AI-Enabled Data-driven Intelligence for Spectrum Demand Estimation

Colin Brown; Mohamad Alkadamani; Halim Yanikomeroglu

eess.SYMar 10, 2026

AI-Enabled Data-driven Intelligence for Spectrum Demand Estimation

Authors: Colin Brown, Mohamad Alkadamani, Halim Yanikomeroglu

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

Abstract

Accurately forecasting spectrum demand is a key component for efficient spectrum resource allocation and management. With the rapid growth in demand for wireless services, mobile network operators and regulators face increasing challenges in ensuring adequate spectrum availability. This paper presents a data-driven approach leveraging artificial intelligence (AI) and machine learning (ML) to estimate and manage spectrum demand. The approach uses multiple proxies of spectrum demand, drawing from site license data and derived from crowdsourced data. These proxies are validated against real-world mobile network traffic data to ensure reliability, achieving an R $^2$ value of 0.89 for an enhanced proxy. The proposed ML models are tested and validated across five major Canadian cities, demonstrating their generalizability and robustness. These contributions assist spectrum regulators in dynamic spectrum planning, enabling better resource allocation and policy adjustments to meet future network demands.

Figures & tables

Fig. 1: Integration of AI-enabled spectrum demand modeling within the broader spectrum planning and resource allocation framework.

Fig. 2: Counts for the number of active user for a single MNO in Toronto taken from crowdsourced data and aggregated at grid tile level.

Fig. 3: Geospatial distribution of proxy discrepancies in Montreal.

Proxy	$R^{2}$	$F$ -Statistic	p-Value
Deployed Bandwidth ( $P_{\text{BW}}$ )	0.72	4477	$<0.001$
Active Users ( $P_{\text{Users}}$ )	0.64	3998	$<0.001$
Combined ( $P_{\text{Combined}}$ )	0.85	$1.33\times 10^{4}$	$<0.001$

TABLE I: Validation Metrics for Proxy Analysis

Model	Proxy	$R^{2}$	Norm. RMSE	Norm. MAE
Baseline (Linear Reg.)	( $P_{\text{Combined}}$ )	0.54	0.031	0.017
XGBoost	( $P_{\text{BW}}$ )	0.84	0.026	0.016
XGBoost	( $P_{\text{Users}}$ )	0.68	0.027	0.018
XGBoost	( $P_{\text{Combined}}$ )	0.89	0.022	0.014

TABLE II: Prediction Performance Metrics for Different Proxies

Fig. 4: Scatter plot of actual vs. predicted spectrum demand values for the Combined Proxy using the best-performing XGBoost model.

Fig. 5: Feature importance heatmap for XGBoost models trained with different proxies.

Explore similar work

Towards Intelligent Spectrum Management: Spectrum Demand Estimation Using Graph Neural Networks
Mar 11, 2026Mohamad Alkadamani, Amir Ghasemi, Halim YanikomerogluWireless Communications Wireless Resource Allocation
Communications Research Centre Canada, Ottawa, ON, Canada · Carleton University, Ottawa, ON, Canada
A Graph-Based Approach to Spectrum Demand Prediction Using Hierarchical Attention Networks
Mar 10, 2026Mohamad Alkadamani, Halim Yanikomeroglu, Amir GhasemiWireless Communications Graph Attention Networks
Department of Data Science, Communications Research Centre Canada · Department of Systems and Computer Engineering, Carleton University
Towards Flexible Spectrum Access: Data-Driven Insights into Spectrum Demand
Mar 10, 2026Mohamad Alkadamani, Amir Ghasemi, Halim Yanikomeroglu6G Networks Spatiotemporal Forecasting
Communications Research Centre, Ottawa, Ontario, Canada · Carleton University, Ottawa, Ontario, Canada

Mar 11, 2026J/K move · Enter open · S save

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

Mohamad Alkadamani, Amir Ghasemi, Halim Yanikomeroglu

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

Wireless Communications Wireless Resource Allocation Graph Neural Networks Graph Attention Networks

Open paper AskPDF arXiv

The growing demand for wireless connectivity, combined with limited spectrum resources, calls for more efficient spectrum management. Spectrum sharing is a promising approach; however, regulators need accurate methods to characterize demand dynamics and guide allocation decisions. This paper builds and validates a spectrum demand proxy from public deployment records and uses a graph attention network in a hierarchical, multi-resolution setup (HR-GAT) to estimate spectrum demand at fine spatial scales. The model captures both neighborhood effects and cross-scale patterns, reducing spatial autocorrelation and improving generalization. Evaluated across five Canadian cities and against eight competitive baselines, HR-GAT reduces median RMSE by roughly 21% relative to the best alternative and lowers residual spatial bias. The resulting demand maps are regulator-accessible and support spectrum sharing and spectrum allocation in wireless networks.

Abstract

Explore similar work

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

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

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