Information Theory
New submissions
[ showing up to 2000 entries per page: fewer  more ]
New submissions for Fri, 15 Jan 21
 [1] arXiv:2101.05373 [pdf, other]

Title: On a Class of TimeVarying Gaussian ISI ChannelsAuthors: Kamyar MoshksarSubjects: Information Theory (cs.IT)
This paper studies a class of stochastic and timevarying Gaussian intersymbol interference (ISI) channels. The $i^{th}$ channel tap during time slot $t$ is uniformly distributed over an interval of centre $c_i$ and radius $ r_{i}$. The array of channel taps is independent along both $t$ and $i$. The channel state information is unavailable at both the transmitter and the receiver. Lower and upper bounds are derived on the WhiteGaussianInput (WGI) capacity $C_{\scriptscriptstyle{WGI}}$ for arbitrary values of the radii $ r_i$. It is shown that $C_{\scriptscriptstyle{WGI}}$ does not scale with the average input power. The proposed lower bound is achieved by a jointtypicality decoder that is tuned to a set of candidates for the channel matrix. This set forms a net that covers the range of the random channel matrix and its resolution is optimized in order to yield the largest achievable rate. Tools in matrix analysis such as Weyl's inequality on perturbation of eigenvalues of symmetric matrices are used in order to analyze the probability of error.
 [2] arXiv:2101.05465 [pdf, ps, other]

Title: Noise Is Useful: Exploiting Data Diversity for Edge IntelligenceComments: 5 pages, 6 figures, to be presented at IEEE Communications LettersSubjects: Information Theory (cs.IT)
Edge intelligence requires to fast access distributed data samples generated by edge devices. The challenge is using limited radio resource to acquire massive data samples for training machine learning models at edge server. In this article, we propose a new communicationefficient edge intelligence scheme where the most useful data samples are selected to train the model. Here the usefulness or values of data samples is measured by data diversity which is defined as the difference between data samples. We derive a closeform expression of data diversity that combines data informativeness and channel quality. Then a joint dataandchannel diversity aware multiuser scheduling algorithm is proposed. We find that noise is useful for enhancing data diversity under some conditions.
 [3] arXiv:2101.05604 [pdf, ps, other]

Title: Decoding of Interleaved Linearized ReedSolomon Codes with Applications to Network CodingComments: 6 pages, 2 figures, submitted to ISIT 2021Subjects: Information Theory (cs.IT)
Recently, MartinezPenas and Kschischang (IEEE Trans. Inf. Theory, 2019) showed that lifted linearized ReedSolomon codes are suitable codes for error control in multishot network coding. We show how to construct and decode lifted interleaved linearized ReedSolomon codes. Compared to the construction by MartinezPenasKschischang, interleaving allows to increase the decoding region significantly (especially w.r.t. the number of insertions) and decreases the overhead due to the lifting (i.e., increases the code rate), at the cost of an increased packet size. The proposed decoder is a list decoder that can also be interpreted as a probabilistic unique decoder. Although our best upper bound on the list size is exponential, we present a heuristic argument and simulation results that indicate that the list size is in fact one for most channel realizations up to the maximal decoding radius.
 [4] arXiv:2101.05607 [pdf, other]

Title: Intelligent Reflecting Surfaces for ComputeandForwardSubjects: Information Theory (cs.IT); Networking and Internet Architecture (cs.NI)
In this paper, we show that using Intelligent Reflecting Surfaces (IRS) can enhance the computing capability of a wireless network scenario. We consider a Multiple Access Channel (MAC) where a number of users aim to send data to a Base Station (BS). The BS is interested in decoding a linear combination of the data from different users in the corresponding finite field. By focusing on the ComputeandForward framework, we show that through carefully choosing the IRS parameters, such a scenario's computation rate will be significantly improved. More specifically, we formulate an optimization problem to maximize the computation rate and tackle the problem via an alternating optimization (AO) approach. Our results confirm the usefulness of IRS technology for future wireless networks  such as 6G  with massive computation requirements.
 [5] arXiv:2101.05629 [pdf, ps, other]

Title: Offgrid Channel Estimation with Sparse Bayesian Learning for OTFS SystemsComments: 30 pages, 9 figures, submitted to IEEE Transactions on Wireless CommunicationsSubjects: Information Theory (cs.IT)
This paper proposes an offgrid channel estimation scheme for orthogonal timefrequency space (OTFS) systems adopting the sparse Bayesian learning (SBL) framework. To avoid channel spreading caused by the fractional delay and Doppler shifts and to fully exploit the channel sparsity in the delayDoppler (DD) domain, we estimate the original DD domain channel response rather than the effective DD domain channel response as commonly adopted in the literature. OTFS channel estimation is first formulated as a onedimensional (1D) offgrid sparse signal recovery (SSR) problem based on a virtual sampling grid defined in the DD space, where the ongrid and offgrid components of the delay and Doppler shifts are separated for estimation. In particular, the ongrid components of the delay and Doppler shifts are jointly determined by the entry indices with significant values in the recovered sparse vector. Then, the corresponding offgrid components are modeled as hyperparameters in the proposed SBL framework, which can be estimated via the expectationmaximization method. To strike a balance between channel estimation performance and computational complexity, we further propose a twodimensional (2D) offgrid SSR problem via decoupling the delay and Doppler shift estimations. In our developed 1D and 2D offgrid SBLbased channel estimation algorithms, the hyperparameters are updated alternatively for computing the conditional posterior distribution of channels, which can be exploited to reconstruct the effective DD domain channel. Compared with the 1D method, the proposed 2D method enjoys a much lower computational complexity while only suffers slight performance degradation. Simulation results verify the superior performance of the proposed channel estimation schemes over stateoftheart schemes.
 [6] arXiv:2101.05681 [pdf, ps, other]

Title: Adaptive Private Distributed Matrix MultiplicationComments: arXiv admin note: text overlap with arXiv:2004.12925Subjects: Information Theory (cs.IT)
We consider the problem of designing codes with flexible rate (referred to as rateless codes), for private distributed matrixmatrix multiplication. A master server owns two private matrices $\mathbf{A}$ and $\mathbf{B}$ and hires worker nodes to help computing their multiplication. The matrices should remain informationtheoretically private from the workers. Codes with fixed rate require the master to assign tasks to the workers and then wait for a predetermined number of workers to finish their assigned tasks. The size of the tasks, hence the rate of the scheme, depends on the number of workers that the master waits for. We design a rateless private matrixmatrix multiplication scheme, called RPM3. In contrast to fixedrate schemes, our scheme fixes the size of the tasks and allows the master to send multiple tasks to the workers. The master keeps sending tasks and receiving results until it can decode the multiplication; rendering the scheme flexible and adaptive to heterogeneous environments. Despite resulting in a smaller rate than known stragglertolerant schemes, RPM3 provides a smaller mean waiting time of the master by leveraging the heterogeneity of the workers. The waiting time is studied under two different models for the workers' service time. We provide upper bounds for the mean waiting time under both models. In addition, we provide lower bounds on the mean waiting time under the workerdependent fixed service time model.
 [7] arXiv:2101.05792 [pdf, ps, other]

Title: Group Testing with a Graph Infection Spread ModelSubjects: Information Theory (cs.IT); Computers and Society (cs.CY); Data Structures and Algorithms (cs.DS); Networking and Internet Architecture (cs.NI); Signal Processing (eess.SP)
We propose a novel infection spread model based on a random connection graph which represents connections between $n$ individuals. Infection spreads via connections between individuals and this results in a probabilistic cluster formation structure as well as a noni.i.d. (correlated) infection status for individuals. We propose a class of twostep sampled group testing algorithms where we exploit the known probabilistic infection spread model. We investigate the metrics associated with twostep sampled group testing algorithms. To demonstrate our results, for analytically tractable exponentially split cluster formation trees, we calculate the required number of tests and the expected number of false classifications in terms of the system parameters, and identify the tradeoff between them. For such exponentially split cluster formation trees, for zeroerror construction, we prove that the required number of tests is $O(\log_2n)$. Thus, for such cluster formation trees, our algorithm outperforms any zeroerror nonadaptive group test, binary splitting algorithm, and Hwang's generalized binary splitting algorithm. Our results imply that, by exploiting probabilistic information on the connections of individuals, group testing can be used to reduce the number of required tests significantly even when infection rate is high, contrasting the prevalent belief that group testing is useful only when infection rate is low.
Crosslists for Fri, 15 Jan 21
 [8] arXiv:2009.01678 (crosslist from math.PR) [pdf, ps, other]

Title: HamiltonJacobi equations for inference of matrix tensor productsComments: 44 pagesSubjects: Probability (math.PR); Information Theory (cs.IT)
We study the highdimensional limit of the free energy associated with the inference problem of finiterank matrix tensor products. In general, we bound the limit from above by the unique solution to a certain HamiltonJacobi equation. Under additional assumptions on the nonlinearity in the equation which is determined explicitly by the model, we identify the limit with the solution. Two notions of solutions, weak solutions and viscosity solutions, are considered, each of which has its own advantages and requires different treatments. For concreteness, we apply our results to a model with i.i.d. entries and symmetric interactions. In particular, for the first order and even order tensor products, we identify the limit and obtain estimates on convergence rates; for other odd orders, upper bounds are obtained.
 [9] arXiv:2101.05544 (crosslist from cs.LG) [pdf, other]

Title: DICE: Diversity in Deep Ensembles via Conditional Redundancy Adversarial EstimationComments: Published as a conference paper at ICLR 2021. 9 main pages, 13 figures, 12 tablesSubjects: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Information Theory (cs.IT)
Deep ensembles perform better than a single network thanks to the diversity among their members. Recent approaches regularize predictions to increase diversity; however, they also drastically decrease individual members' performances. In this paper, we argue that learning strategies for deep ensembles need to tackle the tradeoff between ensemble diversity and individual accuracies. Motivated by arguments from information theory and leveraging recent advances in neural estimation of conditional mutual information, we introduce a novel training criterion called DICE: it increases diversity by reducing spurious correlations among features. The main idea is that features extracted from pairs of members should only share information useful for target class prediction without being conditionally redundant. Therefore, besides the classification loss with information bottleneck, we adversarially prevent features from being conditionally predictable from each other. We manage to reduce simultaneous errors while protecting class information. We obtain stateoftheart accuracy results on CIFAR10/100: for example, an ensemble of 5 networks trained with DICE matches an ensemble of 7 networks trained independently. We further analyze the consequences on calibration, uncertainty estimation, outofdistribution detection and online codistillation.
Replacements for Fri, 15 Jan 21
 [10] arXiv:1702.07409 (replaced) [pdf, other]

Title: Founsure 1.0: An Erasure Code Library with Efficient Repair and Update FeaturesAuthors: Şuayb Ş. ArslanComments: Accepted to Elsevier SoftwareX, 2021Subjects: Information Theory (cs.IT)
 [11] arXiv:1712.01497 (replaced) [pdf, other]

Title: Atomic Norm Based Localization of FarField and NearField Signals with Generalized Symmetric ArraysSubjects: Information Theory (cs.IT); Signal Processing (eess.SP)
 [12] arXiv:1908.02503 (replaced) [pdf, other]

Title: Computational approaches to nonconvex, sparsityinducing multipenalty regularizationComments: 20 pages, 2 figuresSubjects: Information Theory (cs.IT)
 [13] arXiv:1912.11662 (replaced) [pdf, other]

Title: Terahertz MultiUser Massive MIMO with Intelligent Reflecting Surface: Beam Training and Hybrid BeamformingSubjects: Information Theory (cs.IT); Signal Processing (eess.SP)
 [14] arXiv:2006.04581 (replaced) [pdf, other]

Title: Precoder Design and Power Allocation for Downlink MIMONOMA via Simultaneous TriangularizationComments: Accepted for presentation at the 2021 IEEE Wireless Communications and Networking Conference. This paper is the conference version of arXiv:2006.06471 with 6pp, 2 figures, for code, see this https URLSubjects: Information Theory (cs.IT)
 [15] arXiv:2006.06471 (replaced) [pdf, other]

Title: Uplink and Downlink MIMONOMA with Simultaneous TriangularizationComments: Accepted by the IEEE Transactions on Wireless Communications. This is the journal version of the submission arXiv:2006.04581 with 33 pages, 10 figures, and 2 tables. For associated code see this https URLJournalref: IEEE Transactions on Wireless Communications, 2021Subjects: Information Theory (cs.IT)
 [16] arXiv:2011.14741 (replaced) [pdf, ps, other]

Title: Minimax Converse for Identification via ChannelsAuthors: Shun WatanabeComments: 18 pages, no figureSubjects: Information Theory (cs.IT)
 [17] arXiv:2101.03225 (replaced) [pdf, ps, other]

Title: The extended binary quadratic residue code of length 42 holds a 3designComments: 6 pages. Second versionSubjects: Information Theory (cs.IT); Combinatorics (math.CO)
 [18] arXiv:2007.08158 (replaced) [pdf, other]

Title: Channel Estimation for RISAided mmWave MIMO Systems via Atomic Norm MinimizationComments: 30 pages, 10 figures, Submitted to IEEE TWC, under second round reviewSubjects: Signal Processing (eess.SP); Information Theory (cs.IT)
[ showing up to 2000 entries per page: fewer  more ]
Disable MathJax (What is MathJax?)
Links to: arXiv, form interface, find, cs, recent, 2101, contact, help (Access key information)