Remote PHY: A Sense of Reality

Intraway Symphonica: Accelerating new service launches with cloud-native automation

Remote PHY is an approach to distributed access architecture (DAA) in a digital hybrid fiber/coax (HFC) network. It removes the physical layer from the CMTS headend to the network edge. By reducing the distance to the customer’s devices, it obtains higher bandwidth and throughput than a centralized system. It also helps to reduce operational costs with higher bandwidth for Internet services such as voice and video.

Learn how MSOs can reduce the time to implement & manage the configuration of a distributed access architecture (DAA) with a low-code/no code orchestration platform in our Remote Phy Deployment Made Simple webinar.

Remote PHY: A Technical Recap

R-PHY technology pushes the physical RF layer to the edge of the access network. It needs the CCAP to be “split” between the MAC layer and the PHY layer. In a Remote PHY operation, the integrated CCAP is divided into two separate components. The first one is the CCAP Core, and the second component is the RPD. The CCAP Core can include a CMTS Core for DOCSIS technology and an EQAM Core for Video.

Figure: Remote PHY High-Level Architecture

Figure: Remote PHY High-Level Architecture

 

The RPD (Remote PHY Device) contains PHY-related circuitry, such as downstream QAM modulators, upstream QAM demodulators, and pseudowire logic to connect to the CCAP Core. The functions of the RPD device are:

  • To convert downstream DOCSIS, MPEG video and out-of-band signals to analog for transmission over RF.
  • To convert upstream DOCSIS and out-of-band signals received from an analog medium to a CCAP Core.

RPD Functionality

RPDs may be located anywhere, but a usual location for an RPD is the optical node device that is located at the junction of the fiber and coax plants; also, the CCAP stays at the headend. Each one may have its router port, or they may be daisy-chained. A single box can contain multiple RPDs.

An RPD needs to run reliably without onsite intervention, and it has to deal with a possible unsafe physical connection to the CIN (Converged Interconnect Network). It employs encrypted communication to guarantee the authenticity and integrity of the data. Each RPD validates itself as part of its startup sequence; then it uses DHCP server to obtain its IP address.

Software updates are sometimes necessary to fix bugs and enhance security. With the devices located on the provider’s premises, there are no complications, but with equipment that might be a hundred miles away, or even close but outside the MSO’s trusted network, a precise and safe approach is required. Remote PHY states a secure software download protocol. An X.509 certificate validates the source of the update, which builds upon existing DOCSIS methods for updating cable modems. The R-PHY specification from CableLabs details how the software needs to be upgraded. This process is started over the Generic Control Plane (GCP) connection, so based on a secure connection that links the CCAP to the RPD.

If an RPD complies with the specification, it will work with any CCAP that has Remote PHY support.

The Architecture

The converged cable access platform (CCAP) unifies the delivery of voice, video, and data through the DOCSIS protocols. Its previous form is called integrated CCAP or I-CCAP, and it implements OSI layers 1 through 4 to send data over HFC, which includes Layer 1, the physical or PHY layer.

Such a degree of centralization has its limits, as Remote PHY moves Layer 1 from the CCAP to a remote PHY device and connects to it by UDP and TCP/IP over Ethernet. The RPD includes QAM modulators and demodulators. The network connecting to the RPD is referred to as a converged interconnect network or CIN, and it contains routers, switches, or both.

As well as reducing the distance from the node to the customer, the Remote PHY approach improves scalability: the CCAP is no longer limited by the number of lines that can connect to one box.

The RPD includes some Layer 2 and 3 functionality, including media access control (MAC) and pseudowires to the CCAP.

Distributed Access Architecture

Several methods of adjusting networks to digital HFC are underway. All of them involve some DAA, moving some of the functionality closer to the network edge. Remote PHY is attractive due to its simplicity and efficiency in implementation.

PHY hardware implements the physical layer (layer 1) of the OSI model. It moves its functionality from the CMTS headend to the edge of the network. The connection between the CMTS and the PHY uses any implementation of layers 2 and 3.

Some of the benefits of Remote PHY are:

  • Providing compatibility with any IP network architecture at the higher layers.
  • Improving the network’s modularity.
  • Transitioning from existing CMTS architectures in a seamless manner.

The Remote PHY specification provides compatibility among manufacturers’ products so that operators do not have to be tied to a vendor. However, the challenge in a multi-vendor CCAP and/or RPD environment will be for an MSO to have a solution in place to design and orchestrate the deployment of the RPD. In order to be efficient in cost and operational complexity, the MSO will need to have a flexible and vendor-independent OSS Solution (Virtual Infrastructure Manager + Service Order Manager + Network Connector). This solution must help the interoperability of RPD/CCAP of any vendor. 

Alternative Approaches to DAA

While Remote PHY outlines relatively small changes to architectures that are based on linear HFC, other methods decentralize the CMTS more radically.

Remote MAC and PHY move the MAC portion of the architecture to the network edge. A more sophisticated approach is splitting the DOCSIS functionality between the centralized and remote portions. Even though it is a logical continuation of Remote PHY, it may be too radical to obtain support in the short term.

Remote CMTS is more decentralized. Instead of a large, single CMTS, there is one CMTS at each remote node, which is less comprehensive than Remote MAC since it keeps the CMTS together. Nonetheless, it implies shifting system management to a wide array of remote nodes instead of one local one. Among other factors, it poses a new set of security concerns. In the end, as MSOs need to evolve their access networks, many analysis, articles, and POCs (from HW/SW vendors also) are suggesting the pros and cons of the architectural options. They all have one common statement: Remote PHY looks like the leading candidate for DAA, as it is currently available, while the alternatives are still theoretical or without proper ROI.

The Meaning for Telecom Operators

The primary and more obvious benefit of Remote PHY is speed. The architecture does not require moving PHY hardware out to all the nodes in one single operation. A combination of local and remote PHY locations will not present any problems.

The most straightforward approach is to connect the remote PHY device to a port extender configuration. The RPD is still physically close to the central CMTS rather than to the customer. It is the first step in decentralization and allows more coax connections.

Placing the RPD in genuinely remote locations, closer to the customers, will gain a more significant advantage, but it requires deploying fast digital fiber connections to the nodes. Moving the PHY connections out can be a gradual process, spreading the cost over time. Areas with a high concentration of customers will deliver the most significant benefit for the least new cabling, so they make good pilot sites.

Moving to Remote PHY is a significant project that requires careful planning. The benefit of it is that it will let MSOs deliver higher speeds over DOCSIS, allowing more services and producing more significant revenue.

Challenges in Deploying Remote PHY

Customers demand higher data rates. Remote PHY will let MSOs achieve them by moving the QAMs closer to their sites. At the same time, any distributed access architecture (DAA) will pose some considerable challenges. Knowing the issues in advance is essential for precise deployment and to deal with maintenance issues.

Access

The most common problem is physical access. While RPNs have diagnostic capabilities, there are times when hands-on troubleshooting and service are required.

Having as many devices as possible in a central location has benefits for maintenance. An RPN might be located miles away from the primary office. It might be in an inconvenient, outdoor area. Maintenance teams have to add travel time and expenses when they need to service the nodes in person, and they need to have mobile service toolkits.

Remote PHY has to seek a balance between bringing service close to the customer and being able to deal with the maintenance issues. Expanding too quickly is not expensive but could also create logistical issues that the maintenance team is not ready to handle.

Provisioning

The provisioning of customers’ on-premises equipment has not changed significantly within this architecture. IP addresses are still assigned from a central DHCP server, and customer-specific parameters can be set up by the operator or as part of the customer’s online registration procedure.

Remote PHY nodes need to be secured against tampering even if they are in an insecure physical in a location. CableLabs specification requires RPNs to authenticate with the CCAP, guarding against man-in-the-middle attacks. Firmware updates are installed using a reliable procedure.

Hands-on tampering is one of the most challenging attacks to guard against, but the Remote PHY specification includes strong protections against it. To prevent this situation, it is essential to keep the remote nodes and CCAP updated with the latest security patches. 

In terms of having a closed-loop provisioning flow, it is very important that the technician has a field service tool, probably a smartphone type, to support and guide the technician not only through the physical installation but also to provide KPIs for installation quality. These KPIs are usually related to optical power, CPU usage, memory, power levels, IP address, and interface status, among others.

You can learn more about Intraway’s Symphonica Infrastructure Manager here.

Quality of Service

Remote PHY increases bandwidth and QoS by reducing the distance for the RF connection, but the range for the Ethernet connection demands some attention. Definitely, the latency to the RPN should be low and consistent. The quality of the Ethernet connection to the RPN is critical. The longer the path, the more challenging it becomes.

Remote DEPI specification deals with the pseudowire connecting the CCAP to the RPD. The specification defines requirements for buffering and allowable skew, as well as requirements for UDP and MPEG-TS. Installations should ensure an acceptable quality, keeping the number of network hops within limits.

DOCSIS 3.1 incorporates Active Queue Management, which selectively drops packets to keep a consistent data rate. It helps to avoid latency issues when TCP pushes the data rate up to unsustainable levels. The limited dropping of packets signals TCP not to increase its data rate further when the buffers are close to full.

The Demands of DOCSIS 3.1

One of the primary reasons for adopting Remote PHY, of course, is to meet the higher throughput standards of DOCSIS 3.1, and in particular to enable full-duplex communication. Shorter coax distances are not only an advantage but a competitive necessity. As the network becomes more extended, it has to improve the quality of the data path at every step.

The move to Remote PHY, or any other form of DAA, is a major change for cable networks. Deployment and maintenance will require more fieldwork. Change is an inescapable part of data and video networking, and these changes will open up vast opportunities to the providers that deal with it best.

For sure, the deployment of Remote PHY will transform how traditional HFC networks are installed, tested, and maintained, particularly as far as an optical layer is concerned. Remote PHY technology offers considerable benefits to the HFC networks. Adding digital fiber links provides MSOs with solutions that are more cost-effective and well-established in the industry. 

Furthermore, it supports more data capacity due to enhanced spectrum reuse. On the other hand, operators have to handle not only multiple CCAP cores but also numerous nodes having the RPDs that are connected in a leaf-spine architecture.

As the fiber nodes and CCAP cores may be provided by different vendors, it is crucial to make sure that all of the components are interoperable. MSOs looking to migrate to Remote PHY are already facing the challenges of designing, deploying, and managing the CIN. Therefore, operators are involved in careful and constant planning to provide the resources that will guarantee a successful transformation of their networks.

Learn how MSOs can reduce the time to implement & manage the configuration of a distributed access architecture (DAA) with a low-code/no code orchestration platform in our Remote Phy Deployment Made Simple webinar. 

You may also like

GPON Technology

Basics to Understand GPON Technology

Efficient and Economically Viable Deployment of Gigabit Broadband

computer security concept

Virtual CCAP Infrastructure Management

Menu