Random Access Sensor Networks: Field Reconstruction From Incomplete Data
|
|
- Theodora Barnett
- 5 years ago
- Views:
Transcription
1 Random Access Sensor Networks: Field Reconstruction From Incomplete Data Fatemeh Fazel Northeastern University Boston, MA Maryam Fazel University of Washington Seattle, WA Milica Stojanovic Northeastern University Boston, MA Abstract We address efficient data gathering from a network of distributed sensors deployed in a challenging field environment with limited power and bandwidth. Utilizing the low-rank property of the sensing field, we leverage results from the matrix completion theory to integrate the sensing procedure with a simple and robust communication scheme based on random channel access. Results show that the space-time map of the sensing field can be recovered efficiently, using only a small subset of sensor measurements, collected over a fading random access channel. I. INTRODUCTION Wireless sensor networks greatly facilitate long-term monitoring of the natural environment [1]. Such networks comprise a large array of battery-powered sensors, a central collection unit referred to as the Fusion Center (FC) and a communication scheme for the sensors to transmit their data to the FC. Once the network is deployed, access to the sensors is limited, hence re-charging the batteries is not easy. As a result, longterm deployment entails energy-aware sensing and efficient communication. Due to the size of the network, and the energy, bandwidth and time constraints of the data acquisition process, missing or partial information presents a ubiquitous problem in many applications such as geosciences [2], industrial and environmental monitoring, and monitoring surface deformations related to oil fields [3]. Thus, field recovery inevitably has to rely on incomplete data. Based on the principles of compressed sensing and random channel access, in [4], [] we proposed an integrated architecture for sensing and communication, referred to as Random Access Compressed Sensing (RACS). This scheme targets the recovery of slowly varying fields from sub-sampled data, gathered while the process is approximately unchanged. RACS utilizes the fact that most natural signals have a sparse representation in the frequency domain. In the current work, we relax this assumption and take into consideration the temporal variations of the field over much longer periods of time. Employing results from the lowrank matrix recovery problem [6] and the matrix completion theory [7] [9], we address the design of a random access network for long term monitoring of temporally varying fields. Research funded in part by ONR grant N , NSF grant , and NSF CAREER grant ECCS The matrix completion theory addresses the recovery of a low-rank matrix from a subset of its entries. Matrix completion is applied to field monitoring in [], where the authors propose an online algorithm for subspace tracking. The process of gathering the data, which is the focus of our work, was not however in the scope of this paper. In the context of cognitive radios, collaborative spectrum sensing using matrix completion is studied in [11]. In [12], the authors discuss sampling strategies over a grid network to improve energy efficiency. The communication aspect, however, is not addressed, and the time variations of the process are not considered. The sensing and communication aspects of the wireless sensor networks are commonly treated independently. Our contribution is in unifying the sensing scheme, based on lowrank matrix recovery, with a simple and robust communication scheme using random access. The proposed integrated sensing and communication scheme relies only on a few assumptions about the statistical properties of the sensing field, and is thus applicable to a variety of fields. The rest of the paper is organized as follows: In Section II, we integrate the sensing scheme, based on matrix completion, with the communication scheme, using random access. In Section III, a probabilistic model for the system is provided upon which the network design guidelines of Section IV are then based. In Section V, we study the inherent trade-offs in choosing the system parameters. Finally, concluding remarks are provided in Section VI. II. RECONSTRUCTION FROM INCOMPLETE DATA We consider a sensor network with N nodes, where each node measures the field at a given sensing rate λ 1. We will rely on the matrix completion theory to determine the required sensing rate λ 1, which is the objective of the design procedure. A. Field Model The network measures a process u(x, y, t) where x and y denote positions in space and t denotes time. We assume that node i is located at position (x i, y i ), where i {1,..., N}. The observations of the field are available at discrete intervals of time separated by t. Assuming a stationary process, the discretization interval t depends on the temporal autocorrelation of the process R(τ), which quantifies the average correlation between two samples of the process separated
2 by time τ. In other words, R(τ) = 1 N N i=1 R i(τ), where R i (τ) = E{u i (t + τ)u i (t)} denotes the autocorrelation of the process observed by sensor i. The value of t is chosen as the time lag during which the samples of the signal are sufficiently correlated, i.e., t is chosen such that R( t) = qr() (1) where q is the desired level of the correlation. 1 The process thus remains almost unchanged during t, and packets received from distinct nodes during this time interval are regarded as pertaining to the same snapshot of the process. Precise synchronization among nodes, which is difficult to achieve in a large-scale network, is thus not needed. The appropriate value for t strikes a balance between the required sensing rate and the redundancy of sampling. Consequently, as we will show in Section V, it provides a trade-off between the consumed energy and the quality of reconstruction. With a slight abuse of notation, let u i (m) denote the signal measured by sensor node i at any time t [(m 1) t, m t]. Now the sensing field is described by the space-time matrix of sensor measurements as u 1 (1) u 1 (2)... u 1 (M) U =... (2) u N (1) u N (2)... u N (M) The matrix U is of dimension N M, where for large-scale and long-term monitoring applications, both M and N are large. We assume that U is a low rank matrix, i.e., its rank r is much smaller than N and M, as is the case with many natural signals. To illustrate this fact, we generate an example process which will serve as our test case. We assume a large field over which k = sources are randomly placed, each generating an exponentially decaying signal (e.g., heat, sound, etc). A total of N = 2 sensors are distributed over the field to monitor the process. At time t, the observed process at coordinate (x i, y i ) is given by u i (t) = k A j e j p (xi a j ) 2 +(y i b j ) 2 (3) j=1 where (a j, b j ) are the coordinates, A j is the strength, and p j is the decay coefficient of the j-th source, respectively. The map of the process at initial time t = is shown in Fig. 1. The process then evolves over time as the sources move along random trajectories. In order to sample the continuous-time process, and to form the data matrix U, we first determine the appropriate t. Fig. 2 shows the normalized temporal autocorrelation function of the process. For a desired value of q, say q =.98, we note that t = 1 sec. The process is thus unchanged during t = 1 sec. The resulting data matrix U is 1 The function R(τ) may not be completely known a-priori, since it depends on the process being sensed; however, we can assume that it is approximately known from historical data. In Section V, we discuss the dependence of system performance on the choice of t. y x Fig. 1. The map of the process U(x, y, t) at time t =, where the location of the targets is shown by circles. As time progresses, the sources move randomly in the space. The readings of the sensors u i (m) are collected to form the data matrix U (Eq. (2)). normalized autocorrelation time lag τ [sec] Fig. 2. Normalized temporal autocorrelation function of the process, R(τ), averaged over all the sensors. The autocorrelation decreases to.98% of its maximum value at t = 1 sec. formed by joining M consecutive snapshots of the process at the chosen t intervals. The normalized singular values of U are plotted in Fig. 3, for several values of M. It is clear that U has only a small number of significant eigenvalues, i.e., it is low-rank. B. Communication Scheme Each sensor node generates measurement packets randomly at an average rate of λ 1 packets/sec and transmits its packets immediately on a random access channel. Without loss of generality, we assume that packet generation at each node follows an independent Poisson process. Each packet contains the quantized and binary encoded measurement information as well as the location tag for the generating sensor. Once a packet is received, the FC assigns a time stamp to it according to the packet s reception time
3 normalized singular value of the data matrix M=3 M= M= 7 1 index of singular value Fig. 3. Singular values of the data matrix U, normalized with respect to the maximum singular value, plotted for several values of M. We notice that the matrix has a few significant eigenvalues while the rest are negligible. In a random access network, packet collisions are inevitable. Thus, some packets are lost, while others may be received in error due to the communication noise. The FC discards the erroneous packets and collects the rest over an interval M t. It then arranges these packets into an N M matrix X, whose entries correspond to the successfully received packets. Since the sensors transmit at random moments in time, and because packet losses due to collisions and communication noise are random as well, X can be regarded as containing a uniformly random subset of the entries of U. Specifically, the entries of X are x i (m) = u i (m) if the i-th sensor transmitted during the m-th interval, and if the transmission was successful, otherwise x i (m) =. We leverage results from the matrix completion problem which asserts that under some conditions, an unknown lowrank matrix can be recovered from observing a sufficient number of its entries by solving a convex optimization problem, as long as the locations of the observed entries are picked uniformly at random. The number of entries required for recovery is on the order of O(rn(log n) 2 ) where n = max{n, M} [7]. Thus, once the FC collects a sufficient number of packets, it can recover the data matrix U using matrix completion algorithms. C. Recovery The sensing field is reconstructed from the useful packets collected during an observation interval of duration M t, by minimizing the nuclear norm of the matrix U, subject to the constraint u i (m) = x i (m) for the observed entries. This is a convex optimization problem, and a number of numerical algorithms can be used to solve it. Some examples are low-rank matrix fitting algorithm LMaFit [13], fixed point continuation with approximate singular value decomposition (FPCA) [14] and accelerated proximal gradient APGL [1]. In this paper, we use FPCA for recovery, which solves the nuclear norm regularized least squares problem. III. SUCCESSFUL PACKET RECEPTION Given a Poisson distribution of packet arrival times, the probability of n packets arriving concurrently at the FC is P (n) = (2Nλ 1T p ) n e 2Nλ1Tp (4) n! where T p is the packet duration. In a fading channel, successful packet reception is defined as the event in which the received signal-to-interference-and-noise ratio γ stays above a threshold b, where typically b = 2 6 [16]. In other words, the probability of successful reception is given by { } X p s = Prob > b () I n + P N where X represents the power of the desired packet, I n represents the total interference power (caused by the n interfering packets), and P N is the noise power. Assuming a Rayleigh model for the fading statistics, the probability of successful reception is given by [17] p s = e b/γ e 2Nλ 1T p b 1+b (6) where γ is the average received SNR. Within the interval t there may be more than one packet transmitted by a single node. If the FC has successfully received a packet from a given node in this interval, then the extra packet(s) are redundant since the process does not change in t. Hence, not all of the received packets at the FC are useful. Note that if a given packet is lost, a repetition may still be received successfully. For a particular node, let N 1 ( t) denote the number of packets generated during t, that are successfully received. If this number is greater than or equal to 1, the FC will keep one such successfully delivered packet and discard any other packets received from that node. Hence, the number of useful packets generated at each node during t is given by { N, N1 ( t) = ( t) = (7) 1, N 1 ( t) 1 The probability of receiving a useful packet from a node is p g = Prob{N ( t) = 1}, which can be expressed as p g = t Tp l=1 (λ 1 t) l e λ1 t [1 (1 p s ) l ] (8) l! where the first term is the probability that the node generates l packets during t, and the second term is the probability that one or more of the generated packets are received successfully. Note that the maximum number of packets that are generated by a single node during t is t T p 1. The expression (8) then reduces to p g = 1 e psλ1 t (9) From the expression (7), the effective average number of packets received from a given node during t is E{N ( t)} = p g. The average effective arrival rate of useful
4 packets at the FC is thus λ = Np g = N t t (1 e p sλ 1 t ) () The total number of packets that are used in the reconstruction process, K(λ, M t), is the number of received packets left after discarding the unsuccessful and repetitive packets. Thus, the arrival of useful packets follows a Poisson process with an effective average arrival rate λ given by Eq. (). i.e., the probability of receiving k useful packets in the interval t is P K (k) = (λ M t) k e λ M t. (11) k! IV. NETWORK DESIGN The target sampling density η s is defined as the ratio of the required number of observed entries to the total number of entries and is given by η s = Crn(log n) 2 /NM (12) where C is a constant. The actual sampling density, which is a random variable, is obtained by dividing the number of useful measurements collected in M t by the size of the data matrix, i.e., η = K(λ, M t)/nm. (13) A well-designed system thus must satisfy η η s (14) However, since η is random, a probabilistic approach to system design has to be used. We thus define the probability of sufficient sensing as the probability that the target sampling density η s is exceeded at the FC during the observation interval. We then specify the performance requirement as the minimum probability of sufficient sensing, P s, such that Prob {η η s } P s. (1) For large values of the Poisson parameter M tλ, the normal distribution closely approximates the Poisson distribution [18]. Thus, for a given η s and a desired P s, the corresponding λ s can be approximately determined from the condition ( ) MNη s M tλ s Q P s (16) M tλ s where Q( ) is the complementary cumulative distribution function of the standard normal distribution. The resulting λ s represents the minimum average rate of useful packets that meets the sufficient sensing requirement. In other words, λ λ s (17) The design objective is to determine the per-node sensing rate λ 1 that ensures sufficient sensing. The condition (17) implies that λ 1s λ 1 λ 1c (18) where λ 1s and λ 1c are the solutions to λ = λ s, as shown in λ [packet/sec] λ =λ s λ 1s λ 1m λ 1c λ [packet/sec] 1 Fig. 4. The average arrival rate of useful packets, λ, plotted versus the per-node sensing rate λ 1. The solutions to λ = λ s are shown in the figure. Fig. 4. We are only interested in those values of λ 1 for which the system is stable,i.e., for which increasing λ 1 results in an increased number of useful packets. The desired value of the per-node sensing rate thus lies in the stable region λ 1s λ 1 λ 1m (19) where λ 1m is the point at which λ reaches its maximum value. V. DESIGN CONSIDERATIONS We now study the trade-offs in choosing the system parameters. We define two performance metrics: a) the energy consumption; b) the recovery error. The energy consumed by the network to recover one second of the process is given by Γ = Nλ 1 P T T p () where P T is the average transmission power per packet per node. Clearly, a smaller sensing rate λ 1 translates into lower energy consumption. The recovery error is defined as ϵ = U Û F U F (21) where U is the data matrix given by Eq. (2), Û is the recovered data matrix at the FC, and F is the Frobenius norm. The system parameters that need to be determined are M and t. To illustrate the trade-offs, we use the test case of Fig. 1. Specifically, we intend to track this process as it evolves over time, with system parameters given by b = 2, P s =.9, γ = 2 db and T p = 2 msec. For the chosen range of values of M shown in Fig. 3, the rank r of the example data matrix can be approximated as fixed. Fig. shows that there is an optimum value of M, at which λ 1 is smallest, and hence, the energy consumption is the least. This is explained by noting that, assuming a fixed
5 λ 1s [packt/sec].7.6 recovery error M t [sec] Fig.. The per-node sensing rate, λ 1s, plotted versus M, for N = 2. The rank r is assumed to remain fixed at r =. Fig. 7. The recovery error plotted versus t, for the example field (M = N = 2, P s =.9, B = 38. kbps, b = 2, γ = 2 db)..7.6 M= N/2 M=N M=2N λ 1s [packet/sec] y t [sec] x Fig. 6. The per-node sensing rate, λ 1s, plotted versus t, for several values of M, when N = 2. rank r, η s achieves its minimum at M = N. Thus, from an energy point of view, an appropriate choice for the number of columns of the data matrix is M N. For a fixed M and N, Fig. 6 shows that λ 1s decreases with t. Thus, in order to minimize the energy consumption, t has to be chosen as the largest value possible. However, the choice of t also affects the recovery error. The effect of t on the quality of reconstruction depends on the correlation properties of the sensing field, i.e., the rate of change of the field. For the example field, Fig. 7 shows the recovery error plotted versus t. The choice of t thus involves a trade-off between the energy consumption and the desired accuracy of recovery. To visually illustrate the quality of the recovered field, Figs. 8 and 9 show the recovered snapshots of the field, with t = 1 sec and t = 2 sec, where the corresponding recovery errors are 3% and 9%, respectively. In general, both r and t are characteristics of the process that is being sensed, and as Fig. 8. The recovered map with t = 1 sec. We have employed a sensing rate λ 1s =.7 packet/sec, resulting in a sampling density η =.7. The data matrix U is then recovered with a recovery error on the order of 3%. Clearly, increasing the sampling density will reduce the recovery error further. such may not be exactly known and furthermore, may vary with time. The recovery error ϵ is thus influenced by the inaccuracies in estimating r (or equivalently, η s ) and t. VI. CONCLUSIONS Field recovery from incomplete data is a ubiquitous problem in monitoring of environmental and industrial phenomena. Using the matrix completion theory, we proposed a method for data acquisition in a wireless sensor network, deployed for long-term monitoring of large fields. By integrating sensing and communication, an efficient strategy is proposed using which the space-time map of the process is completely recovered from only a small set of measurement packets received over a fading and noisy random access channel. Network design depends on certain properties of the sensing field, such as the sampling interval and the rank of the data matrix, which may not be known a-priori and hence must be
6 y x [14] S. Ma, D. Goldfarb, and L. Chen, Fixed point and bregman iterative methods for matrix rank minimization, Mathematical Programming, pp. 1 33, 9. [1] K. Toh and S. Yun, An accelerated proximal gradient algorithm for nuclear norm regularized least squares problems, Pacific J. Optimization, vol. 6, pp ,. [16] J. Linnartz, Narrowband Land-Mobile Radio Networks. Artech House, Inc., [17] F. Fazel, M. Fazel, and M. Stojanovic, Impact of fading on random access compressed sensing, in 4th Annual Asilomar Conference on Signals, Systems, and Computers, November 11. [18] J. K. Patel and C. B. Read, Handbook of the Normal Distribution. Marcel Dekker, Fig. 9. The recovered map with t = 2 sec. We have employed a sensing rate λ 1s =.22 packet/sec, resulting in a sampling density η =.2. As shown, the quality of reconstruction now deteriorates and the recovery error increase to 9%, as a result of increasing t. estimated. We showed that there is a trade-off between energy consumption and the recovery error, and visually illustrated the reconstruction quality through examples. REFERENCES [1] J. K. Hart and K. Martinez, Environmental sensor networks: A revolution in the earth system science? Earth-Science Reviews, vol. 78, pp , 6. [2] G. Mariethoz and P. Renard, Reconstruction of incomplete data sets or images using direct sampling, Mathematical Geosciences, vol. 42, no. 3, pp ,. [3] A. Tamburini, M. Bianchi, C. Giannico, and F. Novali, Retrieving surface deformation by PSInSAR technology: A powerful tool in reservoir monitoring, International Journal of Greenhouse Gas Control, vol. 4, pp ,. [4] F. Fazel, M. Fazel, and M. Stojanovic, Random access compressed sensing for energy-efficient underwater sensor networks, IEEE Journal on Selected Areas in Communications (JSAC), vol. 29, no. 8, Sept. 11. [], Compressed sensing in random access networks with applications to underwater monitoring, Physical Communication (Elsevier) Journal, 11. [6] B. Recht, M. Fazel, and P. Parrilo, Guaranteed minimum-rank solutions of linear matrix equations via nuclear norm minimization, vol. 2, no. 3,, pp [7] D. Gross, Recovering low-rank matrices from few coefficients in any basis, IEEE Transactions on Information Theory, vol. 7, no. 3, pp , March 11. [8] E. J. Cands and B. Recht, Exact matrix completion via convex optimization, vol. 9, 9, pp [9] E. J. Cands and Y. Plan, Matrix completion with noise, vol. 98, no. 6, 9, pp [] L. Balzano, R. Nowak, and B. Recht, Online identification and tracking of subspaces from highly incomplete information, in Proceedings of the Allerton Conference on Communication, Control and Computing, September. [11] J. J. Meng, W. Yin, H. Li, E. Houssain, and Z. Han, Collaborative spectrum sensing from sparse observations using matrix completion for cognitive radio networks, in Proceedings of the ICASSP,. [12] A. Majumdar and R. K. Ward, Increasing energy efficiency in sensor networks: blue noise sampling and non-convex matrix completion, International Journal of Sensor Networks, vol. 9, no. 3-4, pp , 11. [13] Z. Wen, W. Yin, and Y. Zhang, Solving a low-rank factorization model for matrix completion by a nonlinear successive over-relaxation algorithm, Rice University CAAM, Tech. Rep.,.
Sparse Sensing in Colocated MIMO Radar: A Matrix Completion Approach
Sparse Sensing in Colocated MIMO Radar: A Matrix Completion Approach Athina P. Petropulu Department of Electrical and Computer Engineering Rutgers, the State University of New Jersey Acknowledgments Shunqiao
More informationLow-rank Matrix Completion with Noisy Observations: a Quantitative Comparison
Low-rank Matrix Completion with Noisy Observations: a Quantitative Comparison Raghunandan H. Keshavan, Andrea Montanari and Sewoong Oh Electrical Engineering and Statistics Department Stanford University,
More informationApplication of Tensor and Matrix Completion on Environmental Sensing Data
Application of Tensor and Matrix Completion on Environmental Sensing Data Michalis Giannopoulos 1,, Sofia Savvaki 1,, Grigorios Tsagkatakis 1, and Panagiotis Tsakalides 1, 1- Institute of Computer Science
More informationECE 8201: Low-dimensional Signal Models for High-dimensional Data Analysis
ECE 8201: Low-dimensional Signal Models for High-dimensional Data Analysis Lecture 7: Matrix completion Yuejie Chi The Ohio State University Page 1 Reference Guaranteed Minimum-Rank Solutions of Linear
More informationSelf-Calibration and Biconvex Compressive Sensing
Self-Calibration and Biconvex Compressive Sensing Shuyang Ling Department of Mathematics, UC Davis July 12, 2017 Shuyang Ling (UC Davis) SIAM Annual Meeting, 2017, Pittsburgh July 12, 2017 1 / 22 Acknowledgements
More informationCompressed Sensing and Robust Recovery of Low Rank Matrices
Compressed Sensing and Robust Recovery of Low Rank Matrices M. Fazel, E. Candès, B. Recht, P. Parrilo Electrical Engineering, University of Washington Applied and Computational Mathematics Dept., Caltech
More informationCovariance Sketching via Quadratic Sampling
Covariance Sketching via Quadratic Sampling Yuejie Chi Department of ECE and BMI The Ohio State University Tsinghua University June 2015 Page 1 Acknowledgement Thanks to my academic collaborators on some
More informationRobust Principal Component Analysis
ELE 538B: Mathematics of High-Dimensional Data Robust Principal Component Analysis Yuxin Chen Princeton University, Fall 2018 Disentangling sparse and low-rank matrices Suppose we are given a matrix M
More informationCombining Sparsity with Physically-Meaningful Constraints in Sparse Parameter Estimation
UIUC CSL Mar. 24 Combining Sparsity with Physically-Meaningful Constraints in Sparse Parameter Estimation Yuejie Chi Department of ECE and BMI Ohio State University Joint work with Yuxin Chen (Stanford).
More informationCoprime Coarray Interpolation for DOA Estimation via Nuclear Norm Minimization
Coprime Coarray Interpolation for DOA Estimation via Nuclear Norm Minimization Chun-Lin Liu 1 P. P. Vaidyanathan 2 Piya Pal 3 1,2 Dept. of Electrical Engineering, MC 136-93 California Institute of Technology,
More informationAnalysis of Random Radar Networks
Analysis of Random Radar Networks Rani Daher, Ravira Adve Department of Electrical and Computer Engineering, University of Toronto 1 King s College Road, Toronto, ON M5S3G4 Email: rani.daher@utoronto.ca,
More informationCooperative Spectrum Sensing for Cognitive Radios under Bandwidth Constraints
Cooperative Spectrum Sensing for Cognitive Radios under Bandwidth Constraints Chunhua Sun, Wei Zhang, and haled Ben Letaief, Fellow, IEEE Department of Electronic and Computer Engineering The Hong ong
More informationRecovery of Low-Rank Plus Compressed Sparse Matrices with Application to Unveiling Traffic Anomalies
July 12, 212 Recovery of Low-Rank Plus Compressed Sparse Matrices with Application to Unveiling Traffic Anomalies Morteza Mardani Dept. of ECE, University of Minnesota, Minneapolis, MN 55455 Acknowledgments:
More informationUser s Guide for LMaFit: Low-rank Matrix Fitting
User s Guide for LMaFit: Low-rank Matrix Fitting Yin Zhang Department of CAAM Rice University, Houston, Texas, 77005 (CAAM Technical Report TR09-28) (Versions beta-1: August 23, 2009) Abstract This User
More informationA Power Efficient Sensing/Communication Scheme: Joint Source-Channel-Network Coding by Using Compressive Sensing
Forty-Ninth Annual Allerton Conference Allerton House, UIUC, Illinois, USA September 28-30, 20 A Power Efficient Sensing/Communication Scheme: Joint Source-Channel-Network Coding by Using Compressive Sensing
More informationThree Generalizations of Compressed Sensing
Thomas Blumensath School of Mathematics The University of Southampton June, 2010 home prev next page Compressed Sensing and beyond y = Φx + e x R N or x C N x K is K-sparse and x x K 2 is small y R M or
More informationGoing off the grid. Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison
Going off the grid Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison Joint work with Badri Bhaskar Parikshit Shah Gonnguo Tang We live in a continuous world... But we work
More informationRapid, Robust, and Reliable Blind Deconvolution via Nonconvex Optimization
Rapid, Robust, and Reliable Blind Deconvolution via Nonconvex Optimization Shuyang Ling Department of Mathematics, UC Davis Oct.18th, 2016 Shuyang Ling (UC Davis) 16w5136, Oaxaca, Mexico Oct.18th, 2016
More informationSolving Corrupted Quadratic Equations, Provably
Solving Corrupted Quadratic Equations, Provably Yuejie Chi London Workshop on Sparse Signal Processing September 206 Acknowledgement Joint work with Yuanxin Li (OSU), Huishuai Zhuang (Syracuse) and Yingbin
More informationECE G: Special Topics in Signal Processing: Sparsity, Structure, and Inference
ECE 18-898G: Special Topics in Signal Processing: Sparsity, Structure, and Inference Low-rank matrix recovery via convex relaxations Yuejie Chi Department of Electrical and Computer Engineering Spring
More informationMassive MIMO: Signal Structure, Efficient Processing, and Open Problems II
Massive MIMO: Signal Structure, Efficient Processing, and Open Problems II Mahdi Barzegar Communications and Information Theory Group (CommIT) Technische Universität Berlin Heisenberg Communications and
More informationDistributed Detection and Estimation in Wireless Sensor Networks: Resource Allocation, Fusion Rules, and Network Security
Distributed Detection and Estimation in Wireless Sensor Networks: Resource Allocation, Fusion Rules, and Network Security Edmond Nurellari The University of Leeds, UK School of Electronic and Electrical
More informationADAPTIVE CLUSTERING ALGORITHM FOR COOPERATIVE SPECTRUM SENSING IN MOBILE ENVIRONMENTS. Jesus Perez and Ignacio Santamaria
ADAPTIVE CLUSTERING ALGORITHM FOR COOPERATIVE SPECTRUM SENSING IN MOBILE ENVIRONMENTS Jesus Perez and Ignacio Santamaria Advanced Signal Processing Group, University of Cantabria, Spain, https://gtas.unican.es/
More informationSparse Parameter Estimation: Compressed Sensing meets Matrix Pencil
Sparse Parameter Estimation: Compressed Sensing meets Matrix Pencil Yuejie Chi Departments of ECE and BMI The Ohio State University Colorado School of Mines December 9, 24 Page Acknowledgement Joint work
More informationCompressive Sensing and Beyond
Compressive Sensing and Beyond Sohail Bahmani Gerorgia Tech. Signal Processing Compressed Sensing Signal Models Classics: bandlimited The Sampling Theorem Any signal with bandwidth B can be recovered
More informationBlind Source Separation with a Time-Varying Mixing Matrix
Blind Source Separation with a Time-Varying Mixing Matrix Marcus R DeYoung and Brian L Evans Department of Electrical and Computer Engineering The University of Texas at Austin 1 University Station, Austin,
More informationMatrix Completion for Structured Observations
Matrix Completion for Structured Observations Denali Molitor Department of Mathematics University of California, Los ngeles Los ngeles, C 90095, US Email: dmolitor@math.ucla.edu Deanna Needell Department
More informationBinary Compressive Sensing via Analog. Fountain Coding
Binary Compressive Sensing via Analog 1 Fountain Coding Mahyar Shirvanimoghaddam, Member, IEEE, Yonghui Li, Senior Member, IEEE, Branka Vucetic, Fellow, IEEE, and Jinhong Yuan, Senior Member, IEEE, arxiv:1508.03401v1
More informationThresholds for the Recovery of Sparse Solutions via L1 Minimization
Thresholds for the Recovery of Sparse Solutions via L Minimization David L. Donoho Department of Statistics Stanford University 39 Serra Mall, Sequoia Hall Stanford, CA 9435-465 Email: donoho@stanford.edu
More informationNovel spectrum sensing schemes for Cognitive Radio Networks
Novel spectrum sensing schemes for Cognitive Radio Networks Cantabria University Santander, May, 2015 Supélec, SCEE Rennes, France 1 The Advanced Signal Processing Group http://gtas.unican.es The Advanced
More informationExact Low-rank Matrix Recovery via Nonconvex M p -Minimization
Exact Low-rank Matrix Recovery via Nonconvex M p -Minimization Lingchen Kong and Naihua Xiu Department of Applied Mathematics, Beijing Jiaotong University, Beijing, 100044, People s Republic of China E-mail:
More informationarxiv: v1 [cs.it] 21 Feb 2013
q-ary Compressive Sensing arxiv:30.568v [cs.it] Feb 03 Youssef Mroueh,, Lorenzo Rosasco, CBCL, CSAIL, Massachusetts Institute of Technology LCSL, Istituto Italiano di Tecnologia and IIT@MIT lab, Istituto
More informationStopping Condition for Greedy Block Sparse Signal Recovery
Stopping Condition for Greedy Block Sparse Signal Recovery Yu Luo, Ronggui Xie, Huarui Yin, and Weidong Wang Department of Electronics Engineering and Information Science, University of Science and Technology
More informationThe convex algebraic geometry of rank minimization
The convex algebraic geometry of rank minimization Pablo A. Parrilo Laboratory for Information and Decision Systems Massachusetts Institute of Technology International Symposium on Mathematical Programming
More informationACCORDING to Shannon s sampling theorem, an analog
554 IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL 59, NO 2, FEBRUARY 2011 Segmented Compressed Sampling for Analog-to-Information Conversion: Method and Performance Analysis Omid Taheri, Student Member,
More informationarxiv: v2 [cs.it] 12 Jul 2011
Online Identification and Tracking of Subspaces from Highly Incomplete Information Laura Balzano, Robert Nowak and Benjamin Recht arxiv:1006.4046v2 [cs.it] 12 Jul 2011 Department of Electrical and Computer
More informationOn the Relationship between Transmission Power and Capacity of an Underwater Acoustic Communication Channel
On the Relationship between Transmission Power and Capacity of an Underwater Acoustic Communication Channel Daniel E. Lucani LIDS, MIT Cambridge, Massachusetts, 02139 Email: dlucani@mit.edu Milica Stojanovic
More informationWireless Compressive Sensing for Energy Harvesting Sensor Nodes over Fading Channels
Wireless Compressive Sensing for Energy Harvesting Sensor Nodes over Fading Channels Gang Yang, Vincent Y. F. Tan, Chin Keong Ho, See Ho Ting and Yong Liang Guan School of Electrical and Electronic Engineering,
More informationBreaking the Limits of Subspace Inference
Breaking the Limits of Subspace Inference Claudia R. Solís-Lemus, Daniel L. Pimentel-Alarcón Emory University, Georgia State University Abstract Inferring low-dimensional subspaces that describe high-dimensional,
More informationA Nonuniform Quantization Scheme for High Speed SAR ADC Architecture
A Nonuniform Quantization Scheme for High Speed SAR ADC Architecture Youngchun Kim Electrical and Computer Engineering The University of Texas Wenjuan Guo Intel Corporation Ahmed H Tewfik Electrical and
More informationFinal Report for DOEI Project: Bottom Interaction in Long Range Acoustic Propagation
Final Report for DOEI Project: Bottom Interaction in Long Range Acoustic Propagation Ralph A. Stephen Woods Hole Oceanographic Institution 360 Woods Hole Road (MS#24) Woods Hole, MA 02543 phone: (508)
More informationTransmission Schemes for Lifetime Maximization in Wireless Sensor Networks: Uncorrelated Source Observations
Transmission Schemes for Lifetime Maximization in Wireless Sensor Networks: Uncorrelated Source Observations Xiaolu Zhang, Meixia Tao and Chun Sum Ng Department of Electrical and Computer Engineering National
More informationOn the Relationship between Transmission Power and Capacity of an Underwater Acoustic Communication Channel
On the Relationship between Transmission Power and Capacity of an Underwater Acoustic Communication Channel Daniel E. Lucani LIDS, MIT Cambridge, Massachusetts, 02139 Email: dlucani@mit.edu Milica Stojanovic
More informationROBUST BLIND SPIKES DECONVOLUTION. Yuejie Chi. Department of ECE and Department of BMI The Ohio State University, Columbus, Ohio 43210
ROBUST BLIND SPIKES DECONVOLUTION Yuejie Chi Department of ECE and Department of BMI The Ohio State University, Columbus, Ohio 4 ABSTRACT Blind spikes deconvolution, or blind super-resolution, deals with
More informationContinuous-Model Communication Complexity with Application in Distributed Resource Allocation in Wireless Ad hoc Networks
Continuous-Model Communication Complexity with Application in Distributed Resource Allocation in Wireless Ad hoc Networks Husheng Li 1 and Huaiyu Dai 2 1 Department of Electrical Engineering and Computer
More informationTARGET DETECTION WITH FUNCTION OF COVARIANCE MATRICES UNDER CLUTTER ENVIRONMENT
TARGET DETECTION WITH FUNCTION OF COVARIANCE MATRICES UNDER CLUTTER ENVIRONMENT Feng Lin, Robert C. Qiu, James P. Browning, Michael C. Wicks Cognitive Radio Institute, Department of Electrical and Computer
More informationDIFFUSION-BASED DISTRIBUTED MVDR BEAMFORMER
14 IEEE International Conference on Acoustic, Speech and Signal Processing (ICASSP) DIFFUSION-BASED DISTRIBUTED MVDR BEAMFORMER Matt O Connor 1 and W. Bastiaan Kleijn 1,2 1 School of Engineering and Computer
More informationOnline Identification and Tracking of Subspaces from Highly Incomplete Information
Online Identification and Tracking of Subspaces from Highly Incomplete Information Laura Balzano, Robert Nowak and Benjamin Recht Department of Electrical and Computer Engineering. University of Wisconsin-Madison
More informationProbabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms
Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms François Caron Department of Statistics, Oxford STATLEARN 2014, Paris April 7, 2014 Joint work with Adrien Todeschini,
More informationA Cross-Associative Neural Network for SVD of Nonsquared Data Matrix in Signal Processing
IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 12, NO. 5, SEPTEMBER 2001 1215 A Cross-Associative Neural Network for SVD of Nonsquared Data Matrix in Signal Processing Da-Zheng Feng, Zheng Bao, Xian-Da Zhang
More informationSPARSE signal representations have gained popularity in recent
6958 IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 57, NO. 10, OCTOBER 2011 Blind Compressed Sensing Sivan Gleichman and Yonina C. Eldar, Senior Member, IEEE Abstract The fundamental principle underlying
More informationTransmitter-Receiver Cooperative Sensing in MIMO Cognitive Network with Limited Feedback
IEEE INFOCOM Workshop On Cognitive & Cooperative Networks Transmitter-Receiver Cooperative Sensing in MIMO Cognitive Network with Limited Feedback Chao Wang, Zhaoyang Zhang, Xiaoming Chen, Yuen Chau. Dept.of
More informationMultiple Bits Distributed Moving Horizon State Estimation for Wireless Sensor Networks. Ji an Luo
Multiple Bits Distributed Moving Horizon State Estimation for Wireless Sensor Networks Ji an Luo 2008.6.6 Outline Background Problem Statement Main Results Simulation Study Conclusion Background Wireless
More informationEUSIPCO
EUSIPCO 013 1569746769 SUBSET PURSUIT FOR ANALYSIS DICTIONARY LEARNING Ye Zhang 1,, Haolong Wang 1, Tenglong Yu 1, Wenwu Wang 1 Department of Electronic and Information Engineering, Nanchang University,
More informationLarge-Scale L1-Related Minimization in Compressive Sensing and Beyond
Large-Scale L1-Related Minimization in Compressive Sensing and Beyond Yin Zhang Department of Computational and Applied Mathematics Rice University, Houston, Texas, U.S.A. Arizona State University March
More informationNew Coherence and RIP Analysis for Weak. Orthogonal Matching Pursuit
New Coherence and RIP Analysis for Wea 1 Orthogonal Matching Pursuit Mingrui Yang, Member, IEEE, and Fran de Hoog arxiv:1405.3354v1 [cs.it] 14 May 2014 Abstract In this paper we define a new coherence
More informationAnalysis of Robust PCA via Local Incoherence
Analysis of Robust PCA via Local Incoherence Huishuai Zhang Department of EECS Syracuse University Syracuse, NY 3244 hzhan23@syr.edu Yi Zhou Department of EECS Syracuse University Syracuse, NY 3244 yzhou35@syr.edu
More informationSeismic data interpolation and denoising using SVD-free low-rank matrix factorization
Seismic data interpolation and denoising using SVD-free low-rank matrix factorization R. Kumar, A.Y. Aravkin,, H. Mansour,, B. Recht and F.J. Herrmann Dept. of Earth and Ocean sciences, University of British
More informationRank Determination for Low-Rank Data Completion
Journal of Machine Learning Research 18 017) 1-9 Submitted 7/17; Revised 8/17; Published 9/17 Rank Determination for Low-Rank Data Completion Morteza Ashraphijuo Columbia University New York, NY 1007,
More informationPHASE RETRIEVAL OF SPARSE SIGNALS FROM MAGNITUDE INFORMATION. A Thesis MELTEM APAYDIN
PHASE RETRIEVAL OF SPARSE SIGNALS FROM MAGNITUDE INFORMATION A Thesis by MELTEM APAYDIN Submitted to the Office of Graduate and Professional Studies of Texas A&M University in partial fulfillment of the
More informationIEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 57, NO. 11, NOVEMBER On the Performance of Sparse Recovery
IEEE TRANSACTIONS ON INFORMATION THEORY, VOL. 57, NO. 11, NOVEMBER 2011 7255 On the Performance of Sparse Recovery Via `p-minimization (0 p 1) Meng Wang, Student Member, IEEE, Weiyu Xu, and Ao Tang, Senior
More informationMatrix Completion: Fundamental Limits and Efficient Algorithms
Matrix Completion: Fundamental Limits and Efficient Algorithms Sewoong Oh PhD Defense Stanford University July 23, 2010 1 / 33 Matrix completion Find the missing entries in a huge data matrix 2 / 33 Example
More informationPerformance Analysis for Strong Interference Remove of Fast Moving Target in Linear Array Antenna
Performance Analysis for Strong Interference Remove of Fast Moving Target in Linear Array Antenna Kwan Hyeong Lee Dept. Electriacal Electronic & Communicaton, Daejin University, 1007 Ho Guk ro, Pochen,Gyeonggi,
More informationROBUST BLIND CALIBRATION VIA TOTAL LEAST SQUARES
ROBUST BLIND CALIBRATION VIA TOTAL LEAST SQUARES John Lipor Laura Balzano University of Michigan, Ann Arbor Department of Electrical and Computer Engineering {lipor,girasole}@umich.edu ABSTRACT This paper
More informationGuaranteed Rank Minimization via Singular Value Projection
3 4 5 6 7 8 9 3 4 5 6 7 8 9 3 4 5 6 7 8 9 3 3 3 33 34 35 36 37 38 39 4 4 4 43 44 45 46 47 48 49 5 5 5 53 Guaranteed Rank Minimization via Singular Value Projection Anonymous Author(s) Affiliation Address
More informationSparse and Low Rank Recovery via Null Space Properties
Sparse and Low Rank Recovery via Null Space Properties Holger Rauhut Lehrstuhl C für Mathematik (Analysis), RWTH Aachen Convexity, probability and discrete structures, a geometric viewpoint Marne-la-Vallée,
More informationDistributed Power Control for Time Varying Wireless Networks: Optimality and Convergence
Distributed Power Control for Time Varying Wireless Networks: Optimality and Convergence Tim Holliday, Nick Bambos, Peter Glynn, Andrea Goldsmith Stanford University Abstract This paper presents a new
More informationSubspace Projection Matrix Completion on Grassmann Manifold
Subspace Projection Matrix Completion on Grassmann Manifold Xinyue Shen and Yuantao Gu Dept. EE, Tsinghua University, Beijing, China http://gu.ee.tsinghua.edu.cn/ ICASSP 2015, Brisbane Contents 1 Background
More informationRecovery of Low Rank and Jointly Sparse. Matrices with Two Sampling Matrices
Recovery of Low Rank and Jointly Sparse 1 Matrices with Two Sampling Matrices Sampurna Biswas, Hema K. Achanta, Mathews Jacob, Soura Dasgupta, and Raghuraman Mudumbai Abstract We provide a two-step approach
More informationApplications of Robust Optimization in Signal Processing: Beamforming and Power Control Fall 2012
Applications of Robust Optimization in Signal Processing: Beamforg and Power Control Fall 2012 Instructor: Farid Alizadeh Scribe: Shunqiao Sun 12/09/2012 1 Overview In this presentation, we study the applications
More informationHigh-Rank Matrix Completion and Subspace Clustering with Missing Data
High-Rank Matrix Completion and Subspace Clustering with Missing Data Authors: Brian Eriksson, Laura Balzano and Robert Nowak Presentation by Wang Yuxiang 1 About the authors
More informationOptimisation Combinatoire et Convexe.
Optimisation Combinatoire et Convexe. Low complexity models, l 1 penalties. A. d Aspremont. M1 ENS. 1/36 Today Sparsity, low complexity models. l 1 -recovery results: three approaches. Extensions: matrix
More informationCoherent imaging without phases
Coherent imaging without phases Miguel Moscoso Joint work with Alexei Novikov Chrysoula Tsogka and George Papanicolaou Waves and Imaging in Random Media, September 2017 Outline 1 The phase retrieval problem
More informationConstructing Explicit RIP Matrices and the Square-Root Bottleneck
Constructing Explicit RIP Matrices and the Square-Root Bottleneck Ryan Cinoman July 18, 2018 Ryan Cinoman Constructing Explicit RIP Matrices July 18, 2018 1 / 36 Outline 1 Introduction 2 Restricted Isometry
More informationCompressed sensing. Or: the equation Ax = b, revisited. Terence Tao. Mahler Lecture Series. University of California, Los Angeles
Or: the equation Ax = b, revisited University of California, Los Angeles Mahler Lecture Series Acquiring signals Many types of real-world signals (e.g. sound, images, video) can be viewed as an n-dimensional
More informationMatrix completion: Fundamental limits and efficient algorithms. Sewoong Oh Stanford University
Matrix completion: Fundamental limits and efficient algorithms Sewoong Oh Stanford University 1 / 35 Low-rank matrix completion Low-rank Data Matrix Sparse Sampled Matrix Complete the matrix from small
More informationCompressed Sensing and Sparse Recovery
ELE 538B: Sparsity, Structure and Inference Compressed Sensing and Sparse Recovery Yuxin Chen Princeton University, Spring 217 Outline Restricted isometry property (RIP) A RIPless theory Compressed sensing
More informationQUANTIZATION FOR DISTRIBUTED ESTIMATION IN LARGE SCALE SENSOR NETWORKS
QUANTIZATION FOR DISTRIBUTED ESTIMATION IN LARGE SCALE SENSOR NETWORKS Parvathinathan Venkitasubramaniam, Gökhan Mergen, Lang Tong and Ananthram Swami ABSTRACT We study the problem of quantization for
More informationMatrix Completion from a Few Entries
Matrix Completion from a Few Entries Raghunandan H. Keshavan and Sewoong Oh EE Department Stanford University, Stanford, CA 9434 Andrea Montanari EE and Statistics Departments Stanford University, Stanford,
More informationProbabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms
Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms Adrien Todeschini Inria Bordeaux JdS 2014, Rennes Aug. 2014 Joint work with François Caron (Univ. Oxford), Marie
More informationExploiting Sparsity for Wireless Communications
Exploiting Sparsity for Wireless Communications Georgios B. Giannakis Dept. of ECE, Univ. of Minnesota http://spincom.ece.umn.edu Acknowledgements: D. Angelosante, J.-A. Bazerque, H. Zhu; and NSF grants
More informationEstimation of the Optimum Rotational Parameter for the Fractional Fourier Transform Using Domain Decomposition
Estimation of the Optimum Rotational Parameter for the Fractional Fourier Transform Using Domain Decomposition Seema Sud 1 1 The Aerospace Corporation, 4851 Stonecroft Blvd. Chantilly, VA 20151 Abstract
More informationParameter Estimation for Mixture Models via Convex Optimization
Parameter Estimation for Mixture Models via Convex Optimization Yuanxin Li Department of Electrical and Computer Engineering The Ohio State University Columbus Ohio 432 Email: li.3822@osu.edu Yuejie Chi
More informationEXTENSION OF NESTED ARRAYS WITH THE FOURTH-ORDER DIFFERENCE CO-ARRAY ENHANCEMENT
EXTENSION OF NESTED ARRAYS WITH THE FOURTH-ORDER DIFFERENCE CO-ARRAY ENHANCEMENT Qing Shen,2, Wei Liu 2, Wei Cui, Siliang Wu School of Information and Electronics, Beijing Institute of Technology Beijing,
More informationTRANSMISSION STRATEGIES FOR SINGLE-DESTINATION WIRELESS NETWORKS
The 20 Military Communications Conference - Track - Waveforms and Signal Processing TRANSMISSION STRATEGIES FOR SINGLE-DESTINATION WIRELESS NETWORKS Gam D. Nguyen, Jeffrey E. Wieselthier 2, Sastry Kompella,
More informationShallow Water Fluctuations and Communications
Shallow Water Fluctuations and Communications H.C. Song Marine Physical Laboratory Scripps Institution of oceanography La Jolla, CA 92093-0238 phone: (858) 534-0954 fax: (858) 534-7641 email: hcsong@mpl.ucsd.edu
More informationSIGNAL STRENGTH LOCALIZATION BOUNDS IN AD HOC & SENSOR NETWORKS WHEN TRANSMIT POWERS ARE RANDOM. Neal Patwari and Alfred O.
SIGNAL STRENGTH LOCALIZATION BOUNDS IN AD HOC & SENSOR NETWORKS WHEN TRANSMIT POWERS ARE RANDOM Neal Patwari and Alfred O. Hero III Department of Electrical Engineering & Computer Science University of
More informationSPECTRUM SHARING IN WIRELESS NETWORKS: A QOS-AWARE SECONDARY MULTICAST APPROACH WITH WORST USER PERFORMANCE OPTIMIZATION
SPECTRUM SHARING IN WIRELESS NETWORKS: A QOS-AWARE SECONDARY MULTICAST APPROACH WITH WORST USER PERFORMANCE OPTIMIZATION Khoa T. Phan, Sergiy A. Vorobyov, Nicholas D. Sidiropoulos, and Chintha Tellambura
More informationInformation-Theoretic Limits of Matrix Completion
Information-Theoretic Limits of Matrix Completion Erwin Riegler, David Stotz, and Helmut Bölcskei Dept. IT & EE, ETH Zurich, Switzerland Email: {eriegler, dstotz, boelcskei}@nari.ee.ethz.ch Abstract We
More informationRecent advances in microelectromechanical
SENSOR AND ACTUATOR NETWORKS A Learning-Theory Approach to Sensor Networks Supervised learning might be a viable approach to sensor network applications. Preliminary research shows that a well-known algorithm
More informationWireless Compressive Sensing for Energy Harvesting Sensor Nodes
IEEE TRANSACTIONS ON SIGNAL PROCESSING, VOL. 61, NO. 18, SEPTEMBER 15, 2013 4491 Wireless Compressive Sensing for Energy Harvesting Sensor Nodes Gang Yang, Student Member, IEEE, VincentY.F.Tan, Member,
More informationGuaranteed Minimum-Rank Solutions of Linear Matrix Equations via Nuclear Norm Minimization
Forty-Fifth Annual Allerton Conference Allerton House, UIUC, Illinois, USA September 26-28, 27 WeA3.2 Guaranteed Minimum-Rank Solutions of Linear Matrix Equations via Nuclear Norm Minimization Benjamin
More informationTHE estimation of covariance matrices is a crucial component
1 A Subspace Method for Array Covariance Matrix Estimation Mostafa Rahmani and George K. Atia, Member, IEEE, arxiv:1411.0622v1 [cs.na] 20 Oct 2014 Abstract This paper introduces a subspace method for the
More informationCo-prime Arrays with Reduced Sensors (CARS) for Direction-of-Arrival Estimation
Co-prime Arrays with Reduced Sensors (CARS) for Direction-of-Arrival Estimation Mingyang Chen 1,LuGan and Wenwu Wang 1 1 Department of Electrical and Electronic Engineering, University of Surrey, U.K.
More informationShallow Water Fluctuations and Communications
Shallow Water Fluctuations and Communications H.C. Song Marine Physical Laboratory Scripps Institution of oceanography La Jolla, CA 92093-0238 phone: (858) 534-0954 fax: (858) 534-7641 email: hcsong@mpl.ucsd.edu
More informationEfficient Algorithms for Pulse Parameter Estimation, Pulse Peak Localization And Pileup Reduction in Gamma Ray Spectroscopy M.W.Raad 1, L.
Efficient Algorithms for Pulse Parameter Estimation, Pulse Peak Localization And Pileup Reduction in Gamma Ray Spectroscopy M.W.Raad 1, L. Cheded 2 1 Computer Engineering Department, 2 Systems Engineering
More informationConditions for Robust Principal Component Analysis
Rose-Hulman Undergraduate Mathematics Journal Volume 12 Issue 2 Article 9 Conditions for Robust Principal Component Analysis Michael Hornstein Stanford University, mdhornstein@gmail.com Follow this and
More informationOn the Projection Matrices Influence in the Classification of Compressed Sensed ECG Signals
On the Projection Matrices Influence in the Classification of Compressed Sensed ECG Signals Monica Fira, Liviu Goras Institute of Computer Science Romanian Academy Iasi, Romania Liviu Goras, Nicolae Cleju,
More informationHYPOTHESIS TESTING OVER A RANDOM ACCESS CHANNEL IN WIRELESS SENSOR NETWORKS
HYPOTHESIS TESTING OVER A RANDOM ACCESS CHANNEL IN WIRELESS SENSOR NETWORKS Elvis Bottega,, Petar Popovski, Michele Zorzi, Hiroyuki Yomo, and Ramjee Prasad Center for TeleInFrastructure (CTIF), Aalborg
More informationSpectral k-support Norm Regularization
Spectral k-support Norm Regularization Andrew McDonald Department of Computer Science, UCL (Joint work with Massimiliano Pontil and Dimitris Stamos) 25 March, 2015 1 / 19 Problem: Matrix Completion Goal:
More information