A6s~uct-Thlspm~lavoidsthe complexmatched-filter bank required for optimal detection. Instead, two simple l o w p ? ~ filters are employed, followed by a sampler and a Viterbi algorithm that accounts for ISI. For representative modulations it shown that near-optimal performance can be achieved with just two samples per baud. ~ ( t ) = s ( t ) + n(t). The proposed receiver structure shown in Fig....

We present a generalisation of the standard well known and widely used Metropolis-Hastings algorithm which, unlike the Metropolis-Hastings sampler, can mix well over a target distribution which has separated and possibly narrow modes. The idea arose via a consideration of a space augmented slice sampler where the uniform space required to be sampled for the slice sampler, say I , is augmented w...

We introduce Mean-for-Mode estimation, a variant of an uncollapsed Gibbs sampler that we use to train LDA on a GPU. The algorithm combines benefits of both uncollapsed and collapsed Gibbs samplers. Like a collapsed Gibbs sampler — and unlike an uncollapsed Gibbs sampler — it has good statistical performance, and can use sampling complexity reduction techniques such as sparsity. Meanwhile, like ...

This article provides the results of a study evaluating the sampling efficiency of a flat paddle sampling device against a round surface sampling device in controlled studies. Each device is supplied with identical microbial-growth media, and laboratory studies document equivalent performance of the media in the devices. Both devices perform sampling by the same process (agar transfer of microo...