Large-Scale Spectrum Occupancy Learning via Tensor Decomposition and LSTM Networks

TL;DR

This paper introduces a novel framework combining tensor decomposition and LSTM networks for large-scale spectrum occupancy prediction, effectively capturing correlations and handling noise and missing data.

Contribution

It presents a new tensor-based representation and a combined CP tensor decomposition with LSTM approach for efficient, accurate spectrum occupancy learning.

Findings

Outperforms existing methods in prediction accuracy

Demonstrates robustness against noise and missing data

Achieves computational efficiency on large-scale datasets

Abstract

A new paradigm for large-scale spectrum occupancy learning based on long short-term memory (LSTM) recurrent neural networks is proposed. Studies have shown that spectrum usage is a highly correlated time series. Moreover, there is a correlation for occupancy of spectrum between different frequency channels. Therefore, revealing all these correlations using learning and prediction of one-dimensional time series is not a trivial task. In this paper, we introduce a new framework for representing the spectrum measurements in a tensor format. Next, a time-series prediction method based on CANDECOMP/PARFAC (CP) tensor decomposition and LSTM recurrent neural networks is proposed. The proposed method is computationally efficient and is able to capture different types of correlation within the measured spectrum. Moreover, it is robust against noise and missing entries of sensed spectrum. The superiority of the proposed method is evaluated over a large-scale synthetic dataset in terms of prediction accuracy and computational efficiency.