Federated Learning for 5G Radio Spectrum Sensing

Abstract

Spectrum sensing (SS) is an important tool in finding new opportunities for spectrum sharing. The users, called Secondary Users (SU), who do not have a license to transmit without hindrance, need to employ SS in order to detect and use the spectrum without interfering with the licensed users’ (primary users’ (PUs’)) transmission. Deep learning (DL) has proven to be a good choice as an intelligent SS algorithm that considers radio environmental factors in the decision-making process. It is impossible though for SU to collect the required data and train complex DL models. In this paper, we propose to employ a Federated Learning (FL) algorithm in order to distribute data collection and model training processes over many devices. The proposed method categorizes FL devices into groups by their mean Signal-to-Noise ratio (SNR) and creates a common DL model for each group in the iterative process. The results show that detection accuracy obtained via the FL algorithm is similar to detection accuracy obtained by employing several DL models, namely convolutional neural networks (CNNs), specialized in spectrum detection for a PU signal with a given mean SNR value. At the same time, the main goal of simplification of the SS process in the network is achieved.

Enhanced Federated Learning for Edge Data Security in Intelligent Transportation Systems

“…To tackle this issue, Sabuj et al [153] propose a UAV cognitive radio network (CRN) that utilizes local FL in the edge network to enhance spectral efficiency. Similarly, Wasilewska et al [154] aim to maximize spectral efficiency by exploring the relationship between computational and communication resources in FLbased CRNs. Additionally, to enhance the energy efficiency of FL, Shen et al [155] investigate the determination of the local convergence threshold and optimization of resource allocation to minimize the system's energy consumption in UAV swarms.…”

Secure federated machine learning for distributed spectrum sensing in communication networks

“…Spectrum sensing operations to establish CRNs pose praxeological considerations that make this use-case suited to the implementation of decentralized, horizontal, FML. 8,9 Spectrum sensing operations are distributed within a wireless regional area network (WRAN), with nodes comprising consumer premises devices (CPEs) that are managed by a designated base station (BS). Typically, a WRAN can be conceived to support orthodox FML, under a "hub and spoke" architecture, where each of the CPEs under perform local spectrum sensing and transmit their detected spectrum data to the BS.…”

Geospatial Cognitive Networking Protocols and Sensing Algorithms for 5G NR Beamforming

“…Without these configurations, energy detectors cannot confirm if 5G resources start at fixed (Type A) or flexible (Type B) times or where white space can start or end in the latter case [21]. Recent work on energy detectors transitioning from 4G to 5G ignores this issue entirely [22]. 4G blind control channel detectors like [23], and designs that leverage them [17], also struggle to adapt to 5G without such higher layer information.…”