capable of providing good estimates for the radius of the MCA, allowing the detection of the vasospasm. These changes in arterial radius are being estimated from measurements of CBFV, and CBF is never being measured directly. This is the main advantage of the model-based approach where several interrelations between CBFV, ABP and ICP are taken into account by the differential equations of the model. CONCLUSIONS Our results indicate that arterial radius may be estimated using measurements of ABP, ICP and CBFV, allowing the detection of vasospasm.

Carrier Recovery Enhancement for Maximum-Likelihood Doppler Shift Estimation in Mars Exploration Missions

Abstract: One of the most crucial stages of the Mars exploration missions is the entry, descent, and landing (EDL) phase. During EDL, maintaining reliable communication from the spacecraft to Earth is extremely important for the success of future missions, especially in case of mission failure. EDL is characterized by very deep accelerations, caused by friction, parachute deployment and rocket firing among others. These dynamics cause a severe Doppler shift on the carrier communications link to Earth. Methods have been proposed to estimate the Doppler shift based on Maximum Likelihood. So far these methods have proved successful, but it is expected that the next Mars mission, known as the Mars Science Laboratory, will suffer from higher dynamics and lower SNR. Thus, improving the existing estimation methods becomes a necessity. We propose a Maximum Likelihood approach that takes into account the power in the data tones to enhance carrier recovery, and improve the estimation performance by up to 3 dB. Simulations are performed using real data obtained during the EDL stage of the Mars Exploration Rover B (MERB) mission.

Kalman Filtering and RLS

Abstract: This chapter contains sections titled: Equivalence in Linear Estimation Kalman Filtering and Recursive Least-Squares Appendix: Extended RLS Algorithms ]]>

Steady-State Performance Analysis of a Variable

Abstract: A steady-state performance analysis of the fractional tap-length (FT) variable tap-length least mean square (LMS) algorithm is presented in this correspondence. Based on the analysis, a mathematical formulation for the steady-state tap length is obtained. Some general criteria for parameter selection are also given. The analysis and the asso- ciated discussions give insight into the performance of the FT algorithm, which may potentially extend its practical applicability. Simulation results support the theoretical analysis and discussions. Index Terms—Adaptive filters, steady-state performance analysis, vari- able tap-length LMS algorithm. I. INTRODUCTION The least mean square (LMS) adaptive algorithm has been exten- sively used as a consequence of its simplicity and robustness (1), (2). In many applications of the LMS algorithm, the tap length of the adap- tive filter is kept fixed. However, in certain situations, the tap length of the optimal filter is unknown or variable. According to the analysis in (3) and (4), the mean square error (MSE) of the adaptive filter is likely to increase if the tap length is undermodeled. To avoid such a situation, a sufficiently large filter tap length is needed. However, the computational cost and the excess mean square error (EMSE) of the LMS algorithm will increase if the tap length is too large; thus, a vari- able tap-length LMS algorithm is needed to find a proper choice of the tap length. Several variable tap-length LMS algorithms have been proposed in recent years. Among existing variable tap-length LMS algorithms, some are designed to not only establish a suitable steady-state tap length, but also to speed up the convergence rate (3), (5). These methods are based on the assumption that the unknown optimal filter has an impulse response sequence with an exponentially decaying en- velope, which limits their utility. Other methods are more general and are designed to search for the optimal filter tap length at steady-state (6)-(9); a summary of these works is given in (10). As analyzed in (10), the fractional tap-length (FT) algorithm is more robust and has lower computational complexity when compared with other methods. A convex combination structure of the FT algorithm has been proposed

Affine Projection Algorithm

Abstract: This chapter contains sections titled: Instantaneous Approximation Computational Cost Least-Perturbation Property Affine Projection Interpretation

Robust Adaptive Filters

Abstract: This chapter contains sections titled: A Posteriori-Based Robust Filters ?>-NLMS Algorithm A Priori-Based Robust Filters LMS Algorithm ?> ?> Filters

Optimal Linear Fusion for Distributed Detection via Semidenite Programming

Abstract: Consider the problem of signal detection via multiple distributed noisy sensors. We propose a linear decision fusion rule to combine the local statistics from individual sensors into a global statistic for binary hypothesis testing. The objective is to maximize the probability of detection subject to an upper limit on the probability of false alarm. We employ a divide-and-conquer strategy to divide the decision optimization problem into two subproblems, each of which is a nonconvex program with a quadratic constraint. Through a judicious reformulation and by employing a special matrix decomposition technique, we show that the two nonconvex subproblems can be solved by semidenite programs in a globally optimal fashion. Hence, we can obtain the optimal linear fusion rule for the distributed detection problem. Compared with the likelihood-ratio test approach, the optimized linear fusion rule can achieve comparable detection performance with considerable design e xibility and reduced complexity.

Performance of RLS and Other Filters

Abstract: This chapter contains sections titled: Performance of RLS Performance of Other Filters Performance Table for Small Step-Sizes

Steady-state performance of incremental learning over distributed networks for non-Gaussian data

Abstract: In this paper, the steady-state performance of the distributed least mean-squares (dLMS) algorithm within an incremental network is evaluated without the restriction of Gaussian distributed inputs. Computer simulations are presented to verify the derived performance expressions.

Summary and Notes

Abstract: This chapter contains sections titled: Summary of Main Results Bibliographic Notes

Normalized LMS Algorithm

Abstract: This chapter contains sections titled: Instantaneous Approximation Computational Cost Power Normalization Least-Perturbation Property