Online Supervised Learning for Traffic Load Prediction in Framed-ALOHA Networks

Research output: Contribution to journalArticlepeer-review

21 Citations (Scopus)
131 Downloads (Pure)


Predicting the current backlog, or traffic load, in framed-ALOHA networks enables the optimization of resource allocation, e.g., of the frame size. However, this prediction is made difficult by the lack of information about the cardinality of collisions and by possibly complex packet generation statistics. Assuming no prior information about the traffic model, apart from a bound on its temporal memory, this letter develops an online learning-based adaptive traffic load prediction method that is based on recurrent neural networks (RNN) and specifically on the long short-Term memory (LSTM) architecture. In order to enable online training in the absence of feedback on the exact cardinality of collisions, the proposed strategy leverages a novel approximate labeling technique that is inspired by the method of moments (MOM) estimators. Numerical results show that the proposed online predictor considerably outperforms conventional methods and is able to adapt changing traffic statistics.

Original languageEnglish
Article number8779707
Pages (from-to)1778-1782
Number of pages5
Issue number10
Early online date29 Jul 2019
Publication statusPublished - Oct 2019


  • Method of moments
  • Supervised learning
  • Recurrent neural networks
  • Maximum likelihood estimation
  • Prediction methods
  • Memory architecture
  • Training
  • Traffic load prediction
  • framed-ALOHA
  • online supervised learning
  • recurrent neural network


Dive into the research topics of 'Online Supervised Learning for Traffic Load Prediction in Framed-ALOHA Networks'. Together they form a unique fingerprint.

Cite this