eess.SYMar 10, 2026

Towards Flexible Spectrum Access: Data-Driven Insights into Spectrum Demand

Authors: Mohamad Alkadamani, Amir Ghasemi, Halim Yanikomeroglu

Organizations: Communications Research Centre, Ottawa, Ontario, Canada · Carleton University, Ottawa, Ontario, Canada

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

In the diverse landscape of 6G networks, where wireless connectivity demands surge and spectrum resources remain limited, flexible spectrum access becomes paramount. The success of crafting such schemes hinges on our ability to accurately characterize spectrum demand patterns across space and time. This paper presents a data-driven methodology for estimating spectrum demand variations over space and identifying key drivers of these variations in the mobile broadband landscape. By leveraging geospatial analytics and machine learning, the methodology is applied to a case study in Canada to estimate spectrum demand dynamics in urban regions. Our proposed model captures 70% of the variability in spectrum demand when trained on one urban area and tested on another. These insights empower regulators to navigate the complexities of 6G networks and devise effective policies to meet future network demands.

Figures & tables

Fig. 1: Overview of the spectrum demand estimation methodology.
Fig. 2: Scatter plot comparing deployed bandwidth to download throughput.
Fig. 3: Heatmap of total deployed bandwidth in the GTA.
Data TypeDescription
Population Density (Statistics Canada)Count of individuals per given area during nighttime
Socio-economic Indicators (Statistics Canada)Data reflecting various socio-economic metrics such as income levels, education attainment, and age demographics
Industry Presence (Statistics Canada)Density of different industrial sectors, as defined by NACIS, within a specific geographic area
Daytime Population (Environics Analytics)Count of individuals per given area during daytime
Building Footprints (Microsoft)Geospatial data representing the outlines of building structures
Non-Residential Buildings (OpenStreetMap)Geospatial data representing the footprints of non-residential structures
TABLE I: Feature Engineering Data Summary
Fig. 4: Scatter plot comparing the number of transportation hubs and weighted deployed bandwidth.
ModelScenarioPerformance
Ridge RegressionCombined Urban RegionsR2=0.64R^{2}=0.64 , RMSE = 0.96
Ridge RegressionGTA - VancouverR2=0.53R^{2}=0.53 , RMSE = 1.9
GBRCombined Urban RegionsR2=0.81R^{2}=0.81 , RMSE = 0.51
GBRGTA - VancouverR2=0.70R^{2}=0.70 , RMSE = 0.93
TABLE II: Summary of ML Models’ Performance
Fig. 5: Feature importance for spectrum demand estimation.

Explore similar work

CardsList

  1. AI-Enabled Data-driven Intelligence for Spectrum Demand Estimation

    Mar 10, 2026Colin Brown, Mohamad Alkadamani, Halim YanikomerogluWireless CommunicationsWireless Resource Allocation

    Communications Research Centre, Ottawa, Ontario, Canada · Carleton University, Ottawa, Ontario, Canada

  2. Towards Intelligent Spectrum Management: Spectrum Demand Estimation Using Graph Neural Networks

    Mar 11, 2026Mohamad Alkadamani, Amir Ghasemi, Halim YanikomerogluWireless CommunicationsWireless Resource Allocation

    Communications Research Centre Canada, Ottawa, ON, Canada · Carleton University, Ottawa, ON, Canada

  3. A Graph-Based Approach to Spectrum Demand Prediction Using Hierarchical Attention Networks

    Mar 10, 2026Mohamad Alkadamani, Halim Yanikomeroglu, Amir GhasemiWireless CommunicationsGraph Attention Networks

    Department of Data Science, Communications Research Centre Canada · Department of Systems and Computer Engineering, Carleton University