Name
Abstract | ||
|---|---|---|
In a Bayesian linear model, suppose observation ${\\bf y}={\\bf H}{\\bf x}+{\\bf n}$ stems from independent inputs ${\\bf x}$ and ${\\bf n}$ which are Gaussian mixture (GM) distributed. With known matrix ${\\bf H}$, the minimum mean square error (MMSE) estimator for ${\\bf x}$ , has analytical form. However, its performance measure, the MMSE itself, has no such closed form. Because existing Bayesian MMSE bounds prove to have limited practical value under these settings, we instead seek analytical bounds for the MMSE, both upper and lower. This paper provides such bounds, and relates them to the signal-to-noise-ratio (SNR). |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1109/TSP.2012.2192112 | IEEE Transactions on Signal Processing |
Keywords | Field | DocType |
Bayes methods,Gaussian distribution,least mean squares methods,matrix algebra,Bayesian MMSE bounds,Bayesian linear model,GM distribution,Gaussian mixture distribution,Gaussian mixture statistics,MMSE estimation,SNR,linear model,matrix,minimum mean square error estimator,signal-to-noise-ratio, Gaussian mixture distribution, minimum mean square error estimation,linear model | Matrix (mathematics),Linear model,Signal-to-noise ratio,Minimum mean square error,Gaussian,Estimation theory,Statistics,Mathematics,Estimator,Bayesian probability | Journal |
Volume | Issue | ISSN |
60 | 7 | 1053-587X |
Citations | PageRank | References |
17 | 0.76 | 18 |
Authors | ||
4 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| John T. Flåm | 1 | 35 | 4.68 |
| Saikat Chatterjee | 2 | 17 | 0.76 |
| Kimmo Kansanen | 3 | 195 | 24.36 |
| Torbjörn Ekman | 4 | 33 | 2.21 |