Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/ant-uni-bremen/mmnet_for_cmdnet

MMNet evaluated for the CMDNet journal article

https://github.com/ant-uni-bremen/mmnet_for_cmdnet

Science Score: 23.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ○
    codemeta.json file
  • ○
    .zenodo.json file
  • ✓
    DOI references
    Found 2 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (7.4%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

MMNet evaluated for the CMDNet journal article

Basic Info
  • Host: GitHub
  • Owner: ant-uni-bremen
  • Language: Python
  • Default Branch: master
  • Size: 64.5 KB
Statistics
  • Stars: 0
  • Watchers: 0
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Fork of mehrdadkhani/MMNet
Created over 1 year ago · Last pushed over 1 year ago

https://github.com/ant-uni-bremen/MMNet_for_CMDNet/blob/master/ 

# MMNet

This MMNet fork was used to evaluate MMNet for the [CMDNet journal article](https://doi.org/10.1109/TCOMM.2021.3114682).

MMNet is a massive MIMO signal detection scheme based on light online learning with neural networks that extends to correlated channel scenarios. 

## Table of Contents
0. [Introduction](#introduction)
0. [Citation](#citation)
0. [Repository structure](#repository-structure)

## Introduction
This repository contains MMNet signal detection model, the channels dataset, and benchmarking detection schemes discusssed in the paper "Adaptive Neural Signal Detection for Massive MIMO" (https://arxiv.org/abs/1906.04610). On i.i.d. Gaussian channels, MMNet requires two orders of magnitude fewer operations than existing deep learning schemes but achieves near-optimal performance. On spatially-correlated channels, it achieves the same error rate as the next-best learning scheme (OAMPNet) at 2.5dB lower SNR and with at least 10x less computational complexity. MMNet is also 4--8dB better overall than a classic linear scheme like the minimum mean square error (MMSE) detector.

## Citation
You may cite this project as:
```
@article{khani2019adaptive,
  title={Adaptive Neural Signal Detection for Massive MIMO},
  author={Khani, Mehrdad and Alizadeh, Mohammad and Hoydis, Jakob and Fleming, Phil},
  journal={arXiv preprint arXiv:1906.04610},
  year={2019}
}
```

## Repository structure
Find MMNet and other learning based schemes in ``./learning_based`` directory. Minimum mean square error (MMSE), Approximated message passaing (AMP), Semidefinite relaxation (SDR), Multistage interference cancelation (BLAST), and Maximum-likelihood optimal (ML) are located under ``./classic``. In order to reproduce the simulated correlated channels using 3D-3GPP model please refer to ``./channels`` directory.

Owner

  • Name: Arbeitsbereich Nachrichtentechnik
  • Login: ant-uni-bremen
  • Kind: organization
  • Location: Bremen, Germany

Department of Communication Engineering at University of Bremen

GitHub Events

Total
  • Member event: 1
  • Push event: 1
Last Year
  • Member event: 1
  • Push event: 1