Results 1  10
of
13
Sparsitypromoting sensor selection for nonlinear measurement models
 IEEE Trans. Signal Process. (Submitted
, 2013
"... Abstract—The problem of choosing the best subset of sensors that guarantees a certain estimation performance is referred to as sensor selection. In this paper, we focus on observations that are related to a general nonlinear model. The proposed framework is valid as long as the observations are ind ..."
Abstract

Cited by 12 (9 self)
 Add to MetaCart
(Show Context)
Abstract—The problem of choosing the best subset of sensors that guarantees a certain estimation performance is referred to as sensor selection. In this paper, we focus on observations that are related to a general nonlinear model. The proposed framework is valid as long as the observations are independent, and its likelihood satisfies the regularity conditions. We use several functions of the Cramér–Rao bound (CRB) as a performance measure. We formulate the sensor selection problem as the design of a sparse vector, which in its original form is a nonconvex(quasi) norm optimization problem. We present relaxed sensor selection solvers that can be efficiently solved in polynomial time. The proposed solvers result in sparse sensing techniques. We also propose a projected subgradient algorithm that is attractive for largescale problems. The developed theory is applied to sensor placement for localization. Index Terms—Convex optimization, Cramér–Rao bound, nonlinear models, projected subgradient algorithm, sensor networks,
Giannakis, “Electricity market forecasting via lowrank multikernel learning
 IEEE J. Sel. Topics Sig. Proc
, 2014
"... Abstract—The smart grid vision entails advanced information technology and data analytics to enhance the efficiency, sustainability, and economics of the power grid infrastructure. Aligned to this end, modern statistical learning tools are leveraged here for electricity market inference. Dayahead ..."
Abstract

Cited by 7 (6 self)
 Add to MetaCart
(Show Context)
Abstract—The smart grid vision entails advanced information technology and data analytics to enhance the efficiency, sustainability, and economics of the power grid infrastructure. Aligned to this end, modern statistical learning tools are leveraged here for electricity market inference. Dayahead price forecasting is cast as a lowrank kernel learning problem. Uniquely exploiting the market clearing process, congestion patterns are modeled as rankone components in the matrix of spatiotemporally varying prices. Through a novel nuclear normbased regularization, kernels across pricing nodes and hours can be systematically selected. Even though marketwide forecasting is beneficial from a learning perspective, it involves processing highdimensional market data. The latter becomes possible after devising a blockcoordinate descent algorithm for solving the nonconvex optimization problem involved. The algorithm utilizes results from blocksparse vector recovery and is guaranteed to converge to a stationary point. Numerical tests on real data from the Midwest ISO (MISO) market corroborate the prediction accuracy, computational efficiency, and the interpretative merits of the developed approach over existing alternatives. Index Terms—Blockcoordinate descent, dayahead energy prices, graph Laplacian, kernelbased learning, learning, lowrank matrix, multikernel learning, nuclear norm regularization. I.
Grid topology identification using electricity prices
 in Proc. IEEE Power & Energy Society General Meeting, National Harbor, MD
, 2014
"... ar ..."
(Show Context)
IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS (TO APPEAR) 1 Dynamic Energy Management for SmartGrid Powered Coordinated Multipoint Systems
"... Abstract—Due to increasing threats of global warming and climate change concerns, green wireless communications have recently drawn intense attention towards reducing carbon emissions. Aligned with this goal, the present paper deals with dynamic energy management for smartgrid powered coordinated ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract—Due to increasing threats of global warming and climate change concerns, green wireless communications have recently drawn intense attention towards reducing carbon emissions. Aligned with this goal, the present paper deals with dynamic energy management for smartgrid powered coordinated multipoint (CoMP) transmissions. To address the intrinsic variability of renewable energy sources, a novel energy transaction mechanism is introduced for gridconnected base stations that are also equipped with an energy storage unit. Aiming to minimize the expected energy transaction cost while guaranteeing the worstcase users ’ quality of service, an infinitehorizon optimization problem is formulated to obtain the optimal downlink transmit beamformers that are robust to channel uncertainties. Capitalizing on the virtualqueue based relaxation technique and the stochastic dualsubgradient method, an efficient online algorithm is developed yielding a feasible and asymptotically optimal solution. Numerical tests with synthetic and real data corroborate the analytical performance claims and highlight the merits of the novel approach. Index Terms—CoMP systems, downlink beamforming, smart grids, highpenetration renewables, stochastic optimization, Lyapunov optimization. I.
Decentralized Learning for Wireless Communications and Networking
"... Abstract This chapter deals with decentralized learning algorithms for innetwork processing of graphvalued data. A generic learning problem is formulated and recast into a separable form, which is iteratively minimized using the alternatingdirection method of multipliers (ADMM) so as to gain the ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract This chapter deals with decentralized learning algorithms for innetwork processing of graphvalued data. A generic learning problem is formulated and recast into a separable form, which is iteratively minimized using the alternatingdirection method of multipliers (ADMM) so as to gain the desired degree of parallelization. Without exchanging elements from the distributed training sets and keeping internode communications at affordable levels, the local (pernode) learners consent to the desired quantity inferred globally, meaning the one obtained if the entire training data set were centrally available. Impact of the decentralized learning framework to contemporary wireless communications and networking tasks is illustrated through case studies including target tracking using wireless sensor networks, unveiling Internet traffic anomalies, power system state estimation, as well as spectrum cartography for wireless cognitive radio networks.
Advanced optimization methods for power systems
, 2014
"... Power system planning and operation offers multitudinous opportunities for optimization methods. In practice, these problems are generally largescale, nonlinear, subject to uncertainties, and combine both continuous and discrete variables. In the recent years, a number of complementary theoretical ..."
Abstract
 Add to MetaCart
Power system planning and operation offers multitudinous opportunities for optimization methods. In practice, these problems are generally largescale, nonlinear, subject to uncertainties, and combine both continuous and discrete variables. In the recent years, a number of complementary theoretical advances in addressing such problems have been obtained in the field of applied mathematics. The paper introduces a selection of these advances in the fields of nonconvex optimization, in mixedinteger programming, and in optimization under uncertainty. The practical relevance of these developments for power systems planning and operation are discussed, and the opportunities for combining them, together with highperformance computing and big data infrastructures, as well as novel machine learning and randomized algorithms, are highlighted.
KERNEL SELECTION FOR POWER MARKET INFERENCE VIA BLOCK SUCCESSIVE UPPER BOUND MINIMIZATION
"... Advanced data analytics are undoubtedly needed to enable the envisioned smart grid functionalities. Towards that goal, modern statistical learning tools are developed for dayahead electricity market inference. Congestion patterns are modeled as rankone components in the matrix of spatiotemporal p ..."
Abstract
 Add to MetaCart
(Show Context)
Advanced data analytics are undoubtedly needed to enable the envisioned smart grid functionalities. Towards that goal, modern statistical learning tools are developed for dayahead electricity market inference. Congestion patterns are modeled as rankone components in the matrix of spatiotemporal prices. The new kernelbased predictor is regularized by the square root of the nuclear norm of the sought matrix. Such a regularizer not only promotes lowrank solutions, but it also facilitates a systematic kernel selection methodology. The nonconvex optimization problem involved is efficiently driven to a stationary point following a block successive upper bound minimization approach. Numerical tests on real highdimensional market data corroborate the interpretative merits and the computational efficiency of the novel method. Index Terms — Kernel learning; nuclear norm; multikernel selection; block successive upper bound minimization. 1.
LowRank Kernel Learning for Electricity Market Inference
"... Abstract—Recognizing the importance of smart grid data analytics, modern statistical learning tools are applied here to wholesale electricity market inference. Market clearing congestion patterns are uniquely modeled as rankone components in the matrix of spatiotemporally correlated prices. Upon po ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract—Recognizing the importance of smart grid data analytics, modern statistical learning tools are applied here to wholesale electricity market inference. Market clearing congestion patterns are uniquely modeled as rankone components in the matrix of spatiotemporally correlated prices. Upon postulating a lowrank matrix factorization, kernels across pricing nodes and hours are systematically selected via a novel methodology. To process the highdimensional market data involved, a blockcoordinate descent algorithm is developed by generalizing blocksparse vector recovery results to the matrix case. Preliminary numerical tests on real data corroborate the prediction merits of the developed approach. I.
10.1109/TSP.2014.2332438, IEEE Transactions on Signal Processing 1 Distributed Hybrid Power State Estimation under PMU Sampling Phase Errors
, 2014
"... for more information. ..."
(Show Context)