Estimating the unknown parameters of a statistical model based on the observations collected by a sensor network is an important problem with application in multiple fields. In this setting, distributed processing, by which computations are carried out within the network in order to avoid raw data transmission to a fusion centre, is a desirable feature resulting in improved robustness and energy savings. In the presence of incomplete data, the expectation-maximisation (EM) algorithm is a popular means to iteratively compute the maximum likelihood (ML) estimate. It has found application in diverse fields such as computational biology, anomaly detection, speech segmentation, reinforcement learning, and motion estimation, among others. In this chapter we will review the formulation of the centralised EM estimation algorithm as a starting point and then discuss distributed versions well suited for implementation in sensor networks. The first class of these distributed versions requires specialised routing through the network in terms of a linear or circular path visiting all nodes, whereas the second class does away with this requirement by using the concept of network consensus to diffuse information through the network. Our focus will be on a relevant sensor network application, in which the parameter of a linear model is to be estimated in the presence of an unknown number of randomly malfunctioning sensors.

%B Data Fusion in Wireless Sensor Networks: A statistical signal processing perspective %I The Institution of Engineering and Technology (IET) %C London, UK %P 201-230 %@ 978-1-78561-584-9 %G eng %& 9 %0 Conference Paper %B IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP) %D 2019 %T Flexible spectral precoding for sidelobe suppression in OFDM systems %A Khawar Hussain %A Ana Lojo %A R. López-Valcarce %K spectrum shaping %K winter %XSpectral precoding is a popular approach to reduce out-of-band radiation (OBR) in multicarrier systems in order to avoid adjacent channel interference. Since precoding will introduce signal distortion, appropriate decoding is required at the receiver side. We present a novel linear precoder design with flexibility to trade off OBR reduction, precoding/decoding complexity, and error rate at the receiver. The precoding matrices have low rank, which translates into significant computational savings. In this way, the requirements of different systems can be satisfied with varying complexity levels.

%B IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP) %I IEEE %C Brighton, UK %P 4789 - 4793 %8 04/2019 %G eng %R 10.1109/ICASSP.2019.8683162 %0 Conference Paper %B IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP) %D 2019 %T Frequency-selective hybrid precoding and combining for mmWave MIMO systems with per-antenna power constraints %A Javier Rodríguez-Fernández %A R. López-Valcarce %A Nuria González-Prelcic %K mmWave %K winter %XConfiguring hybrid precoders and combiners is the main challenge to be solved to operate at millimeter wave (mmWave) frequencies. The use of hybrid architectures imposse hardware constraints on the analog precoder that need to be carefully dealt with. In this paper, we develop hybrid precoders and combiners aiming at minimizing the Euclidean distance with respect to the approximate all-digital precoders and combiners maximizing the spectral efficiency under per-antenna power constraints. Numerical results demonstrate the effectiveness of the proposed design method, whose performance is close to that of the all-digital solution.

%B IEEE Int. Conf. Acoustics, Speech and Signal Processing (ICASSP) %I IEEE %C Brighton, UK %P 4794-4798 %8 04/2019 %G eng %R 10.1109/ICASSP.2019.8683831 %0 Journal Article %J Signal Processing %D 2019 %T MMSE filter design for Full-Duplex filter-and-forward MIMO relays under limited dynamic range %A Emilio Antonio-Rodríguez %A Stefan Werner %A R. López-Valcarce %A Risto Wichman %K full-duplex %K winter %XWe study the problem of optimizing the end-to-end performance of a full-duplex filter-and-forward MIMO relay link, consisting

of a source, a relay, and a destination node, by employing linear filtering at each node. The system model accounts for multipath

propagation and self-interference at the relay, as well as transmitter impairments and limited dynamic range at every node. The

design accommodates signals with arbitrary spectra and includes the direct link between the source and destination nodes. Under

the minimum mean square error criterion, the resulting non-convex problem is approximated by a sequence of convex problems

and solved by means of an alternating minimization method. Linear constraints allocate some of the degrees of freedom in the relay

to guarantee a sufficiently small residual self-interference. Simulations quantify the impact of degrees of freedom, the dynamic

range, and the balance between direct and relay paths on the link performance.

%B Signal Processing
%V 156
%P 208-219
%8 03/2019
%G eng
%R doi.org/10.1016/j.sigpro.2018.11.001
%0 Journal Article
%J IEEE Transactions on Wireless Communications
%D 2019
%T Partial-Duplex Amplify-and-Forward Relaying: Spectral Efficiency Analysis under Self-Interference
%A R. López-Valcarce
%A C. Mosquera
%K full-duplex
%K myrada
%K winter
%X We propose a novel mode of operation for Amplify-and-

Forward relays in which the spectra of the relay input and

output signals partially overlap. This partial-duplex relaying mode

encompasses half- and full-duplex as particular cases. By viewing

the partial-duplex relay as a bandwidth-preserving Linear

Periodically Time-Varying system, a spectral efficiency analysis

under self-interference is developed. In contrast with previous

works, self-interference is regarded as a useful informationbearing

component rather than simply assimilated to noise. This

approach reveals that previous results regarding the impact of

self-interference on (full-duplex) relay performance are overly

pessimistic. Based on a frequency-domain interpretation of the

effect of self-interference, a number of suboptimal decoding

architectures at the destination node are also discussed. It is found

that the partial-duplex relaying mode may provide an attractive

tradeoff between spectral efficiency and receiver complexity.

%B IEEE Transactions on Wireless Communications
%V 18
%P 2271-2285
%8 04/2019
%G eng
%N 4
%R 10.1109/TWC.2019.2902390
%0 Conference Paper
%B IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
%D 2018
%T Locally optimal invariant detector for testing equality of two power spectral densities
%A David Ramírez
%A Daniel Romero
%A Javier Vía
%A R. López-Valcarce
%A Ignacio Santamaría
%K cognitive radio
%K winter
%X This work addresses the problem of determining whether two multivariate random time series have the same power spectral density

(PSD), which has applications, for instance, in physical-layer security and cognitive radio. Remarkably, existing detectors for this

problem do not usually provide any kind of optimality. Thus, we study here the existence under the Gaussian assumption of optimal

invariant detectors for this problem, proving that the uniformly most powerful invariant test (UMPIT) does not exist. Thus, focusing on

close hypotheses, we show that the locally most powerful invariant test (LMPIT) only exists for univariate time series. In the multivariate

case, we prove that the LMPIT does not exist. However, this proof suggests two LMPIT-inspired detectors, one of which outperforms

previously proposed approaches, as computer simulations show.

%B IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
%C Calgary, Canada
%G eng
%0 Journal Article
%J Signal Processing
%D 2018
%T Parameter estimation in wireless sensor networks with faulty transducers: A distributed EM approach
%A Silvana Silva Pereira
%A R. López-Valcarce
%A Alba Pagès-Zamora
%K winter
%K wsn
%X We address the problem of distributed estimation of a vector-valued parameter performed by a wireless sensor network in the presence of noisy observations which may be unreliable due to faulty transducers. The proposed distributed estimator is based on the Expectation-Maximization (EM) algorithm and combines consensus and diffusion techniques: a term for information diffusion is gradually turned off, while a term for updated information averaging is turned on so that all nodes in the network approach the same value of the estimate. The proposed method requires only local exchanges of information among network nodes and, in contrast with previous approaches, it does not assume knowledge of the a priori probability of transducer failures or the noise variance. A convergence analysis is provided, showing that the convergent points of the centralized EM iteration are locally asymptotically convergent points of the proposed distributed scheme. Numerical examples show that the distributed algorithm asymptotically attains the performance of the centralized EM method.

%B Signal Processing %V 144 %P 226-237 %8 03/2018 %G eng %U https://authors.elsevier.com/a/1W90XbZX4rsob %R 10.1016/j.sigpro.2017.10.012 %0 Journal Article %J IEEE Trans. Signal Processing %D 2018 %T Testing equality of multiple power spectral density matrices %A David Ramírez %A Daniel Romero %A Javier Vía %A R. López-Valcarce %A Ignacio Santamaría %K winter %B IEEE Trans. Signal Processing %V 66 %P 6268-6280 %8 12/2018 %G eng %N 23 %R 10.1109/TSP.2018.2875884 %0 Conference Paper %B European Signal Processing Conference (EUSIPCO) %D 2017 %T Defending Surveillance Sensor Networks Against Data-Injection Attacks via Trusted Nodes %A R. López-Valcarce %A Daniel Romero %K winter %K wsn %XBy injecting false data through compromised sensors, an adversary can drive the probability of detection in a sensor network-based spatial field surveillance system to arbitrarily low values. As a countermeasure, a small subset of sensors may be secured. Leveraging the theory of Matched Subspace Detection, we propose and evaluate several detectors that add robustness to attacks when such trusted nodes are available. Our results reveal the performance-security tradeoff of these schemes and can be used to determine the number of trusted nodes required for a given performance target.

%B European Signal Processing Conference (EUSIPCO)
%C Kos Island, Greece
%8 08/2017
%G eng
%0 Journal Article
%J IEEE Transactions on Wireless Communications
%D 2017
%T FER Estimation in a Memoryless BSC with Variable Frame Length and Unreliable ACK/NAK Feedback
%A Alberto Rico-Alvariño
%A R. López-Valcarce
%A Carlos Mosquera
%A Robert W. Heath Jr.
%K adaptive signal processing
%K compass
%K Cramér-Rao bound (CRB)
%K frame error rate
%K myrada
%K winter
%X We consider the problem of estimating the frame error rate (FER) of a given memoryless binary symmetric channel by observing the success or failure of transmitted packets. Whereas FER estimation is relatively straightforward if all observations correspond to packets with equal length, the problem becomes considerably more complex when this is not the case. We develop FER estimators when

transmissions of different lengths are observed, together with the Cramer-Rao Lower Bound (CRLB). Although the main focus is on Maximum Likelihood (ML) estimation, we also obtain low complexity schemes performing close to optimal in some scenarios. In a second stage, we consider the case in which FER estimation is performed at a node different from the receiver, and incorporate the impairment of unreliable observations by considering noisy ACK/NAK feedback links. The impact of unreliable feedback is analyzed by means of the corresponding CRLB. In this setting, the ML estimator is obtained by applying the Expectation-Maximization algorithm to jointly estimate the error probabilities of the data and feedback links. Simulation results illustrate the benefits of the proposed estimators.

%B IEEE Transactions on Wireless Communications
%V 16
%P 3661 - 3673
%8 06/2017
%G eng
%N 6
%R 10.1109/TWC.2017.2686845
%0 Conference Paper
%B IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
%D 2017
%T Filter design for delay-based anonymous communications
%A Simon Oya
%A Fernando Pérez-González
%A Carmela Troncoso
%K adversarial signal processing
%K compass
%K Multimedia security
%K winter
%B IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
%G eng
%0 Conference Paper
%B Workshop on Privacy in the Electronic Society (WPES)
%D 2017
%T Is Geo-Indistinguishability what you are looking for?
%A Simon Oya
%A Carmela Troncoso
%A Fernando Pérez-González
%K adversarial signal processing
%K Multimedia security
%K winter
%B Workshop on Privacy in the Electronic Society (WPES)
%G eng
%0 Journal Article
%J IEEE Transactions on Signal Processing
%D 2017
%T Learning Power Spectrum Maps from Quantized Power Measurements
%A Daniel Romero
%A Seung-Jun Kim
%A Georgios Giannakis
%A R. López-Valcarce
%K cognitive radio
%K compass
%K compressed sensing
%K spectrum sensing
%K winter
%K wsn
%X Using power measurements collected by a network of low-cost sensors, power spectral density (PSD) maps are con-

structed to capture the distribution of RF power across space and frequency. Linearly compressed and quantized power measure-

ments enable wideband sensing at affordable implementation complexity using a small number of bits. Strengths of data- and model-

driven approaches are combined to develop estimators capable of incorporating multiple forms of spectral and propagation prior

information while fitting the rapid variations of shadow fading across space. To this end, novel nonparametric and semiparametric

formulations are investigated. It is shown that the desired PSD maps can be obtained using support vector machine-type solvers.

In addition to batch approaches, an online algorithm attuned to real-time operation is developed. Numerical tests assess the performance of the novel algorithms.

%B IEEE Transactions on Signal Processing
%V 65
%P 2547-2560
%8 05/2017
%G eng
%N 10
%R 10.1109/TSP.2017.2666775
%0 Journal Article
%J IEEE Transactions on Information Forensics and Security
%D 2017
%T Number Theoretic Transforms for Secure Signal Processing
%A Alberto Pedrouzo-Ulloa
%A Juan Ramón Troncoso-Pastoriza
%A Fernando Pérez-González
%K lattice cryptography
%K Multimedia security
%K Number Theoretic Transforms
%K Secure Signal Processing
%K Signal Processing in the Encrypted Domain
%K Somewhat Homomorphic Encryption
%K winter
%B IEEE Transactions on Information Forensics and Security
%V 12
%P 1125-1140
%8 05/2017
%G eng
%N 5
%& 1125
%0 Journal Article
%J IEEE Transactions on Information Forensics and Security
%D 2017
%T A Random Matrix Approach to the Forensic Analysis of Upscaled Images
%A David Vázquez-Padín
%A Fernando Pérez-González
%A Pedro Comesaña
%K Multimedia security
%K winter
%B IEEE Transactions on Information Forensics and Security
%V 12
%P 2115-2130
%8 09/2017
%G eng
%N 9
%& 2115
%0 Conference Paper
%B International Tyrrhenian Workshop on Digital Communication 2017
%D 2017
%T Random Matrix Theory for Modeling the Eigenvalue Distribution of Images under Upscaling
%A David Vázquez-Padín
%A Fernando Pérez-González
%A Pedro Comesaña
%K Multimedia security
%K winter
%X The stochastic representation of digital images through a two-dimensional autoregressive (2D-AR) model offers a proper way to approximate the empirical distribution of the eigenvalues coming from genuine images. By considering this model, we apply random matrix theory to analytically derive the asymptotic eigenvalue distribution of causal 2D-AR random fields that have undergone an upscaling operation with a particular interpolation kernel. This eigenvalue characterization is useful in developing new forensic techniques for image resampling detection since we can use theoretical bounds to drive the decision of detectors based on subspace decomposition. Moreover, experimental results with real images show that the obtained asymptotic limits turn out to be excellent approximations, even when working with images of small size.

%B International Tyrrhenian Workshop on Digital Communication 2017 %I Springer International Publishing %C Palermo, Italy %P 109-124 %8 09/2017 %G eng %0 Conference Paper %B IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) %D 2017 %T Robust clustering of data collected via crowdsourcing %A Alba Pagès-Zamora %A Georgios Giannakis %A R. López-Valcarce %A Pere Gimenez-Febrer %K winter %K wsn %XCrowdsourcing approaches rely on the collection of multiple individuals to solve problems that require analysis of large data sets in a timely accurate manner. The inexperience of participants or annotators motivates well robust techniques. Focusing on clustering setups, the data provided by all annotators is suitably modeled here as a mixture of Gaussian components plus a uniformly distributed random variable to capture outliers. The proposed algorithm is based on the expectation-maximization algorithm and allows for soft assignments of data to clusters, to rate annotators according to their performance, and to estimate the number of Gaussian components in the non-Gaussian/Gaussian mixture model, in a jointly manner.

%B IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP) %C New Orleans %P 4014 - 4018 %8 03/2017 %G eng %R 10.1109/ICASSP.2017.7952910 %0 Journal Article %J Signal Processing %D 2017 %T Wideband full-duplex MIMO relays with blind adaptive self-interference cancellation %A Emilio Antonio-Rodríguez %A Stefan Werner %A R. López-Valcarce %A Taneli Riihonen %A Risto Wichman %K adaptive signal processing %K compass %K full duplex %K winter %XWe develop adaptive self-interference cancellation algorithms for both filter-and-forward and decode-and-forward

multiple-input multiple-output relays. The algorithms are blind in the sense that they only exploit the spectral properties

of the transmitted signal to identify the self-interference channel, while dealing with frequency-selective channels and

arbitrary signal spectra. Our approach is non-intrusive in the sense that the algorithms can successfully identify, track,

and cancel the self-interference distortion while the relay is operating in its normal mode. We study the stationary

points of the algorithms and analyze under which conditions they achieve perfect cancellation of the self-interference.

Simulation results show that the algorithms provide residual self-interference levels below the noise floor by using

the time samples of only a few OFDM symbols.

%B Signal Processing
%V 130
%P 74-85
%8 01/2017
%G eng
%& 74
%R 10.1016/j.sigpro.2016.06.010