In decentralized systems like the Internet of Things (IoT) and cyber-physical networks, where data are distributed across multiple nodes, ensuring accurate and robust data analysis is crucial. Existing methods for penalized quantile regression often struggle with asynchronous operations and multiple updates per node, leading to inconsistencies across these distributed nodes. To address these challenges, we propose the Federated Smoothing ADMM (FSAD) algorithm, which integrates non-convex sparse penalties – specifically, the minimax concave penalty (MCP) and smoothly clipped absolute deviation (SCAD)–to effectively identify significant predictors while retaining sparsity. By incorporating a total variation norm within a smoothing ADMM framework, FSAD supports asynchronous updates and ensures model consistency across nodes, thereby overcoming traditional convergence limitations in non-convex, federated settings. Our theoretical analysis provides rigorous convergence guarantees, and extensive simulations confirm that FSAD outperforms existing methods in terms of both accuracy and computational efficiency.

Analyzing consensus algorithms within the context of the γ-nearest ring networks is critical for understanding the efficiency and reliability of large-scale distributed networks. The special properties of the r-nearest neighbor ring offer multiple communication paths, accelerate convergence, and improve the robustness of consensus algorithms. However, this increased connectivity also introduces significant complexity in evaluating the performance of consensus algorithms, since key metrics are typically defined in terms of Laplacian eigenvalues. Especially, estimating the largest eigenvalue of the Laplacian matrix remains a major challenge for the γ-nearest neighbor ring networks. We reformulate the maximization of Laplacian eigenvalue as a minimization of the Dirichlet kernel problem. Firstly, we prove that the first and last lobes of the Dirichlet kernel are the deepest using the shift approach. Next, we apply local smoothness analysis and integer rounding arguments to demonstrate that there is at least one discrete sample to achieve a global minimum in that lobe. This study presents a rigorous analysis to precisely locate and compute the largest eigenvalue, resulting in exact analysis for key performance metrics, including convergence time, first-order network coherence, second-order network coherence, and maximum communication delay, with reduced computational complexity. In addition, our findings illustrate the effect of r in improving the performance of consensus algorithms in large-scale networks.

We investigate the robustness of the recently introduced partialsharing online federated learning (PSO-Fed) algorithm against model-poisoning attacks. To this end, we analyze the performance of the PSO-Fed algorithm in the presence of Byzantine clients, who may clandestinely corrupt their local models with additive noise before sharing them with the server. PSO-Fed can operate on streaming data and reduce the communication load by allowing each client to exchange parts of its model with the server. Our analysis, considering a linear regression task, reveals that the convergence of PSO-Fed can be ensured in the mean sense, even when confronted with model-poisoning attacks. Our extensive numerical results support our claim and demonstrate that PSO-Fed can mitigate Byzantine attacks more effectively compared with its state-of-the-art competitors. Our simulation results also reveal that, when model-poisoning attacks are present, there exists a non-trivial optimal stepsize for PSO-Fed that minimizes its steady-state mean-square error.

Distributed and collaborative machine learning over emerging Internet of Things (IoT) networks is complicated by resource constraints, device, and data heterogeneity, and the need for personalized models that cater to the individual needs of each network device. This complexity becomes even more pronounced when new devices are added to a system that must rapidly adapt to personalized models. Along these lines, we propose a networked federated meta-learning (NF-ML) algorithm that utilizes meta-learning and underlying shared structures across the network to enable fast and personalized model adaptation of newly added network devices. The NF-ML algorithm learns two sets of model parameters for each device in a distributed manner, with devices communicating only with their immediate neighbors. One set of parameters is personalized for the device-specific task, whereas the other is a generic parameter set learned via peer-to-peer communication. The performance of the proposed NF-ML algorithm was validated using both synthetic and real-world data, and the results show that it adapts to new tasks in just a few epochs, using as little as 10% of the available data, significantly outperforming traditional federated learning methods.

Autoencoders (AEs) have attracted significant attention for hyperspectral anomaly detection (HAD) in remote sensing applications due to their ability to unveil small, unique objects scattered across large geographical regions in an unsupervised manner. However, the training and inference processes of AEs are computationally demanding, posing challenges for efficient HAD in resource-constrained onboard applications. Various optimization techniques and parallel computing approaches have been proposed to alleviate the computational burden and enhance the feasibility of AEs for real-time applications in HAD. In this paper, we first present an efficient lightweight autonomous autoencoder (LAutoAE) that addresses the computational challenges of the autonomous hyperspectral anomaly detection autoencoder (AUTO-AD) while maintaining a similar anomaly detection accuracy. To further enhance the accuracy, we introduce LAutoAE+, which integrates kernel principal component analysis (KPCA) based pre-processing methods with the LAutoAE. Experiments on diverse datasets demonstrate that the proposed LAutoAE and LAutoAE+ achieve comparable or superior detection performance compared with conventional Auto-AD, while also achieving reductions of 87% and 89.4%, respectively, in the number of learnable parameters.

Personalized federated learning enables every edge device or group of edge devices within the distributed network to learn a device- or cluster-specific model tailored to their local needs. Data scarcity, however, makes it difficult to learn such individual models, resulting in performance degradation. Since the device- or cluster-specific tasks are distinct but often related, leveraging these similarities through inter-cluster learning alleviates data shortage and enhances learning performance. Although inter-cluster learning can boost performance, uncontrolled intercluster learning may lead to performance degradation due to over- or under-usage of local similarity enforcement. In light of this issue, an intelligent mechanism that performs inter-cluster learning based on device-specific needs is required. To this end, this paper proposes adopting reinforcement learning principles to control device-specific inter-cluster learning in real-time. We propose networked personalized federated learning using reinforcement learning (NPFed-RL) as a general framework and then demonstrate its feasibility by applying it to the ridge regression problem. We conduct numerical experiments to compare the proposed method with the state-of-the-art. The proposed method successfully controls device-specific parameters and offers better learning performance than existing solutions.

Online federated learning (FL) enables geographically distributed devices to learn a global shared model from locally available streaming data. Most online FL literature considers a best case scenario regarding the participating clients and the communication channels. However, these assumptions are often not met in real-world applications. Asynchronous settings can reflect a more realistic environment, such as heterogeneous client participation due to available computational power and battery constraints, as well as delays caused by communication channels or straggler devices. Further, in most applications, energy efficiency must be taken into consideration. Using the principles of partial-sharing-based communications, we propose a communication-efficient asynchronous online FL (PAO-Fed) strategy. By reducing the communication load of the participants, the proposed method renders participation more accessible and efficient. In addition, the proposed aggregation mechanism accounts for random participation, handles delayed updates, and mitigates their effect on accuracy. We study the first- and second-order convergence of the proposed PAO-Fed method and obtain an expression for its steady-state mean square deviation. Finally, we conduct comprehensive simulations to study the performance of the proposed method on both synthetic and real-life data sets. The simulations reveal that in asynchronous settings, the proposed PAO-Fed is able to achieve the same convergence properties as that of the online federated stochastic gradient while reducing the communication by 98%.

This article presents communication-efficient approaches to federated learning (FL) for resource-constrained devices with access to streaming data. In particular, we first propose a partial-sharing-based framework for online federated learning (PSO-Fed), wherein clients update local models from a stream of data and exchange tiny fractions of the model with the server, reducing the communication overhead. In contrast to classical FL approaches, the proposed strategy provides clients who are not part of a global iteration with the freedom to update local models whenever new data arrives. Furthermore, by devising a client-side innovation check, we also propose an event-triggered PSO-Fed (ETPSO-Fed) that further reduces the computational burden of clients while enhancing communication efficiency. We implement the above-mentioned frameworks in the context of kernel regression, where clients perform local learning employing random Fourier features (RFFs)-based kernel least mean squares. In addition, we examine the mean and mean-square convergence of the proposed PSO-Fed. Finally, we conduct experiments to determine the efficacy of the proposed frameworks. Our results show that PSO-Fed and ETPSO-Fed can compete with Online-Fed while requiring significantly less communication overhead. Simulations demonstrate an 80% reduction in PSO-Fed and an 84.5% reduction in ETPSO-Fed communication overhead compared to Online-Fed. Notably, the proposed PSO-Fed strategies show good resilience against model-poisoning attacks without involving additional mechanisms.

Numerous real-life systems exhibit complex nonlinear input-output relationships. Kernel adaptive filters, a popular class of nonlinear adaptive filters, can efficiently model these nonlinear input-output relationships. Their growing network structure, however, poses considerable challenges in terms of their hardware implementation, making them inefficient for real-time applications. Random Fourier features (RFF) facilitate the development of kernel adaptive filters with a fixed network structure. For the first time, this paper attempts to implement the RFF-based kernel least mean square (RFF-KLMS) algorithm on hardware. To this end, we propose several reformulations of the feature functions (FFs) that are computationally expensive in their native form so that they can be implemented in real-time VLSI. Specifically, we reformulate inner product evaluation, cosine, and exponential functions that appear in the implementation of FFs. With these reformulations, the proposed delayed RFF-KLMS (DRFF-KLMS) is then synthesized using 45-nm CMOS technology with 16-bit fixed-point representations. According to the synthesis results, pipelined DRFF-KLMS architectures require minimal hardware increase over the state-of-the-art conventional delayed LMS architecture while significantly improving estimation performance for the nonlinear model. Our results suggest that the cosine feature function-based DRFF-KLMS is appropriate for applications requiring high accuracy, whereas the exponential function-based DRFF-KLMS may be well suited for resource-constrained applications.

Federated learning (FL) literature typically assumes that each client has a fixed amount of data, which is unrealistic in many practical applications. Some recent works introduced a framework for online FL (Online-Fed) wherein clients perform model learning on streaming data and communicate the model to the server; however, they do not address the associated communication overhead. As a solution, this paper presents a partial-sharing-based online federated learning framework (PSO-Fed) that enables clients to update their local models using continuous streaming data and share only portions of those updated models with the server. During a global iteration of PSO-Fed, non-participant clients have the privilege to update their local models with new data. Here, we consider a global task of kernel regression, where clients use a random Fourier features-based kernel LMS on their data for local learning. We examine the mean convergence of the PSO-Fed for kernel regression. Experimental results show that PSO-Fed can achieve competitive performance with a significantly lower communication overhead than Online-Fed.

Many assumptions in the federated learning literature present a best-case scenario that can not be satisfied in most real-world applications. An asynchronous setting reflects the realistic environment in which federated learning methods must be able to operate reliably. Besides varying amounts of non-IID data at participants, the asynchronous setting models heterogeneous client participation due to available computational power and battery constraints and also accounts for delayed communications between clients and the server. To reduce the communication overhead associated with asynchronous online federated learning (ASO-Fed), we use the principles of partial-sharing-based communication. In this manner, we reduce the communication load of the participants and, therefore, render participation in the learning task more accessible. We prove the convergence of the proposed ASO-Fed and provide simulations to analyze its behavior further. The simulations reveal that, in the asynchronous setting, it is possible to achieve the same convergence as the federated stochastic gradient (Online-FedSGD) while reducing the communication tenfold.

From the inaugural steps of McCulloch and Pitts to put forth a composition for an electrical brain, that combined with the conception of an adaptive leaning mechanism by Widrow and Hoff has given rise to the phenomena of intelligent machines, machine learning techniques have gained the status of a miracle solution in a myriad of scientific fields. At the heart of these techniques lies iterative optimisation processes that are derived based on first, and in some cases, second-order derivatives. This manuscript, however, aims to expand the mentioned framework to that of using fractional-order derivatives. The entire format of adaptation is revised form the perspective of fractional-order calculus and the appropriate framework for taking full advantage of the fractional-order calculus in learning and adaptation paradigms is formulated. For rigour, the structure of behavioural analysis and performance prediction of this novel class of learning machines is also forged.

This article proposes an optimal scheduling policy for a system where spatiotemporally dependent sensor observations are broadcast to remote estimators over a resource-limited broadcast channel. We consider a system with a measurement-blind network scheduler that transmits observations, and design scheduling schemes that minimize mean squared error (MSE) by determining a subset of sensor observations to be broadcast based on their information freshness, as measured by their Age of Information (AoI). By modeling the problem as a finite state-space Markov decision process (MDP), we derive an optimal scheduling policy, with AoI as a state variable, minimizing the average MSE for an infinite time horizon. The resulting policy has a periodic pattern that renders an efficient implementation with low data storage. We further show that for any policy that minimizes the overall AoI, the estimation accuracy depends on how the scheduling order relates to the sensor’s intrinsic spatial correlation. Consequently, the estimation accuracy varies from worse than a randomized scheduling approach to near optimal. Thus, we present an additional age-minimizing policy with optimal scheduling order. We also present alternative policies for large state spaces that are attainable with less computational effort. Numerical results validate the presented theory.

We develop a privacy-preserving distributed algorithm to minimize a regularized empirical risk function when the first-order information is not available and data is distributed over a multi-agent network. We employ a zeroth-order method to minimize the associated augmented Lagrangian function in the primal domain using the alternating direction method of multipliers (ADMM). We show that the proposed algorithm, named distributed zeroth-order ADMM (D-ZOA), has intrinsic privacy-preserving properties. Most existing privacy-preserving distributed optimization/estimation algorithms exploit some perturbation mechanism to preserve privacy, which comes at the cost of reduced accuracy. Contrarily, by analyzing the inherent randomness due to the use of a zeroth-order method, we show that D-ZOA is intrinsically endowed with$(\epsilon,\delta)-$differential privacy. In addition, we employ the moments accountant method to show that the total privacy leakage of D-ZOA grows sublinearly with the number of ADMM iterations. D-ZOA outperforms the existing differentially-private approaches in terms of accuracy while yielding similar privacy guarantee. We prove that D-ZOA reaches a neighborhood of the optimal solution whose size depends on the privacy parameter. The convergence analysis also reveals a practically important trade-off between privacy and accuracy. Simulation results verify the desirable privacy-preserving properties of D-ZOA and its superiority over the state-of-the-art algorithms as well as its network-wide convergence.

Conventional adaptive filters, which assume Gaussian distribution for signal and noise, exhibit significant performance degradation when operating in non-Gaussian environments. Recently proposed fractional-order adaptive filters (FoAFs) address this concern by assuming that the signal and noise are symmetric$\alpha $-stable random processes. However, the literature does not include any VLSI architectures for these algorithms. Toward that end, this article develops hardware-efficient architecture for fractional-order correntropy adaptive filter (FoCAF). We first reformulate the FoCAF for its efficient real-time VLSI implementation and then demonstrate that these reformulations cause negligible performance degradation under the 16-bit fixed-point implementation. Using this reformulated algorithm, we design an FoCAF architecture. Furthermore, we analyze the critical path of the design to select the appropriate level of pipelining based on the sampling rate of the application. According to the critical-path analysis, the FoCAF design is pipelined using retiming techniques to obtain delayed FoCAF (DFoCAF), which is then synthesized using$\mathbf {45}$-nm CMOS technology. Synthesis results reveal that DFoCAF architecture requires a minimal increase in hardware over the prominent least mean square (LMS) filter architecture and achieves a significant increase in the performance in symmetric$\alpha $-stable environments where LMS fails to converge.

This work proposes an efficient batch-based implementation for kernel regression on graphs (KRG) using random Fourier features (RFF) and a low-complexity online implementation. Kernel regression has proven to be an efficient learning tool in the graph signal processing framework. However, it suffers from poor scalability inherent to kernel methods. We employ RFF to overcome this issue and derive a batch-based KRG whose model size is independent of the training sample size. We then combine it with a stochastic gradient-descent approach to propose an online algorithm for KRG, namely the stochastic-gradient KRG (SGKRG). We also derive sufficient conditions for convergence in the mean sense of the online algorithms. We validate the performance of the proposed algorithms through numerical experiments using both synthesized and real data. Results show that the proposed batch-based implementation can match the performance of conventional KRG while having reduced complexity. Moreover, the online implementations effectively learn the target model and achieve competitive performance compared to the batch implementations.

Recent years have bore witness to the proliferation of distributed filtering techniques, where a collection of agents communicating over an ad-hoc network aim to collaboratively estimate and track the state of a system. These techniques form the enabling technology of modern multi-agent systems and have gained great importance in the engineering community. Although most distributed filtering techniques come with a set of stability and convergence criteria, the conditions imposed are found to be unnecessarily restrictive. The paradigm of stability and convergence in distributed filtering is revised in this manuscript. Accordingly, a general distributed filter is constructed and its estimation error dynamics is formulated. The conducted analysis demonstrates that conditions for achieving stable filtering operations are the same as those required in the centralized filtering setting. Finally, the concepts are demonstrated in a Kalman filtering framework and validated using simulation examples.

This paper proposes an optimal scheduling policy for a remote estimation problem, where sensor observations of two spatio-temporally correlated processes are broadcasted to two remote estimators. At each time instant only a single observation can be communicated. For this purpose, a system scheduler determines which sensor measurement is communicated. The scheduler cannot observe measurements, and exploits age-of-information (AoI) to calculate the expected estimation error. We derive an optimal scheduling policy, with AoI as state-variable, that minimizes the average mean squared error for an infinite time horizon. The obtained policy yields a periodic scheduling of the sensor measurements, and we show that the AoI for the process with the largest marginal variance does not exceed one.

This work revisits the problem of distributed adaptive filtering in multi-agent sensor networks. In contrast to classical approaches, the formulation relaxes the Gaussian assumption on the signal and noise to the generalized setting of α-stable distributions that do not possess second- and higher-order statistical moments. Most importantly, the considered scenario allows for different characteristic exponents throughout the network. Drawing upon ideas from correntropy-type local similarity measures and fractional-order calculus, a novel class of distributed fractional-order correntropy adaptive filters, that are robust against the jittery behavior of α-stable signals, is derived and their convergence criterion is established. The effectiveness of the proposed algorithms, as compared to existing distributed adaptive filtering techniques, is demonstrated via simulation examples.

We propose a privacy-preserving distributed maximum consensus algorithm where the local state of the agents and identity of the maximum state owner is kept private from adversaries. To that end, we reformulate the maximum consensus problem over a distributed network as a linear program. This optimization problem is solved in a distributed manner using the alternating direction method of multipliers (ADMM) and perturbing the primal update step with Gaussian noise. We define the privacy of an agent as the estimation error of its local state at the adversary and obtain theoretical bounds on the privacy loss for the proposed method. Further, we prove that the proposed algorithm converges to the maximum value at all agents. In addition to the analytical results, we illustrate the convergence speed and privacy-accuracy trade-off through numerical simulations.

We propose a new distributed algorithm to solve the total least-squares (TLS) problem when data are distributed over a multi-agent network. To develop the proposed algorithm, named distributed ADMM TLS (DA-TLS), we reformulate the TLS problem as a parametric semidefinite program and solve it using the alternating direction method of multipliers (ADMM). Unlike the existing consensus-based approaches to distributed TLS estimation, DA-TLS does not require careful tuning of any design parameter. Numerical experiments demonstrate that the DA-TLS converges to the centralized solution significantly faster than the existing consensus-based TLS algorithms.

In this paper we propose a genie-aided strategy to optimize the use of renewable energy (RE) in a community of households with shared access to storage and RE generation facilities. The households are spread over a limited geographical area, and are subject to different time-varying power consumption profiles, and energy prices. We consider a finite number of RE generators and energy storage devices (ESDs), which are deployed in specific locations. The proposed strategy seeks to minimize the energy cost incurred by the participating households by optimizing the rate at which RE is consumed over time. Our model takes into account the power loss incurred in the transmission of energy from the generators to the loads. The optimization problem is cast as a non-convex quadratically constrained quadratic program, which is simplified in order to derive an approximate solution. Numerical results show that transmission losses and differences across price and load can significantly affect the optimal RE allocation among the households. The proposed strategy offers valuable insights for energy planning purposes and can be used to devise real-time RE management algorithms by incorporating the necessary forecasting techniques.

In order to accommodate for modern adaptive filtering applications, the classic adaptive filtering paradigm is considered from a more general perspective. The new formulation allows for time dependent variations in the state of the system and more importantly it relaxes the Gaussian assumption to the generalized setting of α-stable distributions. In this work, based on the principles of gradient descent and fractional-order calculus, a cost-effective technique for tracking the state of such a system is derived. For rigour, performance of the derived filtering technique is analyzed and convergence conditions are established.

Internet of Things technologies will enable smart energy planning, which in turn will expedite the adoption of renewable energy (RE). In this paper, we propose a mathematical framework to optimize the use of RE in a shared solar environment featuring households with access to several RE generators. We consider location and time-dependent electricity prices, and formulate an optimization problem to minimize the energy cost incurred by the households over a finite planning horizon. The proposed framework accounts for transmission losses and battery inefficiencies. We then proposed two approaches to solve the formulated optimization problem. The first approach is based on quadratic programming, and is used to obtain a precision-controllable solution, requiring discretization in time and convex relaxation. The second approach is based on variational methods, which are used to tackle the problem directly in continuous time, thus obtaining a solution in closed form after introducing reasonable simplifications. To ensure full cooperation, we finally derive a fair energy allocation policy, which allocates RE to each household in proportion to its capital investment. The obtained analytical results allow us to evaluate the relationship between achievable performance, RE production, transmission losses, and price variability. Extensive simulations are used to verify the derived analytical results, illustrate the characteristics of the proposed strategies and compare their achievable performance.

A nonlinear adaptive filtering framework for processing complex-valued signals is derived. The introduced adaptive filter extends the fractional-order framework of the authors for dealing with real-valued signals to the complex domain via the augmented statistical approach to complex-valued signal processing. This results in a versatile class of adaptive filtering techniques, which allows the classical Gaussian assumption to be extended to the generalized context of α-stables. For rigor, the performance of the introduced adaptive filtering framework is analyzed, its convergence criteria is established, and its application in tracking signals of chaotic systems is demonstrated using simulations.

In this work, a distributed Kalman filtering and clustering framework for sensor networks tasked with tracking multiple state vector sequences is developed. This is achieved through recursively updating the likelihood of a state vector estimation from one agent offering valid information about the state vector of its neighbors, given the available observation data. These likelihoods then form the diffusion coefficients, used for information fusion over the sensor network. For rigour, the mean and mean square behavior of the developed Kalman filtering and clustering framework is analyzed, convergence criteria are established, and the performance of the developed framework is demonstrated in a simulation example.

A cost-effective framework for distributed adaptive filtering of α-stable signals over sensor networks is proposed. First, the filtering paradigm of α-stable signals through multiple observations made over a network of sensors is revisited and an optimal solution is formulated. Then, an adaptive gradient descent based algorithm for distributed real-time filtering of α-stable signals via multiagent networks is derived. This not only provides an approximation of the formulated optimal solution, but also a cost-effective algorithm that scales with the size of the network. Moreover, performance of the derived algorithm is analyzed and convergence conditions are established.

Summary While the amount of information steadily increases, the requirements on the response time to query these information become more strict. Under those conditions, conventional database systems reach their limits and cannot meet these performance requirements anymore. In recent years, systems with many processing cores are considered to satisfy these demands. Furthermore, these systems include more and more heterogeneous cores tailor‐made to solve one specific task in an efficient manner. However, dedicated hardware accelerators are inflexible and cannot be adapted to the requirements of a dedicated query. Thus, the challenge is orchestrating the diversity of the functionality of all the cores to be optimized for performance/energy efficiency. In this paper, a concept is introduced on how to develop a flexible Field‐Programmable Gate Arrays (FPGA)‐based hardware accelerator to improve the performance of query evaluation in a Semantic Web database. As a first step to the hardware/software system, several joint algorithms are implemented on an FPGA and evaluated against a well‐developed software solution (implemented in C). The comparison shows a significant speedup of up to 10 times. Because of the complexity of the join operator, it is promising that the overall performance of query evaluation can be further enhanced by processing whole queries on an FPGA. Copyright © 2015 John Wiley & Sons, Ltd.

We develop a fully-distributed iterative algorithm for finding a model-distributed least-squares solution of systems of linear equations over sensor networks. Here, model-distributed means the solution vector is distributed across the network rather than being replicated at each node. For this purpose, we devise a dual regularized least-squares problem via a suitable decomposition of the normal equations associated with the original problem. The resultant dual problem can be solved in a fully-decentralized and iterative manner by means of the diffusion-based Pareto optimization strategy. We verify the usefulness of the proposed algorithm via both theoretical analysis and numerical examples.

In this paper, we consider the problem of channel and phase noise estimation for an orthogonal frequency division multiplexing (OFDM) radio link. We solve this problem by first investigating the subspace in which the phase noise spectral vector lies and then exploiting this information during estimation. Building upon earlier works, the phase noise spectral estimate is obtained by minimizing a homogeneous quadratic cost function and the channel estimate in turns depends on the obtained phase noise estimate. We show that, at infinite signal-to-noise ratio, the true phase noise spectral estimate lies in the null space of the matrix associated with the cost function. We utilize this knowledge by imposing constraints that adhere to this null space when minimizing the cost function. In addition, we also propose constraints based on knowledge of the covariance matrix of the phase noise process. Through simulations, we demonstrate lower phase noise mean-square error (MSE) and consequently lower channel MSE when incorporating the subspace information.

Simultaneous reception and transmission in the same frequency, so-called full-duplex operation, causes an infinite feedback loop in an amplify-and-forward relay. The unwanted echoes may result in oscillation at the relay, making it unstable, and distorting the spectrum. This paper presents an adaptive MIMO filtering method for full-duplex amplify-and-forward relays that aims at solving the joint problem of self-interference mitigation and equalization of the source-relay channel. The scheme exploits the knowledge of the autocorrelation of the transmitted signal as the only side information while allowing the relay, in the best case, to implement precoding as if there was not any self-interference or frequency selectivity in the source-relay channel. Finally, the proposed adaptation algorithm is investigated by determining its stationary points and by performing simulations in a MIMO-OFDM framework.

We examine the problem of estimating the frequency of a three-phase power system in an adaptive and low-cost manner when the voltage readings are contaminated with observational error and noise. We assume a widely-linear predictive model for the αβ complex signal of the system that is given by Clarke's transform. The system frequency is estimated using the parameters of this model. In order to estimate the model parameters while compensating for noise in both input and output of the model, we utilize the notions of total least-squares fitting and gradient-descent optimization. The outcome is an augmented gradient-descent total least-squares (AGDTLS) algorithm that has a computational complexity comparable to that of the complex least mean square (CLMS) and the augmented CLMS (ACLMS) algorithms. Simulation results demonstrate that the proposed algorithm provides significantly improved frequency estimation performance compared with CLMS and ACLMS when the measured voltages are noisy and especially in unbalanced systems.

We develop a gradient-descent distributed adaptive estimation strategy that compensates for error in both input and output data. To this end, we utilize the concepts of total least-squares estimation and gradient-descent optimization in conjunction with a recently-proposed framework for diffusion adaptation over networks. The proposed strategy does not require any prior knowledge about the noise variances and has a computational complexity comparable to the diffusion least mean square (DLMS) strategy. Simulation results demonstrate that the proposed strategy provides significantly improved estimation performance compared with the DLMS and bias-compensated DLMS (BC-DLMS) strategies when both the input and output signals are noisy.

Demand-side management will play a crucial role in balancing the energy generation and demand in future smart grids. In this paper, game-theoretic demand-side management algorithms are proposed for energy consumption scheduling under load uncertainty. The demand-side optimization and scheduling problem is formulated as a noncooperative cost minimization game among the endusers and an iterative algorithm that averages over the load uncertainty is proposed for solving it. The proposed algorithm is proven to converge to a Nash equilibrium. Simulation results show that taking into account the uncertainty in the load reduces significantly the load peak-to-average ratio and the hourly variation of the aggregate load profile.

This paper addresses the problem of correlation due to redundancy in cooperative spectrum sensing networks and proposes an algorithm for topology design which improves significantly detection performance. In a recently proposed two-step distributed scheme, redundancy occurs when some nodes contribute more than once for the consensus decision, leading to correlation and consequently degrading performance in the same way as correlated shadowing. To eliminate this type of correlation, we employ two different topologies, primary and complementary, one for each cooperation step. Topology design is accomplished in a distributed manner by stating criteria for user selection. Results show that the proposed double-topology scheme suppresses redundancy and offers similar performance when compared to the case of independent node contributions.

This paper proposes a novel two-step distributed detection scheme for cooperative spectrum-sensing networks. In the first step, individual contributions from nodes within a neighborhood are fused through an adaptive combiner, which updates the weights and makes local decisions iteratively. In the second step, local decisions are shared within a neighborhood to yield a consensus decision. Results are presented in terms of complementary receiver operating characteristic curves and show the good behavior of the proposed scheme when compared to the optimal linear fusion rule, even if correlated node contributions are considered.

We derive the average capacity for an OFDM radio link that is impaired by phase noise and carrier frequency offset at both transmitter and receiver side. It is well known that, for OFDM, phase noise and frequency offset cause intercarrier interference (ICI). We first notice the dependency of the instantaneous capacity on the ICI power and on the fading channel. Using a Taylor series approximation, we show that the ICI power can be well represented in a quadratic form whose probability density function we derive. Using the derived distribution and assuming a Rayleigh-fading channel, we derive analytical expressions of average capacity. Different test cases are presented to study the sensitivity of the capacity on phase noise and frequency offset. The main observations suggest that, for reasonable phase noise levels, the capacity is practically insensitive to the frequency offset.

We propose a discrete-time model for phase noise processes occurring in phase locked loop (PLL) based oscillators. By using linear time-invariant analysis in the phase domain, we arrive at a parallel auto-regressive moving-average phase noise model that captures the major noise sources and corresponding loop transfer functions of the PLL circuitry. The acquired model allows us to derive analytically the distribution of the inter-carrier interference (ICI) power in the receiver, shown to be a sum of correlated gamma random variables with a model-specific covariance matrix. Finally, we utilize the derived ICI distribution and proceed to analyze the effects of phase noise and multipath fading on the performance of OFDM radio links. For example, we evaluate the rate loss without ICI compensation or knowledge at the receiver, and illustrate the trade-off introduced by phase noise between the OFDM symbol length and the overhead caused by the cyclic prefix. Throughout the paper, simulations confirm the accuracy of the analytical expressions.

Predistortion techniques are implemented in OFDM systems to improve power amplifier efficiency while keeping in-band and out- of-band distortion at acceptable levels. Predistorter (PD) design for broadband MIMO-OFDM systems introduces several implementation challenges. Coupling effects cannot be avoided in MIMO transceivers where the paths are implemented on the same chipset. Moreover, baseband PDs are affected by imbalances in the IQ modulator. This paper proposes a MIMO-PD that linearizes the power amplifier (PA) response and compensates for both crosstalk and IQ imbalance. Error vector magnitude (EVM) and adjacent channel power ratio (ACPR) results show a significant improvement compared with conventional PD systems.

While the paradigm of service oriented architectures and also the concept of service migration has been adapted to wireless sensor networks recently, the consistency demands of stateful service migration have not yet been considered in depth. Many services, for example for the aggregation of data or alarm services, require strict consistency when being migrated from one sensor node to another. In previous papers, we described how atomic commit protocols can be used to guarantee strict consistency of migrated services. Additional communication is required to achieve strict consistency. Therefore we varied the network conditions (i.e. loss rate, density) to determine the best solution in different situations. Finally we designed an adaptive approach which uses the ideal combination of transaction protocol and routing protocol to reduce the communication overhead and save energy to increase the lifetime of wireless sensor network deployments.

We analyze the effect of Wiener phase noise on the capacity and signal-to-interference-plus-noise (SINR) ratio. Our analysis includes phase noise at the transmitter and receiver end of an OFDM communication link. We see that the capacity and SINR are random variables whose distribution depends on the phase noise and the fading channel. Using a Taylor series approximation, we show that the random variable, characterizing the phase noise, in the performance metrics can be expressed as a sum of correlated gamma variables with rank-deficient square-root normalized covariance matrix. The approximation holds well when the ratio between the subcarrier spacing and the 3dB bandwidth of the oscillator power spectral density is at least one order of magnitude, which for most practical oscillators is the case. In earlier literature, the probability density function of a sum of correlated gamma variables with full-rank square-root normalized covariance matrix was derived. We extend these results to the rank-deficient case and apply them to the random variable of our case. With the probability density functions characterizing the phase noise and the fading channel at hand, we proceed to obtain closed-form statistical measures of capacity and SINR. The simulations show the good agreement with our analytical expressions.

Focusing on two-antenna infrastructure relays employed for coverage extension, we develop hybrid techniques that switch opportunistically between full-duplex and half-duplex relaying modes. To rationalize the system design, the classic three-node full-duplex relay link is first amended by explicitly modeling residual relay self-interference, i.e., a loopback signal from the transmit antenna to the receive antenna remaining after cancellation. The motivation for opportunistic mode selection stems then from the fundamental trade-off determining the spectral efficiency: The half-duplex mode avoids inherently the self-interference at the cost of halving the end-to-end symbol rate while the full-duplex mode achieves full symbol rate but, in practice, suffers from residual interference even after cancellation. We propose the combination of opportunistic mode selection and transmit power adaptation for maximizing instantaneous and average spectral efficiency after noting that the trade-off favors alternately the modes during operation. The analysis covers both common relaying protocols (amplify-and-forward and decode-and-forward) as well as reflects the difference of downlink and uplink systems. The results show that opportunistic mode selection offers significant performance gain over system design that is confined to either mode without rationalization.

Nonlinear effects of a power amplifier (PA) in communications systems include spectral spreading, which causes adjacent channel interference. For predistorter design, it is crucial to employ a model that can sufficiently model the intermodulation distortion (IMD) caused by the PA. The order of IMD generated by a polynomial is determined by its highest order basis. For a piecewise linear function, the order of nonlinearity cannot be directly deduced from its basis function. This paper investigates how IMD is introduced by the simplicial canonical piecewise linear (SCPWL) function. We examine the operation imposed by the SCPWL basis function on sinusoid and multi-tone signal in the time domain and frequency domain. Our analysis results in expressions that can be used to predict the output spectrum of the SCPWL basis. The SCPWL basis operation involves spreading of the input signal spectrum. The output spectrum of each SCPWL basis includes infinite IMD components. The IMD order of the SCPWL model is not determined by the highest order basis. Instead, the number of SCPWL parameters, i.e., the number of linear affine segments that defines the function, determines how accurately it can model the IMD components.

We study the fundamental rate-interference trade-off between full-duplex and half-duplex transmission modes in a two-hop infrastructure-based decode-and-forward relay link. First, we derive closed-form expressions for average end-to-end capacities. Then, we show that it may be better to tolerate some loop interference in the full-duplex mode than to halve the end-to-end symbol rate by allocating two orthogonal time slots in the half-duplex mode. Furthermore, we evaluate the loop interference power levels below which the full-duplex mode achieves better capacity than the half-duplex mode. Our results indicate that the full-duplex mode is preferable in terms of capacity for many practical signal-to-noise ratios, e.g., above 3-10dB depending on the setup. Finally, we show that full-duplex relays can further improve the end-to-end capacity by adjusting transmit power.

Direct-conversion OFDM transceivers are seriously affected by front-end distortions like IQ imbalance and phase noise. Moreover, OFDM systems are sensitive to nonlinear distortions in power amplifiers and A/D converters, and to carrier frequency offset (CFO). We present a novel baseband compensation technique that compensates these main impairments in OFDM systems. Unlike compensation scheme that target a subset of impairments, our method jointly mitigates the effects of IQ imbalance, phase noise, carrier frequency offset and nonlinear distortion. Simulations of the proposed scheme with large levels of CFO, IQ imbalance and phase noise show an excellent performance in terms of bit error rate.

This paper studies the performance of a two-hop infrastructure-based relay link. The relay operates in a half-duplex mode using an amplify-and-forward protocol. Considering transmission of OFDM signals with nonlinear power amplifiers (PAs), we determine exact closed-form expressions for end-to-end bit error probabilities (BEPs). The expressions allow us to analyze the effect of nonlinear distortion as well as the special case of linear PAs. Furthermore, we derive asymptotic BEP floor values which illustrate how the end-to-end performance is limited by both the nonlinear PA distortion and the weakest hop. Our discussion indicates that the influence of the nonlinear distortion can be mitigated by properly adjusting the PA back-offs.

We study the physical-layer performance of a wideband multihop system in which a chain of amplify-and-forward relays is deployed for routing data from a source to a destination. The novelty of the system model w.r.t. prior work is in assuming frequency-selective multipath channels for all hops. We propose a new approach in which the channels are specified in terms of clustered exponentially decaying power-delay profiles (PDPs). Focusing on relays that apply time-domain signal processing, the distinctive form of the end-to-end source-destination PDP motivates us to introduce a new hypoexponential PDP and study it in detail. Then, we analyze the performance of orthogonal frequency division multiplexing (OFDM) transmission over the multihop relay link. The type of the adopted analysis is suitable mainly for investigating the typical time-domain issues of OFDM transmission such as excessive multipath delay spread, non-ideal time syncronization, and interference due to insufficient cyclic prefix duration. In particular, we derive new closed-form expressions for evaluating the signal-to-interference and noise ratio. Our analysis confirms that multihopping increases considerably the delay spread of the end-to-end PDP and makes the channel more frequency selective. This imposes significant effect on the design of receiver time synchronization and pilot structures, and on the choice of OFDM physical layer parameters.

We consider a downlink OFDMA transmission system in which an infrastructure-based relay node is deployed for extending the coverage of a base station. Focusing on decode-and-forward (DF) operation, we study how different relay functionalities affect the system performance. The functionalities include fixed or adaptive subcarrier pairing, information redistribution, buffering, and adjustment of the time shares allocated for the two hops. In particular, it is of great interest for system design to be able to evaluate the performance gains due to each functionality, because they come at the expense of increased complexity and cost of the relay architecture. The main contribution of the paper is to analyze the performance of the proposed protocols by deriving new closed-form expressions for the average end-to-end capacity. The calculated expressions facilitate the performance comparison between the different functionalities.

This paper proposes selective update and cooperation strategies for parameter estimation in distributed adaptive sensor networks. A set-membership filtering approach is employed that results in reduced complexity for updating parameter estimates at each network node, a significant reduction in information exchange between cooperating nodes, and an optimal strategy to obtain consensus estimates. The proposed strategies and the estimation algorithm offer a new way to explore cooperation in adaptive distributed sensor networks.

We study the performance of a two-hop full-duplex OFDM relay link, where the relay, or in-band repeater, amplifies and forwards its input signal on the same channel as the main transmitter. We derive new closed-form expressions for the end-to-end signal-to-interference and noise ratio (SINR) when the multipath channels are modeled with clustered exponential power-delay profiles. Our system model incorporates the deleterious effect of loop interference associated with full-duplex relays due to signal leakage from transmission to reception. In particular, our analysis shows that the relay gain is a central parameter in the mitigation of residual loop interference and for the optimization of the end-to-end performance.

This paper studies receiver techniques for nonlinear amplifier distortion compensation in an OFDM relay-assisted cooperative communication system. The system model includes a nonlinear amplifier at the amplify-and-forward relay, modeled as a solid state power amplifier (SSPA). A maximum ratio combiner (MRC) is introduced that includes the effects of the amplifier distortions. The MRC is obtained by a proper modeling of the nonlinear distortion noise. Furthermore, we introduce a power amplifier nonlinearity cancellation (PANC) technique that is suitable for cooperative systems. Our simulation results confirm that the new MRC and PANC techniques offer substantial performance improvements if the relayed signal is subject to nonlinear distortion.

We study the fundamental capacity trade-off between full-duplex and half-duplex transmission modes in a two- hop communication system with a fixed infrastructure-based amplify-and-forward relay. First, we derive closed-form expressions for the average end-to-end capacity in the relay link. We show that it may be better to tolerate some loop interference with the full-duplex mode than to consume channel resources by allocating two orthogonal channels with the half-duplex mode. Furthermore, we evaluate the maximum loop interference power levels that still allow the full-duplex mode to achieve the same capacity as the half-duplex mode. Our results indicate that with practical signal-to-noise ratio values, the full-duplex mode is preferable in terms of capacity.

This paper considers usage of a single-frequency mode for cell coverage extension in infrastructure-based relay links. We provide new closed-form expressions for outage probability by taking into account practical constraints such as interference due to frequency reuse and signal leakage between the relay transmitter and receiver. The analysis covers both decode-and-forward and amplify-and-forward protocols both in downlink and in uplink. For amplify-and-forward relaying, variable gain and fixed gain methods for transmit power normalization are discussed. Simulations show excellent agreement with theory and confirm that the single-frequency mode can be applied at the cost of tolerable signal degradation.

This letter derives new gain control schemes for an amplify-and-forward single-frequency relay link in which loop interference from the relay transmit antenna to the relay receive antenna has to be tolerated. The proposed gain control schemes take into account the effect of residual loop interference that remains after imperfect loop interference cancellation. As a result of our gain control strategy, the signal-to-interference and noise ratio can be maximized while, at the same time, transmit power is decreased. Finally, we evaluate system performance by deriving closed-form outage probability expressions for the gain control schemes.

This paper proposes a clustering approach to parameter estimation in distributed sensor networks. The proposed approach is an alternative to the conventional centralized and decentralized approaches. This is made possible by the unique adaptive estimation architecture, U-SHAPE, stemming from set-membership adaptive filtering. At the expense of a slightly degraded mean-square error performance (comparing to the least-squares approach), the proposed approach offers improved data processing flexibility in a distributed sensor network, reduced signal processing hardware and reduced communication bandwidth and power requirements.

QR-decomposition-based least-squares lattice (QRD-LSL) algorithms do not provide the transversal weight vector in explicit form. These weights can be computed from the variables of the QRD-LSL algorithm using the Levinson-Durbin (LD) recursion. If the prediction coefficients do not vary over time, a reduced complexity but approximate solution can be obtained. Nonetheless, this approximate solution requires algorithm convergence and infinite memory support (forgetting factor equal to one). To obtain the exact weights at any time instant and for any choice of the forgetting factor, the computational complexity of the true LD recursion increases by an order of magnitude. In this letter, we show that an exact solution can be obtained with a reduced computational complexity and without any added restriction. Simulation results show that the solutions obtained using the proposed method and the exact LD recursion are the same up to the precision used, whereas the weights from the approximate method always deviate from the true solution.

This paper proposes a novel split predistorter structure to remove nonlinear distortion caused by nonlinear power amplifier with memory. Unlike conventional techniques, the new technique does not require an estimation of the memory model at the transmitter leading to a reduced implementation complexity. Simulations have been carried out using the SUI3 channel model developed for IEEE802.16 standard. The results verify the good performance of the proposed predistorter for several relevant power amplifier models.

This work proposes a new partial update filtering technique tailored for adaptive transmit beamforming with low feedback rate. A signal dependent selection rule is derived that singles out one component of the beamforming vector to be updated. This provides an efficient way to perform the update, which escalates the performance with the number of available feedback bits without increasing the computational complexity. An operation count for the proposed algorithm and existing solutions is provided. Simulations show that the proposed scheme outperforms the tracking capabilities of existing codebook solutions in low mobility scenarios, while having comparable bit error probability performance and a computational complexity reduction of 90%.

Fast QR decomposition RLS (FQRD-RLS) algorithms are well known for their good numerical properties and low computational complexity. However the FQRD-RLS algorithms do not provide access to the filter weights, and so far their use has been limited to problems seeking an estimate of the output error signal. In this paper we present a novel technique to obtain the filter weights of the FQRD-RLS algorithm at any time instant. As a consequence, we extend the range of applications to include problems where explicit knowledge of the filter weights is required. The proposed weight extraction technique is tested in a system identification setup. The results verify our claim that the extracted coefficients of the FQRD-RLS algorithm are identical to those obtained by any RLS algorithm such as the inverse QRD-RLS algorithm

This paper proposes a combined frequency and time domain channel estimation method for MIMO OFDM systems. Initial channel estimation is performed by first estimating the channel response in frequency domain, exploiting a set of dedicated pilot carriers, followed by an interpolation step. In order to reduce the interpolation error and improve the bit error rate performance, we propose to refine the channel estimates in time domain using the equalized signal from the frequency domain processing. Simulations have been carried out using the spatial channel model proposed under the 3GPP framework. The proposed method has been tested in a wide range of mobile speeds in conjunction with several standard MIMO equalizers. The results show that the proposed estimator outperforms widely used time and frequency domain channel estimation approaches

This paper proposes a new data-selective affine projection algorithm for echo cancellation. The algorithm generalizes the concepts of the conventional set-membership affine-projection by incorporating a lower bound on the output error in order to prevent undesirable attenuation of the far-end signal. It is shown that the echo signal can more reliably be removed from the far-end user signals by employing the new algorithm in double talk situations. In addition, the proposed algorithm retains the fast convergence of the conventional SM-AP algorithm while keeping a reduced number of coefficient updates. Simulation results, using the ITU-T G.168 recommendation setup parameters, are presented in order to confirm the good features of the proposed algorithm.

Fast QR decomposition RLS (FQRD-RLS) algorithms are well known for their good numerical properties and low computational complexity. The FQRD-RLS algorithms do not provide access to the filter weights, and their uses have so far been limited to problems seeking an estimate of the output error signal. In this paper we present techniques which allow us to reproduce the equivalent output signal corresponding to any input-signal applied to the weight vector of the FQRD-RLS algorithm. As a consequence, we can extend the range of applications of the FQRD-RLS to include problems where the filter weights are periodically updated using training data, and then used for fixed filtering of a useful data sequence, e.g., burst-trained equalizers. The proposed output-filtering techniques are tested in an equalizer setup. The results verify our claims that the proposed techniques achieve the same performance as the inverse QRD-RLS algorithm at a much lower computational cost.

This paper proposes a data-selective affine projection algorithm. The algorithm generalizes the ideas of the conventional set-membership affine-projection (SM-AP) algorithm to include a variable data-reuse factor. By utilizing the information provided by the data-dependent step size, we propose an assignment rule that automatically adjust the number of data reuses. A particular reduced-complexity implementation of the proposed algorithm is also considered in order to reduce the dimensions of the matrix inversions involved in the computation of update. Simulations show that a significant reduction in the overall complexity can be obtained with the algorithm as compared with the conventional SM-AP algorithm. In addition, the proposed algorithm retains the fast convergence of the conventional SM-AP algorithm, and the low steady-state error of the SM-NLMS algorithm.

This work proposes a novel adaptive transmit beamforming algorithm, suitable for single-beam multiple-input multiple-output (MIMO) systems employing closed-loop feedback channels with low rate. A low computational complexity solution is achieved by means of a partial update strategy, in which only one of the complex baseband weights of the transmit antenna array is modified at each instance. The new algorithm can operate sequentially over the antennas, or employ a signal-dependent criterion that determines which antenna coefficient to update. The solution for the weight under consideration maximizes the signal-to-noise ratio at the receiver, given the current channel state and beamforming weights. Simulations show that the proposed algorithm reaches near optimal bit error probability (BEP) performance in low mobility scenarios, and offers advantages over non-recursive codebook solutions on a wide range of fading rates. The formulas presented in this paper specialize to a new solution in the single receive antenna multiple-input single-output (MISO) case, which allows an intuitive geometrical interpretation. The new MISO formulation is shown to outperform a recently proposed partial update beamforming scheme for single receive antenna systems.

This paper proposes a new channel estimation approach for MIMO-OFDM systems subject to power amplifier nonlinearities. The harmful effect of nonlinear distortion on the system performance can be reduced by applying an iterative distortion cancellation technique. However, such an approach requires accurate channel estimates to yield good performance. In our approach, initial channel estimation is carried out in frequency domain (FD) followed by time domain (TD) processing of the received signal. In the TD step, the equalized signals from the FD processing is used to remove the nonlinear distortion and to improve the initial channel estimate. Simulations verify that the combined channel estimation and nonlinear cancellation strategy has a performance close to the case of perfectly known channel. Furthermore, a substantial improvement in the system performance is achieved with the proposed technique when compared with conventional methods in terms of mean square error and BER

The paper describes the implementation of the multistage Wiener filter (MWF) through a series of nested Householder transformations applied to the input signal as a means to form the analysis filter-bank part of the structure. The input signal is successively projected onto an appropriate subspace whereby the mutual information is maximized at each step, as required by the MWF structure. The method presented is described-as a constrained optimization problem, where the constraints are imposed on the input signal, and not on the coefficients of the filter. The method is applicable to reduced-rank as well as to full-rank implementations of the MWF. The Householder transformation assures that the equivalent blocking matrices for all the stages are efficiently implemented via single reflections and that only unitary reflections are employed; robustness against deleterious finite-precision effects is, therefore, improved. Simulations of a DS-CDMA interference suppression receiver illustrate the robust behavior of the proposed scheme when implemented in finite precision as compared to the conventional MWF using non-orthogonal blocking matrices.

This letter compares the transients of the constrained recursive least squares (CRLS) algorithm with the generalized sidelobe canceler (GSC) employing the recursive least squares (RLS) algorithm. We prove that the two adaptive implementations are equivalent everywhere regardless of the blocking matrix chosen. This guarantees that algorithm tuning is not affected by the blocking matrix. This result differs from the more restrictive case for transient equivalence of the constrained least mean-square (CLMS) algorithm and the GSC employing the least mean square (LMS) algorithm, for in this case the blocking matrix needs to be unitary.

Partial-update adaptive filtering algorithms only update part of the filter coefficients at each time instant, leading to reduced computational complexity as compared with their conventional counterparts. In this paper, the ideas of the partial-update NLMS-type algorithms found in the literature are extended to the framework of set-membership filtering, from which data-selective NLMS type of algorithms with partial update are derived. The new algorithms combine data-selective updating from set-membership filtering with the reduced computational complexity from partial updating. Simulation results verify the good performance of the new algorithms in terms of convergence speed, final misadjustment, and reduced computational complexity.

This paper introduces a constrained version of the recently proposed set-membership affine projection algorithm based on the set-membership criteria for coefficient update. The algorithm is suitable for linearly constrained minimum-variance filtering applications. The data-selective property of the proposed algorithm greatly reduces the computational burden as compared with a nonselective approach. Simulation results show the good performance in terms of convergence, final misadjustment, and reduced computational complexity.

This letter presents a new data selective adaptive filtering algorithm, the set-membership affine projection (SM-AP) algorithm. The algorithm generalizes the idea of the set-membership NLMS (SM-NLMS) algorithm to include constraint sets constructed from the past input and desired signal pairs. The resulting algorithm can be seen as a set-membership version of the affine-projection (AP) algorithm with an optimized step size. Also, the SM-AP algorithm does not trade convergence speed with misadjustment and computational complexity as most adaptive filtering algorithms. Simulations show the good performance of the algorithm, especially for colored input signals, in terms of convergence, final misadjustment, and reduced computational complexity.

This paper proposes a new approach to linearly-constrained adaptive filtering, where successive Householder transformations are incorporated in the algorithm update equation in order to reduce computational complexity and coefficient-error norm. We show the derivation of two new algorithms, namely the unnormalized and the normalized Householder-transform constrained LMS algorithms (HCLMS and NHCLMS, respectively). Although the derivation is carried out based on the constrained LMS (CLMS) algorithm, the technique can be applied to other constrained algorithms as well. Simulation results of a linearly-constrained minimum-variance problem show that in finite-precision implementation the coefficient-error norms obtained with the new algorithms are smaller than those obtained with the CLMS and the normalized CLMS algorithms.

Pipelined implementation of an adaptive direct-sequence code division multiple access (DS-CDMA) receiver is proposed when multiple antennas are utilized for mobile communications. Adaptive multiple-antenna receivers can provide insensitivity to the interfering powers and room for more users or require a smaller number of antennas than the matched filter solution. A number of approximation techniques are utilized to pipeline the adaptive algorithm used for the proposed multiple-antenna receiver. The resulting pipelined receiver requires minimal hardware increase and achieves a higher throughput or requires lower power as compared to the receiver using the serial algorithm. Simulation results illustrate the signal-to-interference ratio (SIR) versus the relative interfering power for different number of antennas and different levels of pipelining.

This paper examines the utilization of multiple antennas for direct-sequence code division multiple access (DS-CDMA) mobile communications at the mobile receiver. A linear single-user multiple-antenna receiver is derived and its performance is compared with that of the conventional matched filter. An adaptive implementation of the receiver is also considered. The results show that the multiple-antenna receiver is insensitive to the interfering powers and can provide room for more users or a smaller number of antennas than the matched filter solution. Using the adaptive algorithm, the performance even with a single antenna is often much better than that of a matched filter with 4 antennas.