Enabling Mobile Base Stations in 5G via Wireless Access Backhaul (WAB): A Multi-Band Experimental Study

TL;DR

This paper demonstrates a multi-band Wireless Access Backhaul (WAB) testbed for 5G, validating its effectiveness in mobile scenarios and extending coverage in challenging conditions using open-source tools.

Contribution

First experimental realization of a multi-band WAB testbed combining FR2 and FR1, showcasing mobility support and coverage extension with commercial components.

Findings

WAB effectively mitigates FR2 limitations in uplink and NLOS scenarios.

The testbed extends FR2 coverage while maintaining legacy FR1 compatibility.

Experimental results confirm WAB's scalability and practicality for vehicular networks.

Abstract

Highly dynamic and mobile applications, such as vehicular networks, require stable connectivity, which is often challenging to achieve. Network densification is a key approach to address this issue and can be achieved cost-effectively through mobile base stations and wireless relaying. However, existing solutions rely on rigid and complex architectures that hinder deployment in dynamic scenarios. The recently standardized Wireless Access Backhaul (WAB) architecture represents a key evolution, enabling flexible and modular wireless relay networks with native support for mobility and multi-technology wireless backhaul. This paper presents the first experimental realization of a multi-band WAB testbed, combining an FR2 backhaul and an FR1 access link using open-source software and commercial off-the-shelf components. The proposed framework validates end-to-end WAB operation under mobility and demonstrates the extension of FR2 coverage while maintaining compatibility with legacy FR1 user equipment. Experimental campaigns in vehicular and outdoor-to-indoor scenarios confirm that WAB effectively mitigates FR2 limitations, particularly in uplink and Non-Line-of-Sight conditions. These results highlight WAB as a practical and scalable approach for vehicular and next-generation wireless networks.