Time filter

Source Type

Redwood City, CA, United States

Ardestanizadeh E.,University of California at San Diego | Ardestanizadeh E.,ASSIA Inc | Franceschetti M.,University of California at San Diego
IEEE Transactions on Automatic Control

Feedback communication is studied from a control-theoretic perspective, mapping the communication problem to a control problem in which the control signal is received through the same noisy channel as in the communication problem, and the (nonlinear and time-varying) dynamics of the system determine a subclass of encoders available at the transmitter. The MSE exponent is defined to be the exponential decay rate of the mean square decoding error and is used for analysis of the reliable rate of communication. A sufficient condition is provided under which the MMSE capacity, the supremum achievable MSE exponent, is equal to the information-theoretic capacity, the supremum achievable rate. For the special class of stationary Gaussian channels and linear time-invariant systems, a simple application of Bode's integral formula shows that the feedback capacity, recently characterized by Kim, is equal to the maximum instability that can be tolerated by any linear controller under a given power constraint. Finally, the control mapping is generalized to the N-sender AWGN multiple access channel. It is shown that Kramer's code for this channel, which is known to be sum rate optimal in the class of generalized linear feedback codes, can be obtained by solving a linear quadratic Gaussian control problem. © 2012 IEEE. Source

Ardestanizadeh E.,University of California at San Diego | Ardestanizadeh E.,ASSIA Inc | Minero P.,University of Notre Dame | Franceschetti M.,University of California at San Diego
IEEE Transactions on Information Theory

A code for communication over the k-receiver complex additive white Gaussian noise broadcast channel (BC) with feedback is presented and analyzed using tools from the theory of linear quadratic Gaussian optimal control. It is shown that the performance of this code depends on the noise correlation at the receivers and it is related to the solution of a discrete algebraic Riccati equation. For the case of independent noises, the sum rate achieved by the proposed code, satisfying average power constraint P , is characterized as 1/2 log (1+Pφ), where the coefficient φ [1,k] quantifies the power gain due to the presence of feedback. This includes a previous result by Elia and strictly improves upon the codes by Ozarow and Leung and by Kramer. When the noises are correlated, the prelog of the sum capacity of the BC with feedback can be strictly greater than 1. It is established that for all noise covariance matrices of rank r the prelog of the sum capacity is at most k-r+1 and, conversely, there exists a noise covariance matrix of rank r for which the proposed code achieves this upper bound. This generalizes a previous result by Gastpar et al. for the two-receiver BC. © 1963-2012 IEEE. Source

Bhagavatula R.,ASSIA Inc | Heath Jr. R.W.,University of Texas at Austin
IEEE Transactions on Signal Processing

Base station cooperation can exploit knowledge of the users' channel state information (CSI) at the transmitters to manage co-channel interference. Users have to feedback CSI of the desired and interfering channels using finite-bandwidth backhaul links. Existing codebook designs for single-cell limited feedback can be used for multicell cooperation by partitioning the available feedback resources between the multiple channels. In this paper, a new feedback-bit allocation strategy is proposed, as a function of the delays in the communication links and received signal strengths in the downlink. Channel temporal correlation is modeled as a function of delay using the Gauss-Markov model. Closed-form expressions for bit partitions are derived to allocate more bits to quantize the stronger channels with smaller delays and fewer bits to weaker channels with larger delays, assuming random vector quantization. Cellular network simulations are used to show that the algorithm presented in the paper yields higher sum-rates than the equal-bit allocation technique. © 2011 IEEE. Source

Bhagavatula R.,ASSIA Inc | Heath R.W.,University of Texas at Austin
IEEE Transactions on Wireless Communications

Base station cooperation can use knowledge of the users' channel state information (CSI) at the transmitters to manage co-channel interference. A reasonable way to provide CSI to the base stations is through a finite rate limited feedback channel. Existing multicell limited feedback techniques require a large amount of feedback, which incurs an overhead penalty on the uplink. In this paper, a new feedback approach based on predictive vector quantization (PVQ) is proposed to reduce feedback requirements in multicell systems and provide high resolution CSI at base stations by exploiting temporal correlation in the channels. Transmitter and receiver structures are proposed to implement predictive limited feedback accounting for delay, for signals on the Grassmann manifold. Simulations show that the proposed PVQ framework yields higher sum-rates than memoryless quantization approaches for multicell limited feedback, in a cooperative system using intercell interference nulling. © 2002-2012 IEEE. Source

Galli S.,ASSIA Inc
IEEE Transactions on Communications

We report here that channel power gain and Root-Mean-Square Delay Spread (RMS-DS) in Low/Medium Voltage power line channels are negatively correlated lognormal random variables. Further analysis of other wireline channels allows us to report a strong similarity between some properties observed in power line channels and the ones observed in other wireline channels, e.g. coaxial cables and phone lines. For example, it is here reported that channel power gain and logarithm of the RMS-DS in DSL links are linearly correlated random variables. Exploiting these results, we here propose a statistical wireline channel model where tap amplitudes and delays are generated in order to reflect these physical properties. Although wireline channels are considered deterministic as their impulse response can be readily calculated once the link topology is known, a statistical wireline channel model is useful because the variability of link topologies and wiring practices give rise to a stochastic aspect of wireline communications that has not been well characterized in the literature. Finally, we also point out that alternative channel models that normalize impulse responses to a common (often unitary) power gain may be misleading when assessing the performance of equalization schemes since this normalization artificially removes the correlation between channel power gain and RMS-DS and, thus, Inter-Symbol Interference (ISI). © 2011 IEEE. Source

Discover hidden collaborations