Multiple Importance Sampling for Efficient Symbol Error Rate Estimation

Elvira, Vı́ctor; Santamarı́a, Ignacio

doi:10.1109/lsp.2019.2892835

Cited by 19 publications

(10 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Adaptive multiple importance sampling has been utilised in a variety of research fields, including population genetics (16), environment illumination computations (17), and signal communications (18).…”

Section: Background and Methodsmentioning

confidence: 99%

Integrating geostatistical maps and transmission models using adaptive multiple importance sampling

Retkute

Touloupou

Basáñez

et al. 2020

Preprint

View full text Add to dashboard Cite

The Adaptive Multiple Importance Sampling algorithm (AMIS) is an iterative technique which recycles samples from all previous iterations in order to improve the efficiency of the proposal distribution. We have formulated a new statistical framework based on AMIS to sample parameters of transmission models based on high-resolution geospatial maps of disease prevalence, incidence, or relative risk. We tested the performance of our algorithm on four case studies: ascariasis in Ethiopia, onchocerciasis in Togo, HIV in Botswana, and malaria in the Democratic Republic of the Congo.

show abstract

Section: Background and Methodsmentioning

confidence: 99%

Integrating geostatistical maps and transmission models using adaptive multiple importance sampling

Retkute

Touloupou

Basáñez

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…A deeper theoretical analysis of ALOE can be found in [28]. ALOE has recently been used for SER estimation of singleinput single-output (SISO) channels with non-square twodimensional constellations in [29]. With two-dimensional lattices or constellations the Voronoi regions are determined by just a few hyperplanes, which can easily be computed, and hence an exact implementation of ALOE is possible.…”

Section: Multiple Importance Sampling and The Aloe Algorithmmentioning

confidence: 99%

“…A multiple importance sampling technique called ALOE ("At Least One rare Event") [28], has recently been applied to SER estimation in single-input single-ouput systems [29] transmitting non-square 2D constellations [30]. ALOE is extremely efficient to estimate the integral of a Gaussian in a region defined by a union of half-spaces, which is precisely the error event in a digital communications system.…”

Section: Introductionmentioning

confidence: 99%

Multiple Importance Sampling for Symbol Error Rate Estimation of Maximum-Likelihood Detectors in MIMO Channels

Elvira

Santamarı́a

2021

IEEE Trans. Signal Process.

Self Cite

View full text Add to dashboard Cite

In this paper we propose a multiple importance sampling (MIS) method for the efficient symbol error rate (SER) estimation of maximum likelihood (ML) multiple-input multipleoutput (MIMO) detectors. Given a transmitted symbol from the input lattice, obtaining the SER requires the computation of an integral outside its Voronoi region in a high-dimensional space, for which a closed-form solution does not exist. Hence, the SER must be approximated through crude or naive Monte Carlo (MC) simulations. This practice is widely used in the literature despite its inefficiency, particularly severe at high signal-to-noiseratio (SNR) or for systems with stringent SER requirements. It is well-known that more sophisticated MC-based techniques such as MIS, when carefully designed, can reduce the variance of the estimators in several orders of magnitude with respect to naive Monte Carlo in rare-event estimation, or equivalently, they need significantly less samples for attaining a desired performance. The proposed MIS method provides unbiased SER estimates by sampling from a mixture of components that are carefully chosen and parametrized. The number of components, the parameters of the components, and their weights in the mixture, are automatically chosen by the proposed method. As a result, the proposed method is flexible, easy-to-use, theoretically sound, and presents a high performance in a variety of scenarios. We show in our simulations that SERs lower than 10 −8 can be accurately estimated with just 10 4 random samples.

show abstract

“…The procedure for the efficient simulation from a generic truncated Gaussian distribution is described in [11]. ALOE has recently been used for the SER estimation of single-input single-output (SISO) AWGN channels with non-square two-dimensional constellations in [19]. With twodimensional lattices or constellations the Voronoi regions are determined by just a few hyperplanes, which can easily be computed, and hence an exact implementation of ALOE is possible.…”

Section: B Multiple Importance Samplingmentioning

confidence: 99%

Efficient SER Estimation for MIMO Detectors via Importance Sampling Schemes

Elvira

Santamarı́a

2019

2019 53rd Asilomar Conference on Signals, Systems, and Computers

Self Cite

View full text Add to dashboard Cite

In this paper we propose two importance sampling methods for the efficient symbol error rate (SER) estimation of maximum likelihood (ML) multiple-input multiple-output (MIMO) detectors. Conditioned to a given transmitted symbol, computing the SER requires the evaluation of an integral outside a given polytope in a high-dimensional space, for which a closedform solution does not exist. Therefore, Monte Carlo (MC) simulation is typically used to estimate the SER, although a naive or raw MC implementation can be very inefficient at high signal-to-noise-ratios or for systems with stringent SER requirements. A reduced variance estimator is provided by the Truncated Hypersphere Importance Sampling (THIS) method, which samples from a proposal density that excludes the largest hypersphere circumscribed within the Voronoi region of the transmitted vector. A much more efficient estimator is provided by the existing ALOE (which stands for "At Least One rare Event") method, which samples conditionally on an error taking place. The paper describes in detail these two IS methods, discussing their advantages and limitations, and comparing their performances.

show abstract

Multiple Importance Sampling for Efficient Symbol Error Rate Estimation

Cited by 19 publications

References 33 publications

Integrating geostatistical maps and transmission models using adaptive multiple importance sampling

Integrating geostatistical maps and transmission models using adaptive multiple importance sampling

Multiple Importance Sampling for Symbol Error Rate Estimation of Maximum-Likelihood Detectors in MIMO Channels

Efficient SER Estimation for MIMO Detectors via Importance Sampling Schemes

Contact Info

Product

Resources

About