arXiv subjects
Publications and source records attributed to Hans-Andrea Loeliger.
The partition function of a factor graph can sometimes be accurately estimated by Monte Carlo methods. In this paper, such methods are extended to factor graphs with negative and complex factors.
The paper proposes Monte Carlo algorithms for the computation of the information rate of two-dimensional source/channel models. The focus of the paper is on binary-input channels with constraints on the allowed input configurations. The problem of numerically computing the information rate, and even the noiseless capacity, of such channels has so far remained largely unsolved. Both problems can be reduced to computing a Monte Carlo estimate of a partition function. The proposed algorithms use tree-based Gibbs sampling and multilayer (multitemperature) importance sampling. The viability of the proposed algorithms is demonstrated by simulation results.
A factor-graph representation of quantum-mechanical probabilities is proposed. Unlike standard statistical models, the proposed representation uses auxiliary variables (state variables) that are not random variables.
Polynomial remainder codes are a large class of codes derived from the Chinese remainder theorem that includes Reed-Solomon codes as a special case. In this paper, we revisit these codes and study them more carefully than in previous work. We explicitly allow the code symbols to be polynomials of different degrees, which leads to two different notions of weight and distance. Algebraic decoding is studied in detail. If the moduli are not irreducible, the notion of an error locator polynomial is replaced by an error factor polynomial. We then obtain a collection of gcd-based decoding algorithms, some of which are not quite standard even when specialized to Reed-Solomon codes.
It is shown how expectation maximization (EM) may be viewed as a message passing algorithm in factor graphs. In particular, a general EM message computation rule is identified. As a factor graph tool, EM may be used to break cycles in a factor graph, and tractable messages may in some cases be obtained where the sum-product messages are unwieldy. As an exemplary application, the paper considers linear Gaussian state space models. Unknown coefficients in such models give rise to multipliers in the corresponding factor graph. A main attraction of EM in such cases is that it results in purely Gaussian message passing algorithms. These Gaussian EM messages are tabulated for several (scalar, vector, matrix) multipliers that frequently appear in applications.
Elaborating on prior work by Minka, we formulate a general computation rule for lossy messages. An important special case (with many applications in communications) is the conversion of "soft-bit" messages to Gaussian messages. By this method, the performance of a Kalman equalizer is improved, both for uncoded and coded transmission.