Enhancing Massive MIMO Symbol Detection in Unknown Noise Environments: A Generative Modeling Approach
Bibliographic record
Abstract
This paper presents a novel symbol detection method for massive Multiple-Input Multiple-Output (m-MIMO) systems, addressing the challenges posed by unknown additive noise distributions. While the optimal MIMO detector under uniform priors is the Maximum Likelihood (ML) detector, its implementation depends on accurate knowledge of the noise distribution which is often inaccessible. Furthermore, for some types of additive noise, such as impulsive noise, the probability density function (PDF) does not admit a closed-form expression, making ML detection infeasible. In our approach, we exploit the favorable propagation properties of m-MIMO systems to obtain a reliable initial estimate of the transmitted symbol vector using a simple zero-forcing (ZF) detector. We then generate a limited number of random points from the input symbol constellation in a restricted neighborhood around the ZF estimate. These points are subsequently used to obtain samples from the unknown noise distribution, which are mapped to a latent space typically (but not necessarily) characterized by a Gaussian distribution, where approximate ML detection can be performed. We benchmark our proposed detector, called Zero-Forcing based Latent Space Symbol Detector (ZF-LSSD) against existing methods across various m-MIMO configurations and noise scenarios. Numerical simulations show that our detector consistently outperforms these methods in diverse additive noise environments.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".