Connectivity is experiencing a revolution. Low Earth orbit (LEO) broadband has moved with lightning speed from a futuristic concept to reality, and it’s rapidly becoming a cornerstone of modern connectivity.

SpaceX's Starlink, the leader and only commercially operational player in this field, besides OneWeb, already serves almost 5 million subscribers in 117 countries. The company has provided critical emergency infrastructure during recent hurricanes and wildfires.

This rise of LEO broadband shows no signs of slowing down: Starlink is further pushing the envelope on network capacity as it prepares to deploy its latest V3 satellites on the new, much larger Starship rocket. V3’s supposedly twentyfold increased capacity would even enable gigabit services once sufficiently deployed. Provided Starlink’s claims of “breakeven cash flow” stand investors’ scrutiny, the company will also have the funds to further expand its constellation and service offering. Competitors such as Amazon's Project Kuiper and China's burgeoning LEO programs are hot on Starlink’s heels, and in the coming years could provide robust alternatives.

But what does this satellite internet boom mean for the terrestrial broadband industry? There’s no doubt that satellite will bring both opportunities and challenges, forcing operators like CableLabs’ member companies to ask some hard questions.

A new addition to CableLabs’ strategy brief series on LEO discusses how to analyze the technology’s capacity and examines scenarios for Starlink’s ramp-up. It is available exclusively for CableLabs member operators here.

Below, we provide a high-level look at five considerations for terrestrial broadband operators to keep in mind and also pose five questions for them to consider as the satellite broadband industry moves forward.

Competitive Threat

Starlink’s growing capacity will drive further rising broadband subscriber numbers. So far, most Starlink customers have no terrestrial alternative, so the company’s offer effectively expands the broadband market to remote areas. But once Starship deploys V3 at scale, there would be enough capacity for Starlink to become a significant competitor for home broadband subscribers everywhere.

Question 1: Where is cable market share vulnerable to LEO providers, and how far?

We examine this further in the new members-only strategy brief.

Truly Seamless Converged Low-Latency Connectivity

Starlink already has nine global mobile network operator (MNO) partners beginning to complement its terrestrial coverage with satellite-based connectivity. In the mid-term, this broadened coverage will provide “direct-to-cell” service to any reasonably up-to-date phone anywhere, making dead zones a thing of the past and ubiquitous connectivity table stakes. At the same time, optical free space communication between satellites will enable a major leap in reducing latency.

Question 2: How can broadband operators with a converged network vision integrate LEO as a complementary layer to enhance their service offerings?

“Coopetitive” Ecosystem Building and Coordination

It seems obvious that (at most) a few global LEO connectivity providers will prevail, and any communications service provider (CSP) wanting to offer seamless connectivity will have to partner. Any LEO satellite will be serving on its orbit as a “cell tower in the sky” for many operators in quick succession, and LEO players will simultaneously be competing with CSP’s core service offering.

Question 3: What are the best ecosystem and competitive strategies with LEO players in the mid and long-term? How can we coordinate spectrum use and allocation across borders?

Transactional Connectivity and Capacity Trading

Highly automated transactional trading will become a key mechanism to ensure that the emerging multi-layered ecosystems provide users with the right connectivity option at any time, using, for example, automated roaming and dynamic spectrum allocation.

Question 4: How can networks and users negotiate connectivity and capacity transactions effectively?

Edge Computing

LEO satellite constellations will effectively function as a new edge location with high computing capacity that is only one millisecond-hop away from any point on the planet. Already now, compute capacity on satellites handles complex image processing tasks. Amazon Kuiper has announced to integrate their offer with their AWS footprint. CableLabs’ Next-Gen Systems team is exploring the possibilities of this “Cloud in the Sky.”

Question 5: What use cases will this new satellite edge enable? How can it interplay with the evolving terrestrial edge footprint?

How CableLabs Can Help

This exciting time demands proactive engagement to stay ahead of the curve and chart a successful course towards sustained differentiation.

CableLabs experts have insights, tools and research at the ready to help our members deepen their understanding of the LEO broadband landscape, assess market risks and competitive threats, and drive innovation. Whether you’re a strategist, an engineer, a product manager or just someone fascinated by the LEO possibilities, the message is clear: Pay attention, engage with the experts and be ready to adapt. LEO broadband isn’t just on the rise — it’s redefining the horizon.

For more in-depth analysis, check out the new members-only strategy brief about LEO and subscribe to the CableLabs blog to stay tuned for future updates.

Seamless connectivity is increasingly becoming an integral part of the connectivity service offerings of multiple system operators (MSOs). MSOs with their hybrid network infrastructure and varied service offerings may supplement their connectivity offerings with other wireless operators — for example, mobile virtual network operator (MVNO) agreements with mobile network operators (MNOs) — in addition to leveraging their own infrastructure. Depending on the infrastructure they own, MSOs may have to contend with disparate sets of wireless infrastructures.

In such scenarios, the transition of the user’s data traffic between these disparate networks becomes critical for ensuring a consistent user experience. It is also key to enforcing uniform and personalized policies as users move in and out of coverage of these networks with regards to:

• The transition being optimally triggered, considering not just the received signal strength indicator (RSSI) but also the network congestion, availability of and interference from neighboring networks and the traffic requirements (based on traffic/application type).
• The transition being as quick and seamless as possible (without any noticeable service disruption to the user) while maintaining security and privacy of user data.

Addressing these challenges will enable seamless and consistent connectivity for subscribers, dynamically adapting to evolving user needs and network conditions.

CableLabs’ Work In the Seamless Connectivity Space

CableLabs recognizes the evolving mobile industry landscape driven by the introduction of 5G and the availability of new and innovative spectrum options. Aware of our members’ growing mobile subscriber base and a need to complement their existing broadband offerings, we understand how critical it is to resolve the pain points that they face today (or may face in the near future).

Considering this, CableLabs, in collaboration with our members, started a Seamless Connectivity working group (WG) to understand and validate some of these issues. The aim of the group is to come up with multiple solutions that could cater to specific member needs by aligning with their infrastructure and deployment strategies. Seamless connectivity is a key theme of the Technology Vision for the future of the industry.

One potential solution for addressing the seamless connectivity issue is ATSSS, or Access Traffic Steering, Switching and Splitting. To learn more about the standardized ATSSS feature, watch our demo video below. You can watch the full video, “Harnessing ATSSS: Seamless Traffic Switching for Uninterrupted Connectivity,” here.

To learn more, check out the paper on seamless connectivity that we published at SCTE TechExpo24.

If you have any questions on the ATSSS demo or seamless connectivity initiative or if you would like to participate in the working group and collaborate with us, please reach out to me or my colleagues Sanjay Patel, John Bahr and Neeharika Jesukumar.

At the recent DOCSIS® Provisioning of XGS-PON Interop·Labs event at CableLabs, suppliers of PON optical line terminals (OLTs) exercised their systems’ ability to configure XGS-PON optical network units (ONUs) via the use of DOCSIS-style configuration files. The event took place February 24–28 at our headquarters in Louisville, Colorado.

CableLabs member operators have long been familiar with the method of using a configuration file to provision services on a DOCSIS cable modem. Beginning a decade or so ago, DOCSIS-style configuration was also utilized by many operators to simplify the provisioning of IEEE EPON systems compliant to the DOCSIS Provisioning of EPON (DPoE) specifications. These DPoE virtual cable modem (vCM) config files enabled an operator to use the same service provisioning back-office systems that they originally deployed for DOCSIS cable modems to provision services on DPoE EPON networks.

Fast-forward to today, and many CableLabs member operators are deploying or preparing to deploy ITU-T-based PON technologies, including XGS-PON, 25GS-PON and 50G HSP.

Some XGS-PON OLT suppliers have now also adopted the vCM config file method for provisioning ONU services by adding a DOCSIS adaptation layer (DAL) software function to their system offering. The DAL solution performs the well-known cable modem DHCP and TFTP transactions on behalf of each ONU registering on each OLT managed by the system. The DAL parses the contents of each config file and translates those contents into directives on the OLT. This provides the OLT with the necessary details to complete the service configuration of a given ONU.

Using an OLT with a DAL capability isn’t the only method to configure services on each ONU. Other methods include the use of SDN-based procedures and protocols such as Netconf or OLT supplier-provided RESTful APIs. But the use of the DAL and DOCSIS-style config files can ease the transition of member operators into ITU-T-based PON technologies, without the need for wholesale replacement of the operator’s service provisioning back-office.

Supplier Participation in the Provisioning of XGS-PON Interop

We thank the supplier participants who helped make the February interop event so successful. The primary participants were XGS-PON OLT suppliers who showcased their DAL implementations.

These OLT suppliers included Calix, Harmonic and Nokia. In particular, Calix tested using their E7-2 OLT and DPx DAL. Harmonic brought their Pier OLT and cOS virtualized core software platform. And Nokia tested using their Lightspan MF-2 OLT with Altiplano controller and a DAL from their solution partner, ZCorum. As a bonus, Nokia demonstrated that the DOCSIS provisioning method can also be used to provision services on a 25GS-PON optical network unit (ONU).

In addition to the OLT supplier participants, MaxLinear was in attendance with their reference-design XGS-PON ONU, and TraceSpan participated with their NG-PON Xpert analyzer, whose DPoX Validation capability was instrumental for validating the correctness of the OMCI configurations that were set by the OLTs.

How the XGS-PON ONU Test Cases Unfolded

During the interop event, CableLabs provided an identical set of XGS-PON ONUs of various makes and models to each of the three OLT suppliers. CableLabs also provided a set of standard config files — one for each of the test cases defined in the interop test plan.

By utilizing a common set of ONUs and config files, we could see how each OLT-plus-DAL system reacted to a given config file on a given ONU. The overall goal of the event was to work towards a given ONU having an identical service configuration — for a given test case — across all the OLTs.

The test plan executed at this event was derived from the DOCSIS Provisioning of ITU-T PON tech report, which CableLabs published in December 2024. Test cases included basic HSD configuration of ONUs with a single Ethernet port as well as ONUs with multiple Ethernet ports. When configuring services on ONUs with multiple ports, a config file construct called a Cable Modem Interface Mask (CMIM) Classifier was utilized to control which ports on the ONU were active. Testing also included a triple-play (HSD, IP Voice and IP Video) test case, which utilized Differentiated Services Code Point (DSCP) traffic classifiers in the config file. A final test case that was executed from the interop test plan included the use of an ONU with an integrated eRouter functionality.

In cases where an OLT supplier was not initially able to successfully configure a given ONU for a given test case, we allowed the supplier to make customizations to the standard config file. This enabled the supplier to execute the test case and note where changes in their software might be needed to successfully use the standard config file in the future.

Each test case executed during the event made use of the lab’s 100GE traffic generation system, the lab’s standard DHCP and TFTP server infrastructure, and a PON analyzer to decode the ITU-T-specified ONU Management and Control Interface (OMCI) message exchanges across the PON. Many thanks to Charter Communications and TraceSpan Communications for supplying additional NG-PON Xpert analyzer units, enabling each OLT supplier test setup to utilize a dedicated analyzer.

Beyond the primary DOCSIS provisioning focus of this February event, we were pleasantly surprised that one of the supplier participants in this event brought OLT and ONU software versions that demonstrated partial compliance to the Cable OpenOMCI I01 specification. We look forward to seeing many more suppliers follow suit in the near future.

Applying the Lessons Learned

Overall, the results of the event were encouraging, with both established as well as new supplier entries into this product space showcasing their abilities to provide common service configurations across a wide variety of ONU makes and models.

The implementations exercised at this event indicate that these suppliers believe that a DOCSIS Provisioning of XGS-PON solution will be an important part of their product portfolio for the foreseeable future. We plan to take the lessons learned from this event and apply them to a future update of the DOCSIS Provisioning of ITU-T PON tech report.

Join the Next PON Interop

CableLabs is planning three more PON Interop·Labs events to be held in our Louisville, Colorado, labs in 2025. Our next event is scheduled for the week of April 28, 2025.

The April event will provide an opportunity for OLT suppliers to return with their DAL solutions to once again test ONU and config file interoperability. In addition, ONU and OLT suppliers with software implementations compliant to the Cable OpenOMCI I01 specification will be invited to exercise the capabilities defined in that specification.

There has been momentum in the broadband industry to provide fiber to the premises (FTTP) solutions as a part of an operator’s service portfolio. While optical technologies have long been a part of the cable network, FTTP is the chosen architecture for all-fiber access networks.

For FTTP, the premises can be a subscriber’s home, a commercial location, a campus environment, a multi-dwelling unit (MDU) or another location. Over the past couple of decades, operators have embraced the passive optical network (PON) technology for their fiber-based access network implementations. The point-to-multipoint topology of PON lends itself nicely to the current and future designs of the broadband network.

More recently, many operators have begun deploying PON technologies defined by the International Telecommunications Union Telecommunications Standardization (ITU-T). As a part of its standards development, ITU-T has released standards for 10 Gigabit Symmetrical PON (XGS-PON).

With broadband operators beginning to deploy multiple gigabit service, there’s a growing current of enthusiasm and interest for the efficient deployment, management and maintenance of those access networks. Furthermore, the speed at which technology is moving is impressive and expensive. It is challenging to keep pace with these advancements, which require a matrix of expertise and decision-making support.

Building on PON’s Momentum

PON is one of the technologies that keeps marching forward. CableLabs has participated in the development of PON-based standards and specifications for over a decade, and we’re continuing in that vein to help operators lower barriers for deploying and operating FTTP solutions.

Common provisioning and management of PON in the broadband industry typically requires support of legacy systems that have been in place for decades, e.g., DOCSIS® operations support system (OSS) or integration with newer back-office systems. Near-term objectives for the work CableLabs is beginning include XGS-PON and 25GS-PON support, covering applicability for next-gen PON flavors with special focus on vendor neutrality through device interoperability.

CableLabs has created two working groups dedicated to optimizing the integration of ITU-T PON technologies into cable networks. The two working groups are the Common Provisioning and Management of PON (CPMP) and Optical Operations and Maintenance (OOM). These two groups are complementary in their activities. The CPMP group is focused on supporting the back-office provisioning and management of XGS-PON as well as the interoperability of optical network units (ONUs) with optical line terminals (OLTs). The OOM working group is focused on the operations and maintenance of the underlying optical networks.

Hurdles in Interoperability

There are always tradeoffs with technology, whether it’s price, timing or things out of an operator’s control, such as product availability. However, when customer premises equipment (CPE) is required, there is always an underlying benefit to the interoperability of that device with the network that it connects to.

Interoperability provides necessary competition, which results in pricing benefits, innovation and choice for operators. Having the choice of which ONU is connected to the OLT is instrumental in providing lower cost services with the ability to help foster innovation.

Interoperability involves several technical and logistical challenges. Addressing these hurdles typically involves a combination of adherence to standards, thorough testing and collaboration between vendors to ensure equipment from different manufacturers can work together seamlessly. These hurdles to interoperability include:

• Standardized technology: Standards are not necessarily the end-all-be-all for successful interoperability. In the context of XGS-PON, there is the ONU Management and Configuration Interface (OMCI). This OMCI information is defined in the ITU-T G.988 standard. This is the accepted way to configure and manage ONU equipment via the OLT. OMCI is extremely comprehensive and as such, it is very difficult to provide simple and extensible ways to support interoperability.
• Vendor-specific implementations: Vendors may support proprietary features or extensions in their products. While these can offer enhanced performance, additional functionality and vendor differentiation, they can also create interoperability issues if these features are not supported universally across different vendor equipment.
• Operator requirements: Each operator deployment is different, and each operator requires a specific implementation to support their business objectives. While certain configuration parameters, fault reporting and performance monitoring are common among different implementations, there are always variations, and this requires different configurations for network components and CPE.
• Network configuration, service activation, and management: XGS-PON networks require precise network management and configuration to ensure proper operation. Differences in management systems and configuration approaches between vendors can create integration challenges.
• Testing and certification: Comprehensive testing is required to ensure that equipment from different vendors works together as expected. Many operators and vendors don’t have the infrastructure to support such testing, which becomes a barrier when choosing equipment suppliers. A significant impact on time-to-market is testing. Testing and validation of requirements is a significant cost and effort.

Mitigating Interoperability Hurdles

CableLabs, with the support from vendor partners and member operators, has long had a successful formula for developing specifications that have transferred to CPE interoperability. Device interoperability is the primary objective when producing interface specifications between network components. Vendor neutrality through that device interoperability is a key intention for operators deploying XGS-PON in cable networks.

Developing documentation to standardize technology with the intent for interoperability is a fine line to walk when working with multiple operators and equipment manufacturers. To add fuel to the fire, when standards have already been written, developing a method for interoperability can require modifications to the original formula. By leveraging the existing G.988 OMCI standard, CableLabs is developing a Cable OpenOMCI specification to support ONU configuration in an interoperable way within the industry.

While it is always a goal to develop common processes and requirements that encourage the vendor ecosystem, sometimes vendor-specific parameters are required to support vendor differentiation. It’s important for vendors to differentiate their products, which provides their teams to innovate. The work CableLabs is doing to support device interoperability includes the ability for vendor-specific configurations when needed.

Each operator is different, and each operator has its own set of business objectives that translates to the products it buys, the services it offers, and the networks it builds. When developing specifications to support a myriad of operator requirements that span the globe, we must always be cognizant on ways to support those requirements. During our specification development we work together to identify and accommodate those differences needs across the industry.

While each network is different, for interoperability, some network configuration, service activation, and management must be standardized. This can be required anywhere in the network from the back office all the way to the end of the network with CPE. When building a program to support interoperability this entire ecosystem is in play. The coordination effort and the development of a common set of processes to support interoperability with device configuration, service activation and management is key.

In February, we hosted a DOCSIS® Provisioning of XGS-PON Interop·Labs event for members of the CPMP working group. Testing and certifying devices are very important for interoperability. To help remove the barriers of cost and reduce time-to market with comprehensive testing, CableLabs provides formal interoperability events like this as well as ongoing interoperability activities. Additionally, if needed, CableLabs will develop a certification program to test and certify ONUs against the specification.

Pushing Toward XGS-PON Adoption

While XGS-PON is gaining interest with cable operators, the industry is keen to develop procedures and specifications to support ONU interoperability with OLTs. While there are always common hurdles to interoperability, they are never enough to stop interoperability. Since the XGS-PON technology has already been defined and standardized by ITU-T, this does bring a different hurdle to the work.

Typically, CableLabs and its vendor partners and member operators develop the specifications that support interoperability. However, in this instance, we must leverage existing standards from another SDO. It won’t be impossible, but it will be a heavy lift.

Learn more about our working groups and how CableLabs members and our vendor community can help us chart the way forward for technologies like this.

Field operations teams are essential for ensuring the seamless performance of modern networks, including hybrid fiber coax (HFC), fiber and mobile infrastructures. However, as networks grow in complexity, traditional troubleshooting methods — manual workflows, static documentation and reliance on expert technicians — struggle to keep pace.

Enter Agentic AI, a transformative approach that embeds intelligent, autonomous AI-driven agents into field operations. These agents act as subject matter experts (SMEs), each specializing in a specific aspect of network troubleshooting, decision support and workflow optimization. By leveraging this structured, multi-agent AI system, organizations can scale expertise, reduce resolution time and enhance operational efficiency.

The Evolution of AI in Field Operations

From AI Assistants to Structured Agentic AI

While AI-powered virtual assistants have been used to support technicians with knowledge retrieval, they lack structured decision-making capabilities. Agentic AI introduces a multi-agent system where each AI agent specializes in a distinct area of expertise, much like human SMEs within an organization.

These specialist AI agents are grouped together into teams based on their expertise, allowing groups of specialist AI agents to collaboratively solve problems in a shared area of expertise, like human teams within an organization. This multi-agent team approach allows for efficient and accurate decision-making to address field operations tasks.

How Agentic AI Operates as a Team of SMEs

Agentic AI is structured around multiple specialized agents, each performing specific roles within the troubleshooting and network maintenance process. These agents collaborate dynamically, ensuring that every decision is informed by real-time data and domain knowledge.

Key AI agents include:

• Knowledge Retrieval Agent (SME in Network Standards & Best Practices)
  • Retrieves industry standards, vendor documentation and best practice guides.
  • Cross-references sources such as CableLabs specifications and SCTE standards. 
• Telemetry Analysis Agent (SME in Real-Time Network Monitoring)
  • Continuously monitors network logs, telemetry data, meter measurements and service degradation patterns.
  • Detects anomalies like signal degradation, upstream noise or fiber attenuation.
• Troubleshooting Workflow Agent (SME in Guided Resolutions)
  • Generates step-by-step troubleshooting workflows based on real-time conditions.
  • Adapts workflows dynamically based on technician feedback and sensor inputs.
• Decision Support Agent (SME in Root Cause Analysis & AI-Driven Recommendations)
  • Synthesizes insights from multiple agents to determine the most effective resolution.
  • Suggests alternative troubleshooting paths if the initial fix does not resolve the issue.
• Proactive Network Maintenance Agent (SME in Proactive Network Health & Failure Prevention)
  • Uses historical patterns and AI-driven models to detect potential failures before they occur.
  • Recommends preemptive maintenance to avoid service disruptions.

These agents are combined into agentic teams that are tailored to different areas of expertise, such as impairment types, enabling targeted collaboration and troubleshooting. By structuring AI in this multi-agent, SME-like framework, Agentic AI mirrors the way expert teams collaborate in real-world field operations, ensuring that each aspect of troubleshooting and maintenance is handled with precision.

Agentic AI in Action: Enhancing Field Operations

AI-Driven Troubleshooting for Faster Resolution

With Agentic AI, network troubleshooting shifts from manual trial-and-error approaches to data-driven, AI-guided processes. When a technician encounters an issue, the AI agents work together to provide precise, real-time recommendations.

Example: Resolving Signal Impairments in HFC Networks

1. The Telemetry Analysis Agent detects signal impairments from network telemetry.
2. The Knowledge Retrieval Agent pulls relevant troubleshooting workflows from specs, standards and vendor manuals.
3. The Troubleshooting Workflow Agent generates a guided resolution process, suggesting tests with field meters.
4. The Decision Support Agent analyzes technician input and network readings, refining recommendations dynamically.
5. If the issue is a recurring fault, the Proactive Maintenance Agent flags it for proactive intervention.

This real-time, multi-agent collaboration ensures that field technicians receive expert-level guidance instantly, reducing mean time to resolution (MTTR) and improving service quality.

Scaling Expertise with AI-Driven SMEs

Transforming Field Training & Knowledge Retention

A major challenge in field operations is scaling knowledge across teams. Traditionally, new technicians rely on classroom training and shadowing experienced engineers. With Agentic AI, expertise is available on demand — every technician, regardless of experience level, can access AI-powered SMEs for troubleshooting guidance.

Key Benefits:

• Faster onboarding: New hires gain instant access to SME-level knowledge, reducing training time.
• Standardized troubleshooting: AI ensures consistent best practices across teams based on SCTE Learning and Development guidelines.
• Knowledge retention: AI continuously learns from past cases, preserving institutional expertise.

By deploying Agentic AI as a structured knowledge system, organizations can scale expertise at unprecedented levels.

Beyond Troubleshooting: AI-Powered Proactive Maintenance

Future Work: Predicting & Preventing Failures Before They Occur

Instead of reacting to service disruptions, Agentic AI enables a shift toward proactive and predictive network maintenance.

• The Predictive Maintenance Agent continuously analyzes historical network performance trends.
• AI identifies early warning signs of network failures, such as cable degradation, fiber attenuation, or RF noise issues.
• The system recommends preemptive maintenance actions, reducing truck rolls and service downtime.

By integrating predictive analytics into field operations, network operators can lower costs, minimize disruptions and improve customer experience.

As the world of broadband connectivity evolves, so does the technology that powers it. The evolution of DOCSIS technology has been a cornerstone of high-speed internet delivery, enabling service providers to meet the growing demands for faster and more reliable internet access. DOCSIS 4.0 technology represents the latest milestone in this journey, offering impressive speeds and new capabilities.

The latest Interop·Labs event, which took place Feb. 10–13 at CableLabs in Louisville, Colorado, showcased these capabilities and more, focusing on speed and stability — crucial factors in determining the quality and reliability of an internet connection.

A Bright Future for DOCSIS 4.0 Technology

DOCSIS 4.0 technology is poised to revolutionize broadband connectivity, offering significantly faster speeds, better efficiency and enhanced security. Its backward compatibility with existing DOCSIS 3.0/3.1 infrastructure makes it a flexible solution for both service providers and consumers.

As with any major technological leap, ensuring smooth interoperability between old and new systems is key to a successful transition. By offering a scalable, future-proof solution that can be deployed gradually, DOCSIS 4.0 technology allows operators to gracefully migrate to these new products.

Why Speed and Stability Are Critical

Speed refers to how quickly data is downloaded or uploaded, often measured in megabits per second (Mbps), and plays a key role in activities such as streaming, gaming and video conferencing. Stability, on the other hand, is about the consistency of that connection.

A stable connection ensures that users experience minimal interruptions or slowdowns, even during peak usage times. Both high speed and stable performance are essential for seamless online experiences, especially in households or businesses that rely on multiple devices or heavy internet usage.

Key Contributors and Participants

The DOCSIS 4.0 specifications describe both frequency division duplex (FDD) and full duplex (FDX) modes of operation. For this interop, the first of 2025, new modems and RPDs that supported both modes (on the same device) were participants for the first time.

Attendance at the interop was high and included new suppliers and products, as well as two operators that joined to observe demonstrations, interact with the suppliers and discuss their own DOCSIS 4.0 network progress.

Suppliers included CommScope and Harmonic, who brought DOCSIS 4.0 CCAP cores to the event. For the first time, we saw new Remote PHY Device (RPD) platforms from three separate suppliers: Harmonic, Teleste and Vecima. Seven DOCSIS 4.0 modem suppliers — Arcadyan, Askey, Gemtek, Hitron, Sagemcom, Sercomm and Ubee — brought multiple cable modem models. Calian participated with its test solutions, and Microchip participated with its clock and timing system.

Broadcom has developed new chipsets that enable a device to operate in either FDD or FDX mode, under the control of the network. Another chipset supplier, MaxLinear, was also in attendance, exhibiting its latest DOCSIS 4.0 modem innovations.

Testing scenarios involved using a virtual core from one supplier, and RPDs and DOCSIS 4.0 cable modems from various other suppliers. The products were mixed and matched to verify interoperability scenarios and speeds through the system. As before, DOCSIS 3.1 and DOCSIS 4.0 devices were combined to demonstrate the cross-compatibility of existing and new technology. Suppliers providing test equipment used these setups to verify their solutions.

Broadband Service Enhancement

The February interop was another opportunity to drive interoperability between some of these new products/platforms and some existing ones, ensuring that the different systems and devices would seamlessly work together to exchange information. Interoperability is vital to enabling these devices to connect and interact. By fostering interoperability, the DOCSIS ecosystem continues to provide innovation and solutions to enhance broadband services, ultimately leading to better outcomes for consumers.

Remote PHY Interoperability

Although modems were the focus, the event also looked at the interoperability between DOCSIS 4.0 cores and RPDs. We branched into more and different configurations, moving beyond one-size-fits-all configurations. These products are going to be deployed in many scenarios, and going deeper into these various configurations will ensure flexibility as DOCSIS 4.0 technology moves into the field.

Join Us Soon

The next DOCSIS 4.0 interop is planned for the week of March 31 at CableLabs.

If you are a member or vendor attending CableLabs Winter Conference, join us for our two-part session “The What, Why and How of Future HFC Solutions.” We’ll explore the potential of future generations of DOCSIS networks, exploring how the evolution of the technology will support tomorrow’s hybrid fiber coax networks. This two-part session begins at 12:45 Tuesday, March 11.

With DOCSIS 4.0 technology, the future of broadband looks brighter, faster and more connected than ever before. The technology’s seamless integration into current systems ensures that consumers will continue to benefit from fast, reliable internet — no matter where they are on their broadband journey.

Modern networks deliver impressive speeds — often reaching gigabits per second — yet they still suffer from unpredictable delays that can disrupt interactive applications. Whether it’s video conferencing, cloud gaming or remote collaboration, these inconsistencies can lead to frustrating user experiences. As network operators strive to enhance reliability and responsiveness, a more effective solution is needed.

To address this need, the Internet Engineering Task Force (IETF) — an organization responsible for developing open internet standards — has specified the Low Latency, Low Loss and Scalable (L4S) throughput architecture. L4S enables applications to implement a new mechanism to ensure that they are sending their data as fast as the network can support, but no faster. The result is efficient capacity usage with minimal queuing delay and low packet loss.

This shift to a more comprehensive quality of service (QoS) model is essential for delivering smooth and uninterrupted performance across a wide range of services, from gaming and video streaming to cloud computing and augmented reality.

What Is L4S?

The power of L4S stems from new congestion control algorithms that adapt to new fine-grain notifications of congestion at the IP layer across various network elements along an end-to-end (E2E) path. While L4S can be deployed on each element of the network, its most significant impact is at points of congestion, also referred to as network bottlenecks — where the rate of incoming packets can exceed the departure rate.

The cable industry already adopted support for L4S, as part of the Low Latency DOCSIS® 3.1 specifications (and it carries forward into DOCSIS 4.0 gear as well). Operators are beginning to enable this functionality in their networks, and more are expected to do so over the coming months. That said, the broadband access network segment is only one potential bottleneck. There are others on the E2E path.

The Need for L4S in Wi-Fi Networks

Wi-Fi networks, in particular, require L4S support as they are frequently a point of congestion in E2E networks. Indeed, although Wi-Fi networks often advertise their maximum capacity that is greater than the broadband access connection, actual performance is significantly influenced by factors such as the distance between clients and the access point (AP), as well as the number of APs and clients operating on the same channel. This need for L4S support is even more critical given that a substantial portion of internet traffic is transmitted over Wi-Fi.

WBA L4S Implementation Guidelines and NS3 Simulator

Wi-Fi presents unique challenges for L4S implementation compared to wired technologies such as DOCSIS® networks. While wired networks primarily see only buffering delays, Wi-Fi additionally introduces media access delays, which can be significant in congested environments. To tackle these challenges, CableLabs worked within the End-to-End QoS Working Group of the Wireless Broadband Alliance (WBA) to produce a set of guidelines to implement L4S in current Wi-Fi products.

The guidelines cover:

• An overview of L4S technology, explaining its mechanics and benefits.
• The importance of L4S support in Wi-Fi equipment for improving E2E application performance.
• Implementation strategies for Wi-Fi equipment suppliers to enable L4S functionality in their products.
• Simulation and test results demonstrating the advantages of L4S in real-world scenarios

The simulation results are based on a Wi-Fi NS3 model developed by CableLabs, designed to evaluate L4S performance in Wi-Fi networks. The model is open-source and available to industry and research players to support L4S deployment and assess its impact on various use cases. In addition, CableLabs provided field test data that were conducted with a Nokia AP.

The Future of L4S in Wi-Fi

Wi-Fi equipment suppliers today can leverage the L4S Implementation Guidelines to develop support on their existing platforms (e.g. Wi-Fi 7 devices). Several proposals from industry leaders, including CableLabs, aim to incorporate L4S support into the 802.11 Wi-Fi standard, ensuring native support for L4S across future Wi-Fi generations.

In addition, as the ecosystem matures, CableLabs will continue refining the NS3 model to expand its applicability to more scenarios and use cases. This ongoing effort is being advanced in collaboration with the WBA E2E QoS working group.

L4S is a critically important next step in the evolution of the internet that solves many of the issues that cause frustrations today, where it seems that bandwidth alone hasn’t fully enabled reliable and responsive interactive application experiences.

To fully take that step, the segments of the network that are the likely bottlenecks in residential deployments — the access network and the Wi-Fi segment — both need built-in support for L4S.

Are you ready to strategize and share at Winter Conference 2025? In just a few weeks, an exclusive gathering of CableLabs member operators, approved vendors and industry thought leaders will meet in Orlando, Florida, to discuss and collaborate on how to move the industry forward.

Join us March 10–13 as we work to unlock the value of seamless connectivity and capitalize on the opportunities ahead. Member registration closes Monday, February 24. Secure your spot to take part in charting a course for success in the next era of connectivity.

The Path Ahead

Winter Conference sessions will be an in-depth exploration of the Technology Vision, with each session highlighting individual aspects of the framework — the architectures, protocols, technologies and strategies required for building and evolving the network of the future together. Hear from the experts on topics such as Network Evolution, Security Evolution, Differentiated Services and AI and take home practical insights to inform and inspire your own journey.

By aligning on a shared vision and working together to foster more robust, efficient and sustainable industry ecosystems, our member operators and the vendor community can collaboratively unlock new levels of agility and create solutions at scale. The Technology Vision, developed in collaboration with our members, lays out a roadmap for this more connected, intelligent and secure tomorrow.

Key Winter Conference Highlights

Register today to join us for high-impact sessions, plus hours of additional networking and vendor connections during provided meals and Exhibit Floor receptions. Members-only project meetings will be held Thursday, March 13.

Highlights this year for members and exhibiting vendors include Zack Kass, our keynote speaker. The AI futurist and former head of go-to-market at OpenAI will share his vision for a future powered by AI, offering actionable strategies for leveraging technology to accelerate innovation. Also, new this year, discover how four global startups are disrupting connectivity, focusing on solutions to enhance network performance and reduce latency during our neXus: Startup and Scaleup Spotlight.

View the full agenda for more details. All sessions and demos at a glance:

Tuesday, March 11

Sessions

• Shedding Light on Tomorrow’s Optical Access Networks
• The Rise of AI Automation: A Catalyst for Network Innovation
• Broadband Lunch: Insights Into Forecasting Network Demand
• The What, Why and How of Future HFC Solutions (two parts)
• Shaping the Security Evolution: Strategies from Industry CISOs
• The Power of NaaS: Simplifying Connectivity With Zero Touch Onboarding
• The Next Generation of Wi-Fi Access: Transforming the Connected Home and Beyond
• Breaking Barriers: Buildling Scalable Telemetry for Multi-Access Network Innovation

Demonstrations

• Mobile Wi-Fi Steering to Cellular
• Spectrum Monitoring
• Adversarial Machine Learning and Machine Augmented Penetration Testing
• neXus: Startup and Scaleup Spotlight: Redefining the Connected Home Experience
• neXus: Startup and Scaleup Spotlight: Making Wi-Fi Smarter Through AI-Powered Spectrum Slicing
• NaaS Featuring Traveler Wi-Fi

Wednesday, March 12

Sessions

• What Lies Ahead: Shaping the Future of Connectivity
• Spectrum Policy: Opening a New World of Connectivity Options
• API-First Networks: Building the Foundation for Future Services
• Redefining the Road to Reliability
• Security Evolution: Threats, Technology and Product Differentiation
• A Coherent Path to FTTP Deployments
• The Quest for Seamless Connectivity
• Context Is King: Unlocking Differentiated Services With Smart Networks
• Balancing Economic Trade-Offs in the Last Mile

Demonstrations

• neXus: Startup and Scaleup Spotlight — Unlocking the True Potential of Advanced Networks
• neXus: Startup and Scaleup Spotlight — Revolutionizing Network Operations with Agentic AI
• Harnessing ATSSS: Seamless Traffic Switching for Uninterrupted Connectivity
• Mobile Wi-Fi Steering to Cellular
• Spectrum Monitoring
• Adversarial Machine Learning and Machine Augmented Penetration Testing

Smaller Market Conference

Ahead of Winter Conference, we invite small and mid-tier multi-system operators to join the Smaller Marketing Conference on Monday, March 10, to discuss the issues that are most important to them and their teams. This event is available for CableLabs members and NCTA guests. Registration for Smaller Market Conference is separate from Winter Conference, so if you are planning to attend, please be sure to register for both.

Make Your Mark at Winter Conference

Are you ready to impact the future? Join us at Winter Conference 2025, and be part of the conversation on guiding the industry’s path forward. Beyond simply being an opportunity to

Attendees will not only learn more about the Technology Vision. They will also take part in important conversations that will shape the direction and influence the success of that strategic vision.

Registration for Winter Conference 2025 is open for members through Monday, Feb. 24. Secure your spot today.

As we dive into a new year, new administration and new Congress, one thing remains the same: we will need more spectrum to power our increasingly data-centric lives. Almost everything we do today is online — from working, learning, accessing health care and buying groceries to streaming content. Our everyday activities rely on the exchange or consumption of data. As discussed in our prior blog, “The Case for Additional Unlicensed Spectrum,” we see that case only growing stronger every day.

Traditionally, policymakers, including the Federal Communications Commission (FCC), focused spectrum decisions around coverage, to make sure that consumers across the United States could access communications and the internet. Now that almost all of the country is covered by communications and internet services, we need to shift our priorities to focus on meeting consumers’ and the industry’s capacity and performance needs. A helpful starting point is to look at how people and businesses are consuming data now and where technology is driving more demand.

The workhorse of connectivity, Wi-Fi carries more than 90 percent of all consumer internet traffic. Looking at just mobile devices and the nation’s largest wireless providers, Open Signal found that for every one bit carried on a mobile network, nearly 9 bits are carried on Wi-Fi. Wireless providers and their networks fundamentally rely on Wi-Fi to manage their capacity and traffic because the majority of smartphone usage occurs at home, where they are predominantly connected to Wi-Fi. The overwhelming reliance on Wi-Fi is expected to continue for the foreseeable future.

The Economic Impact of Surging Wi-Fi Demand

Unsurprisingly, given the heavy consumer reliance on Wi-Fi, another recent study found that the annual U.S. economic value of Wi-Fi is expected to grow from $1.6 trillion in 2024 to $2.4 trillion by 2027, including an estimated $514 billion in consumer benefit, $624 billion in producer surplus and $1.29 billion in GDP contribution. The same study found that the FCC’s April 2020 decision to open additional unlicensed spectrum to support Wi-Fi growth (in the 6 GHz spectrum band) generated $870 billion in economic value, just from 2023-2024, and that number is expected to increase to $1.2 trillion by 2027.

The annual CES in Las Vegas always provides a preview of our increasingly connected world — from wider adoption of health care applications to richer and more immersive entertainment to new smart home applications. One primary enabler for all these applications is the ubiquitous, plentiful and inexpensive wireless connectivity provided through Wi-Fi.

Not only will these innovations add more devices and applications to the already significant load on Wi-Fi frequencies, but they will also be increasingly data intensive. The new devices and applications will increasingly require higher speeds and lower latencies to work. A recent ABI Research report projects that current-standard (6 GHz-supported) Wi-Fi devices will grow from 95 million in 2024 to 367 million in 2029 — an increase of 288 percent in just five years — in North America alone.

That’s why the Wi-Fi industry has already developed a new Wi-Fi standard that will use wider-bandwidth channels to support these types of applications and more efficiently use available unlicensed spectrum. The catch is that they need more contiguous unlicensed spectrum bands to make enough wide channels to keep pace with consumer needs and technology advances.

How We Can Evolve With Growing Demands

Without more unlicensed spectrum, Wi-Fi performance will degrade as more devices, applications and users come online. Diminished performance is expected first in areas like dense commercial and residential areas, such as universities, apartments and office buildings that host high concentrations of users and devices. As the number of devices grows and more data-intense applications become widely adopted, Wi-Fi demand will continue to grow across the country.

The expected Wi-Fi spectrum exhaustion in densely populated environments creates a problem, as roughly a third of Americans live in apartments, condos, connected townhomes or other multi-family buildings, and even more people work in office buildings with more than one tenant. Wi-Fi exhaust could start to appear within a small space, or even a block, where multiple Wi-Fi networks are operating. These networks could be supporting multiple devices per household or office and multiple users, all sharing the same unlicensed frequencies.

Initially, degraded Wi-Fi performance will occur at peak usage times. As more devices, users and more data-intensive applications — such as telemedicine and augmented reality/virtual reality (AR/VR) — start to compete for the same channels, that degraded performance will extend for more time and more broadly to more locations.

More unlicensed spectrum is the key to averting this looming problem.

Fortunately, policymakers already have a successful model to follow — the FCC’s 2020 decision to make the U.S. the first country in the world to open the 6 GHz band to unlicensed commercial use by sharing with incumbent licensees, including public safety, utility and broadcast links. Other governments around the world are following the United States’ leadership, given the significant economic and technological success.

Thanks to the U.S.’s leadership and forward-thinking, American consumers are the first to experience better, faster and more reliable Wi-Fi service almost everywhere they go. But technology will not stop advancing, and consumers will only expect better connectivity. Given the astounding growth and demand for Wi-Fi, the U.S. must again lead the way to create room for continued unlicensed innovation, establish a foothold in the Wi-Fi and unlicensed global technology markets, and model how other allied countries can create economic value, expansive access and innovation.

Below, we outline some of the technical considerations as we look more closely at the looming Wi-Fi spectrum exhaustion issue.

Unlicensed Spectrum Efficiencies Across Dimensions of Frequency, Time and Geography

Wi-Fi and other unlicensed technologies are highly efficient stewards of limited spectrum, designed to maximize the utility of every MHz of available spectrum. This enables unlicensed technologies to support a wide variety and intensity of devices, applications and use cases. Wi-Fi’s efficient use of spectrum is built on low-power use — minimizing (but not eliminating) geographic overlap of uncoordinated networks, equitable time-sharing through “listen before talk” or similar contention-management techniques, and the support for the full channelization of all available unlicensed frequencies.

• Geography: Most notably, the FCC’s 6GHz rules constrain the transmission powers of both access points and clients, limiting the coverage area of these Wi-Fi networks, therefore reducing the number of geographically overlapping networks in an area.
• Time: Wi-Fi and other similar unlicensed technologies incorporate contention-management techniques that enable geographically overlapping Wi-Fi networks to use the same (or partially overlapping) frequencies. Networks then equitably share those frequencies in time between the two or more overlapping networks.

Figure 1: In general, when there are physically overlapping Wi-Fi networks on the same channel, only one network can access that channel at a time.

• Frequency: The 6 GHz band provides seven non-overlapping 160 MHz channels or three nonoverlapping 320 MHz channels. This has enabled geographically overlapping networks, like those in neighboring apartments, to reduce but not eliminate frequency overlap increasing available capacity to each Wi-Fi network.

Figure 2: Current Wi-Fi channelization of the 6 GHz band and potential future Wi-Fi channelization of the 7/8 GHz band. (Click image to enlarge.)

Through the dimensions of geography, time and frequency, Wi-Fi provides an incredible amount of data-carrying capacity, but it isn’t inexhaustible.

As we look to the future, these techniques, as well as further advances in Wi-Fi technology, are unlikely to meet growing demand from the devices and applications that rely on Wi-Fi — without additional unlicensed spectrum.

With additional unlicensed spectrum, however, Wi-Fi would have access to additional channels, further reducing conflict and performance degradation in networks that are close to each other and increase the instantaneous bandwidth and throughput available to applications. This would further reduce the likelihood of collisions in time.

Growing Demands Will Strain Available Unlicensed Spectrum

The explosion in the number of connected devices and the increased throughput, latency and performance demands of existing and new applications will exceed the available and future Wi-Fi capacity from currently available spectrum. As ABI Research explains, by the end of the decade, most households will have 6 GHz-capable Wi-Fi 7 access points, and we’ll begin to see the transition to Wi-Fi 8 access points.

• Total Number of Connected Devices: In 2024, the average household had 18 connected devices, but as more devices add Wi-Fi connections, such as appliances and lightbulbs, we regularly see households with 40, 50, 75 or more. ABI Research forecasts that the number of connected devices will continue to grow, with total annual shipments of Wi-Fi-connected devices to North America growing by 35.1 percent each year from 2023 through 2029. Although many of these devices are relatively low bandwidth (e.g., thermostats, lighting controls, appliances), each regularly transmits and receives data, increasing the probability of crowding and contention with other devices on the network. As the number of total connected devices increases, the amount of unused unlicensed spectrum in a home or business decreases. This will most negatively affect latency-sensitive applications, such as video conferencing and immersive, online gaming, which require ready access to uncongested spectrum.
• More Demanding Applications and Devices: The ceaseless march toward higher picture quality and the growing incorporation of AI capabilities will drive increased Wi-Fi traffic in the coming years. For instance, we see all the major streaming platforms moving toward some version of 4K Ultra HD, high dynamic range (HDR) formats (e.g., Dolby Vision and HDR10+). In parallel, smart television manufacturers are rapidly producing devices that support these more data-intensive formats. These new formats and future formats and associated devices will continue to drive and require increased bitrates. For instance, streaming a Dolby Vision video to a smart TV can easily have a bit rate of 25–30 Mbps. We expect even higher bitrates for video formats being streamed to AR/VR goggles as higher resolution and picture quality are key to these experiences. Moreover, we expect the widespread adoption of AI capabilities by device manufacturers and consumers to further drive Wi-Fi traffic, including from increased upstream video and telemetry data for AI processing in the cloud, and increased frequency and size of software updates to support locally deployed AI models.
• New Client-to-Client Connections: To further enhance the utility of 6 GHz, the FCC recently enabled very low power (VLP) unlicensed devices to use the band. With much lower power levels, VLP connections are very short range. The primary use cases for VLP include wearables, such as watches, rings and earbuds, and other short-range use cases such as connectivity between AR/VR goggles and a TV or between a gaming controller and console. What sets many of these connections apart is that the same bit will make multiple wireless hops in the home before it reaches the Wi-Fi access point and the broader internet, driving substantially more demand for unlicensed spectrum capacity.

We anticipate that the growth and increasingly intense and demanding use of Wi-Fi will first lead to unlicensed spectrum exhaust and degraded performance in the most dense environments, like apartments, condos, schools, and office buildings, before manifesting in less dense environments.

Open Additional Unlicensed Spectrum!

To enable the connection-rich near future for all end users, the U.S. government (and governments around the world) must identify additional unlicensed spectrum. Today’s growing number of Wi-Fi devices, and the next generation of devices, that support applications in education, health care and entertainment all fundamentally rely on ubiquitous, plentiful and inexpensive wireless connectivity that Wi-Fi and other unlicensed technologies provide.

Without additional unlicensed spectrum, the flow of transformative technology and applications will begin to slow down. CableLabs looks forward to continuing to work with policymakers and the broader industry to help open additional unlicensed spectrum.

The CableLabs Strategy team, collaborating with our Technology team counterparts, develops Key Business Questions (KBQs) specifically focused on strategy and impact on the bottom line. These strategy KBQs provide a solid strategic underpinning to the innovations in the Technology Vision, and they can distill our members’ opportunities and challenges as they make critical technology decisions and investments.

Considering the various technology options available for access network upgrades — combined with evolving consumer needs and exciting opportunities created by technology innovations — now is the right time for CableLabs' member operators to leverage these KBQs to inform the analyses we produce for our members. By developing strategic questions that are tailored to their organizations' unique goals, members can position themselves for long-term success.

Strategy KBQ Goals

Strategy KBQs align the Strategy team’s work with the needs of our individual members, ensuring maximum impact with focused CableLabs resources that can complement member companies’ existing strategy teams. In this way, CableLabs can help inform the critical strategic decisions that members make on innovations and technologies. The strategy KBQs enable our members’ boards, executives and execution-level leaders to make decisions and take actions for long-term success.

CableLabs offers our members optionality both in technologies and in strategies. The strategy KBQs don’t seek one-size-fits-all answers; rather, they lead to a flexible set of analyses that members can tailor with us according to the particulars of their businesses, markets, competition and goals.

Strategy KBQs in the Real World

The CableLabs Strategy team has been carefully distilling strategic issues for network evolution into strategy KBQs like these:

• How will peak bandwidth demand grow in the future and what are the implications for capacity planning and product development?
• What are the economic and performance comparisons of the access network technology options (HFC, FTTH, Wireless) for cable operators?
• How do low earth orbit (LEO) satellite broadband economics scale with increased capacity, increased usage and increased subscribers?
• How do Fixed Wireless Access (FWA) broadband economics scale when adding loaded spectrum to meet demand for increased usage and subscribers?

Each strategy KBQ comes with a cascade of relevant questions that addresses issues at different layers within a member’s organization, such as strategy, product and engineering. The operator can use the analyses driven by KBQs to inform critical decisions and operational actions.

How can members apply strategy KBQs and the analyses that answer them? Here are some examples:

• Knowing peak demand requirements for 2035, operators can design their broadband access networks and deploy technologies to meet the requirements specific to their markets, instead of investing purely in greater capacity for capacity’s sake.
• Knowing the options for product positioning, such as low-cost, no-frills provider or differentiated high-end provider, and how different network technologies satisfy those positions, operators can make decisions on how to position their offerings most successfully in their markets.
• Knowing revenue impacts together with total cost of ownership that results from positioning options, competition, and broadband access network technology choices, members can form a solid view of the economic returns specific to their markets for different technological investments.

No two broadband access networks are alike — we’ve searched for the “typical network” and found there is no such thing! The same is true for the answers to the strategy KBQs; while answers may be similar, there will be no one-size-fits-all strategy for cable operators. Members can utilize strategic insights from CableLabs, addressing a range of results from different choices, to develop the answers that are right for their markets.

Translating to Action

CableLabs has structured Winter Conference 2025 to focus on the six Technology Vectors. In each session, members can learn more about the Technology Vision and Technology Vectors, and they can obtain a preview of the strategy KBQs for each Technology Vector. Panelists will share their differing perspectives on technology choices, future use cases and key business questions.

Just as our members’ markets are always evolving, the strategy KBQs will also evolve. Members can contribute to refining the strategy KBQs going forward, just as with the Technology Vision and Technology Vectors.