Results 1 
4 of
4
1SUPER: Sparse signals with Unknown Phases Efficiently Recovered
"... Abstract—Compressive phase retrieval algorithms attempt to reconstruct a “sparse highdimensional vector ” from its “lowdimensional intensity measurements”. Suppose x is any lengthn input vector over C with exactly k nonzero entries, and A is an m × n (k < m n) phase measurement matrix over C ..."
Abstract

Cited by 1 (0 self)
 Add to MetaCart
(Show Context)
Abstract—Compressive phase retrieval algorithms attempt to reconstruct a “sparse highdimensional vector ” from its “lowdimensional intensity measurements”. Suppose x is any lengthn input vector over C with exactly k nonzero entries, and A is an m × n (k < m n) phase measurement matrix over C. The decoder is handed m “intensity measurements” (A1x,..., Amx) (corresponding to componentwise absolute values of the linear measurement Ax) – here Ai’s correspond to the rows of the measurement matrix A. In this work, we present a class of measurement matrices A, and a corresponding decoding algorithm that we call SUPER, which can reconstruct x up to a global phase from intensity measurements. The SUPER algorithm is the first to simultaneously have the following properties: (a) it requires only O(k) (orderoptimal) measurements, (b) the computational complexity of decoding is O(k log k) (near orderoptimal) arithmetic operations, (c) it succeeds with high probability over the design of A. Our results hold for all k ∈ {1, 2,..., n}. I.
Sublinear Time Compressed Sensing for Support Recovery using SparseGraph Codes
, 2015
"... We address the problem of robustly recovering the support of highdimensional sparse signals1 from linear measurements in a lowdimensional subspace. We introduce a new compressed sensing framework through carefully designed sparse measurement matrices associated with low measurement costs and lowc ..."
Abstract
 Add to MetaCart
We address the problem of robustly recovering the support of highdimensional sparse signals1 from linear measurements in a lowdimensional subspace. We introduce a new compressed sensing framework through carefully designed sparse measurement matrices associated with low measurement costs and lowcomplexity recovery algorithms. The measurement system in our framework captures observations of the signal through welldesigned measurement matrices sparsified by capacityapproaching sparsegraph codes, and then recovers the signal by using a simple peeling decoder. As a result, we can simultaneously reduce both the measurement cost and the computational complexity. In this paper, we formally connect general sparse recovery problems in compressed sensing with sparsegraph decoding in packetcommunication systems, and analyze our design in terms of the measurement cost, computational complexity and recovery performance. Specifically, by structuring the measurements through sparsegraph codes, we propose two families of measurement matrices, the Fourier family and the binary family respectively, which lead to different measurement and computational costs. In the noiseless setting, our framework recovers the sparse support of any Ksparse signal in time2 O(K) with 2K measurements obtained by the Fourier family, or in time O(K logN) using K log2N + K measurements obtained by the binary family. In the presence of noise, both measurement and
1“InformationFriction ” for Noiseless Compressive Sensing Decoding Researchers
"... The fundamental problem considered in this paper is, basically, “what is the energy consumed for the implementation of compressive sensing algorithm on a circuit? " We use the proposed “bitmeters1 " measure as a proportional measurement for energy, i.e., the product of number bits transmi ..."
Abstract
 Add to MetaCart
(Show Context)
The fundamental problem considered in this paper is, basically, “what is the energy consumed for the implementation of compressive sensing algorithm on a circuit? " We use the proposed “bitmeters1 " measure as a proportional measurement for energy, i.e., the product of number bits transmitted and the distance of information transport. Be Analogous to the friction imposed on relative motion between two surfaces, this model is socalled “informationfriction " model. By using this “informationfriction " model, a fundamental lower bound for the implementation of compressive sensing algorithms on a circuit is provided. Further, we explore and compare a series of decoding algorithms based on different implementationcircuits. As our second main result, an ordertight asymptotic result on a fixed precision bits for the regime m = O(k) (m is the number of measurements and k is the number of nonzero entries in the compressible vector) is provided. We thus attest in this paper that the asymptotic lower bound is orderoptimal with a sublinear sparsity k such that k = n1−β (β ∈ (0, 1), n is the total number of input entries) since a proposed algorithm with corresponding construction of implementationcircuit can achieve an upper bound with the same order.
FRANTIC: A Fast Referencebased Algorithm for Network Tomography via Compressive Sensing
"... Abstract—We study the problem of link and node delay estimation in undirected networks when at most k out of n links or nodes in the network are congested. Our approach relies on endtoend measurements of path delays across prespecified paths in the network. We present a class of algorithms that w ..."
Abstract
 Add to MetaCart
(Show Context)
Abstract—We study the problem of link and node delay estimation in undirected networks when at most k out of n links or nodes in the network are congested. Our approach relies on endtoend measurements of path delays across prespecified paths in the network. We present a class of algorithms that we call FRANTIC. The FRANTIC algorithms are motivated by compressive sensing; however, unlike traditional compressive sensing, the measurement design here is constrained by the network topology and the matrix entries are constrained to be positive integers. A key component of our design is a new compressive sensing algorithm SHOFAINT that is related to the SHOFA algorithm [1] for compressive sensing, but unlike SHOFA, the matrix entries here are drawn from the set of integers {0, 1,...,M}. We show that O(k logn / logM) measurements suffice both for SHOFAINT and FRANTIC. Further, we show that the computational complexity of decoding is also O(k logn / logM) for each of these algorithms. Finally, we look at efficient constructions of the measurement operations through Steiner Trees. I.