Software innovation could shift cost, time for in-building wireless deployments

By Louis Jacob, CTO and SVP of Technology, iBWAVE and Scott Lewis, Vice President of Next Generation Solutions & Real Estate, Phoenix Tower

Note: This blog was produced under WIA’s Innovation and Technology Council (ITC). The ITC is the forum for forecasting the future of the wireless industry. Participants explore developments in the wider wireless industry, from 5G network monetization trends and streamlining infrastructure deployment to future spectrum needs and cell site power issues. The group is publishing a series of thought-leadership pieces throughout 2024.

For more than three decades, the wireless industry has done a good job of connecting people almost everywhere they live, work and play, including within large public venues, sports arenas and airports. In the past, mobile network operators (MNOs) have participated in a model in which they pay to install in-building networks in the biggest venues and most important airports to ensure their customers have seamless connectivity everywhere. 

But that dynamic is shifting, particularly as in-building deployments move into the next tier of buildings, including enterprises, hospitals, hotels, warehouses and multifamily dwellings. Building owners and landlords know their tenants expect cellular connectivity, and MNOs are motivated to make sure their customers can connect to their public networks or private wireless networks wherever they are. 

In some cases, MNOs continue to be willing to shoulder or share in the cost of deploying in-building infrastructure, particularly when it allows them to leverage their brand with naming-rights arrangements. But for the vast majority of venues that need in-building wireless coverage, MNOs now expect property owners to participate in building distributed antenna systems and other in-building wireless infrastructure – and pay for it.

In-building network design is still largely done the way it has been done for years. Software predicts how RF signals in a specific spectrum band for a particular MNO will behave in the physical indoor environment. Engineers then use various design tools  to plot that information and analyze it. 

Meanwhile, in-building networks are becoming more complex for a variety of reasons. There are more areas to cover, including parking garages, campuses and even the rugged environs of mines. More equipment is needed to handle increasing voice and especially data usage. In-building networks are now being used not only to connect people but also machines, including thermostats, lighting and security cameras. The post-COVID hybrid working environment has necessitated interconnectivity between offices and remote work environments, and the skyrocketing use of video conferencing applications is taxing existing networks. New sustainable building materials change how wireless signals enter and move throughout spaces, and higher frequency signals behave differently than traditional signals. 

All of these factors add time and cost to designing and deploying in-building wireless networks with the same precision we have accomplished in the past. What took a day or an hour to simulate a few years ago can now take a week or more. 

In-building networks continue to require the same physical equipment and deployment process as they have for years: fiber must still be pulled, antennas still need to be placed, and indoor networks must still connect to a carrier network. Therefore, the key to streamlining deployments while maintaining a high level of accuracy is likely to require advances in software and innovations such as machine learning and artificial intelligence.

There has not been a fundamental technical innovation in in-building wireless in almost 30 years since the invention of the distributed antenna system that has enabled mobile network operators to be inside of buildings ubiquitously in a way that serves their customers at a relatively reasonable cost. But while we have innovated so much in software in the overall telecoms industry, and we have innovated so much in the way mobile devices work, there is still a big roadblock in bringing mobility inside 99 percent of the world’s buildings. 

Although Wi-Fi networks are widely available, end-users increasingly prefer to connect to their mobile carrier’s network or a private wireless network for security reasons. Many times, these connections are not available. In the post-COVID hybrid work environment, people are often working from a variety of different places, including hotels when they are on vacation. CTOs at hotels and other locations are feeling the pressure from guests who need secure wireless connectivity. Could the mobile industry replicate the successful software-enabled Wi-Fi topology in use today for private wireless networks?

Infrastructure companies, like Phoenix Tower International, can already build neutral-host networks in buildings and on campuses. These networks leverage one design and one set of equipment to allow connectivity with multiple MNOs on different spectrum bands. Software allows changes and updates to be made to these networks without having to replace antennas and other equipment. This is a significant savings of both time and money and protects the initial investment.

One problem we as an industry are still trying to solve is creating software that allows shared spectrum to be deployed on off-the-shelf servers with a radio head in a building. If the major MNOs agree on a standard for how that will work, in-building networks can offload mobile traffic using Multiple Operator Core Network (MOCN), a technology that could bridge the gap between traditional DAS networks and new private wireless networks. This scenario will require mobile operators and regulators to agree on a joint standard that allows radio spectrum to be shared, similar to CBRS.

AI can streamline tasks from simple to complex. For example, AI tools could make recommendations for how much equipment a specific design will need by ingesting the details of an in-building design as well as the specifications of various equipment and then creating an estimate of how much wiring, cabling and other equipment will be needed to actually build out the design.

Another avenue companies including iBwave, the global leader in network planning, are exploring to leverage AI/ML is the idea of using software to create a first draft of an in-building design with acceptable precision that engineers can then tweak and optimize. Imagine the value of trading time plotting data on paper from scratch to reviewing and fine-tuning a design that is already 95 percent complete.

iBwave is already leveraging automatic features such as Automatic Small Cell/Access Point Placement and Auto Channel Assignment. These features enable users to set coverage criteria, with the software calculating and placing access points and small cells on the floor plan. This saves users a significant amount of time on their designs while reaching the desired coverage that can be further optimized.

It may be possible at some point for AI/ML to learn from existing real-world in-building deployments to build models for different buildings that share characteristics. Maybe the building is in the same location as another building but has fewer floors. Could one office or warehouse network design be adapted to another office or warehouse? It has been challenging in the past to leverage data because of how fast wireless technology, spectrum and devices have changed. But with AI/ML, tools may be able to use the science behind the data, such as how signals interact with concrete, rather than the data itself to build accurate models quickly. 

In the next few years, it is possible that the major mobile operators in this country will be able to work with technology innovators to find a way to deploy hundreds if not thousands of software-enabled solutions inside buildings that don’t require an engineer to sift through 200-page design documents for approval.