Closed Loop Capacity Optimization for Extended Spectrum DOCSIS (2020)

By Dr. Thushara Hewavithana, Intel Corporation; Dr. Rainer Strobel; Nader Foroughi, Shaw Communications Inc

Achievable data rates in the fixed access network keep increasing. While passive optical networks (PON) move to 25 or 50 Gbit/s with IEEE 802.3ca and MG fast targets 10 Gbit/s aggregated point-to-point rate over twisted pair, it is time to evaluate HFC technology as a successor for 10 Gbit/s capable DOCSIS, using full duplex or 1.8 GHz bandwidth.

Figure 1 compares the data rate trends for different access technologies. DSL, as a point-to-point technology is at lower rates, but with a higher growth rate. While DOCSIS and PON, both shared medium technologies, follow a similar trend with lower growth rate of the aggregated rate, which is compensated by reducing the number of subscribers sharing the bandwidth as an additional measure. From Figure 1, aggregated data rates around 30 Gbit/s are a competitive choice for a future DOCSIS generation, which is herein called extended spectrum DOCSIS (ESD). This will allow 20-25 Gbit/s downstream (DS) and 5-10 Gbit/s upstream (US) rates, which is comparable to a single 25G PON wavelength service that is shown in to serve future access network requirements. Following the arguments of, this will allow for 10 Gbit/s services and cover the bandwidth growth predicted by Nielsen’s law.

The cable industry has recognized ESD as a viable path to extend competitiveness of DOCSIS network at a fraction of cost compared to fiber deployments going forward. Under the 10G DOCSIS initiative, 1.8GHz frequency division duplexing (FDD) DOCSIS and 1.2 GHz full duplex DOCSIS options have been included in DOCSIS 4.0 as two possible ways of getting to 10 Gbit/s node throughput, enabling low single digit Gbit/s services. We can consider 1.8 GHz FDD to be an intermediate step to get to the 3 GHzESD and 10 Gbit/s services. In this paper, we will focus more on 1.8 GHz ESD when describing algorithms and evaluating performance results. Nonetheless we will maintain the forward compatibility of our algorithms for a future 3 GHz ESD solution.

One of the key enablers of ESD is the advancement of power amplifier (PA) technology that can support multi GHz transmit signal. However, total composite power (TCP) of these PAs does not scale with the increased spectrum beyond 1218 MHz. Therefore, the ESD communication system is limited in its transmit power. Optimal allocation of available transmit power and appropriate bit-loading (profile definition) is needed to get the maximum capacity out of the network.

In this paper, we outline a framework for closed loop optimization of the capacity of ESD systems subjected to the TCP constraint mentioned above. Cable modems provide the node with channel estimate and signal to noise ratio estimate data and the intelligent node uses this data to calculate the optimal power allocation and bit-loading for the downstream. This can be made part of the profile management application running on a virtual Cable Modem Termination System (vCMTS). We have shown that combining careful allocation of channels, closed loop optimization of transmit power, and adaptive bit loading achieves considerable gains in data rate and reduction in TCP for network topologies currently present in MSO networks.

By clicking the "Download Paper" button, you are agreeing to our terms and conditions.

Similar Papers

Blueprint for 3 GHz, 25 Gbps DOCSIS
By John T Chapman & Hang Jin, Cisco Systems; Thushara Hewavithana, Intel Corporation; Rainer Hillermeier, Qorvo
2019
Distributed Gain Architecture: Increased Performance, Decreased Power Draw
By Nader Foroughi, Shaw Communications; Jason Rupe, CableLabs
2020
HFC Evolution: The Best Path Forward
By Nader Foroughi, Shaw Communications
2018
Dr. Strangeleak Returns
By Ted E. Hartson, Post-Newsweek Cable, Inc.
1990
Upgrading the Plant to Satisfy Traffic Demands
By Nader Foroughi, Shaw Communications
2019
Extended Spectrum DOCSIS: A Pragmatic Approach
By Steve Condra, Teleste Intercept LLC; Kari Maki & Arttu Purmonen, Teleste Corporation
2019
The Grown-up Potential Of A Teenage Phy
By Dr. Robert Howald, Robert Thompson, Dr. Amarildo Vieira, Motorola Mobility
2012
Using Mixed Integer Linear Programming for RF Bandwidth Optimization
By Tom Holloran, Charter Communications
2019
Implement Closed-Loop Network Decisioning Now with Big Data Analytics and Fuel Future-State SDN Use Cases Through a Common Platform Deployment
By Brennen Lynch and Anukool Lakhina, Guavus, Inc.
2014
Blueprint for Mobile Xhaul over DOCSIS
By Jennifer Andreoli-Fang, PhD, CableLabs; John T. Chapman & Tong Liu, PhD, Cisco; Damian Poltz, Shaw Communications
2019
More Results >>