Microsoft Word Solving The Han Firmware Distribution Problem

Solving the HAN Firmware
Distribution Problem
Where we are today and where we must go
By Marcia Martin

Abstract
Energy utilities are beginning to consider the deployment of consumer home-
area networks that operate in conjunction with the Advanced Metering
Infrastructure to enable consumers to actively and selectively monitor and
control their energy consumption and participate in partnership with the
utilities in demand management programs. As these deployments reach the
millions of households in number, a network-based mechanism for distributing
firmware updates and managing device compatibility must be found.

Many issues must be settled to accomplish this in a way that ensures the
behavior of the entire HAN, in the presence of a device upgrade, is
predictable. Standards and institutions must be established to preserve
consumer autonomy while ensuring the security and availability of the Smart
Grid. This whitepaper surveys the problem space and provides the reader
with a basis for further investigation.

303–453–8380
magpie@MagpieTI.com
www.MagpieTI.com/smart-energy-practice

October 2009

Solving the HAN Firmware Distribution Problem

Why Worry about Firmware Distribution?
Providing the ability to remotely upgrade Home Area Network
(HAN) devices used for energy management is a critical need
for the Smart Grid. Why? HANs were originally something of
an adult toy, providing the means for homeowners to
centrally and remotely control their high-end home video,
audio, and lighting equipment. Home automation devices
contained firmware, but it was supported by the AV reseller
and the consumer. Devices were upgraded, if at all, at the discretion of the
consumer. Typically, they worked as installed for the life of the system.

Today, however, the same home automation technology is being put to a
much more serious use – opening up a two-way channel of communication
between the consumer and the public utility. This will enable consumers to
better control their energy costs, and become the utility’s partner in the
crucial task of managing energy consumption to create a more stable,
sustainable grid.
Contents
Why worry about Firmware What can Energy Management HANs do?
Distribution? ..................................... 2 An Energy Management HAN connects individual devices in the home to the
What can Energy Management
Advanced Metering Infrastructure or AMI network. Utilities are deploying AMIs in
HANs Do? ...................................... 2
major cities around the United States already. The AMI data communications
Managing the HAN as Critical
network connects the utility meter on every house in the utility service area with
Infrastructure ............................... 3
Requirements on the Smart Grid ......... 4 the utility’s data center, providing a two-way communications channel. The utility
Emerging Deployment Model ................ 5 collects data about the home’s energy consumption in more detail than has ever
Conclusions ........................................... 7 been possible. The utility can also transmit commands and information to the
Abbreviations and Acronyms ........... 8 “smart” meter. However, the meter alone can’t do much to affect the household’s

2
October 2009 www.MagpieTI.com/smart-energy-practice


energy consumption. The smartest of smart meters have the ability to
communicate with home-area networked devices inside the home. Connecting the
smart meter to the HAN opens up a world of possibilities for home energy
management. This energy management strategy can:

• Collect data about not only how much power the whole house is using, but
determine which devices in the house are consuming, and when.

• Display information to the homeowner about current rates, so that
consumers can make informed decisions about when to perform high-
consumption activities.

• Alert consumers about upcoming maintenance events, anticipated peak
consumption periods, and so on.

• Most importantly, communicate directly with individual smart devices in
the home to remotely modify the home’s energy consumption. This is
called Demand Management. It will become an increasingly important tool
for utilities in keeping the energy grid stable and available, and in
controlling cost to consumers.

Managing the HAN as Critical Infrastructure
Now that the Home Area Network is playing a critical role not only in the
household but in the community, it has to be managed in a different way from
when it was just a toy. These are the three main differences to consider:

• Security. There was little motivation for anyone to “hack” a HAN when
gaining access could, at worst, allow the hacker to turn the home’s stereo
on and off. But an Energy Management HAN could be collecting and
transmitting data that consumers might wish to keep private between
themselves and the utility. Worse, spoofing demand-response commands
to many households at once could actually destabilize or damage the grid.

• Availability. As soon as consumers become dependent on the enhanced
service levels they can enjoy with the Smart Grid, then they will consider it
a service failure, for example, if their electric bill jumps because the
devices in their home have failed to carry out their energy management
policies. They will hold the utility responsible.

• Currency. Utilities will need to be able to manage households uniformly
in order to maintain the security and availability of the energy
management network. This problem becomes more complex if many
versions of the several communications protocols they employ must be
supported. Just as with the software on your business’s desktops,
everything works better when everyone’s up-to-date.

3


Requirements on the Smart Grid
The critical need to maintain HAN Security, Availability, and Currency in the
firmware on HAN devices imposes some requirements on the way utilities and
energy service providers must deploy the data communications portion of the
Smart Grid that enables connectivity between Home Area Networks and the
utilities’ or Energy Service Provider’s network operations center (NOC).

Enabling Firmware Distribution
What mechanisms need to be put in Need Requirements State of the Art
place to make it happen?
• HAN devices need to support • NEMA SG-AMI 1-2009 defines
a standard protocol for requirements for
accepting a code load and commanding smart meters to
restarting the device. accept a network code load,
• If NEMA SG-AMI 1-2009 but does not define a
Standards requirements regarding code standard protocol.
loads were applied by • Device vendors have
analogy to HAN devices, it proprietary methods.
would facilitate safe firmware • Many HAN devices deployed
upgrades. today do not meet the
analogous requirements.

• Nothing forces consumers to • Smart Energy Profile (SEP) is
deploy only devices approved an emerging de facto
by their utility. So industry- standard for energy HANs.
wide interoperability testing • On 9/14/2009 Landis+Gyr
Testing and certification is a must. announced a HAN device
interoperability certification
program to SEP 1.0.
• SEP 1.0 does not specify any
code-load protocol.

• All energy management HANs • Connecting HANs to the
must be connected to a internet is expensive.
network that can support the • AMI infrastructures today are
load from firmware updates proprietary and diverse. Most
Network and its normal workload. can support networked
• Two candidate networks are firmware updates, but some
the Internet and the AMI need careful management to
network. do so.

• Some entity must control • “Who owns the HAN?”
distribution of firmware • Homeowners feel they do.
updates to maintain Security, • Utilities want easy support.
Owner Availability, and Currency in
• Manufacturers want control.
the utility service area.
• NEMA SG-AMI 1-2009 implies
central distribution center.

4


Emerging Deployment Model
A workable combined model for maintaining energy management HANs
consistently within a utility service area seems to be emerging.

HAN support deployment model
The public utility, commercial Energy
Service Providers, and consumers will Advanced
be partners in maintaining the home Metering
Infrastructure
delivery segments of the Smart Grid,
including smart devices inside the
home.

Neighborhood Area
Concentrator
Utility
Consumer Energy
Cellular or Services
Portal
Fiber Provider (ESP)
Backhaul

Public
Internet Utility

In this picture, the home area networks in these homes are connected to the
Advanced Metering Infrastructure via the utility meters on their houses. AMI
networks lack the uniformity and transparency we’ve become used to in the World
Wide Web. A number of different vendors are competing to supply utilities with
network infrastructure for their Smart meters. Several different physical
architectures have already been deployed in the United States and Europe. In the
US, a plurality of vendors favors the RF Mesh architecture. This has the lowest
deployment cost, and poses relatively few technical and regulatory issues because
the RF signal used is in the unregulated 900 Mhz band, the same frequency band
used by older cordless home telephones.

The table below summarizes the leading AMI network architectural models. RF
Mesh AMI networks are of most concern it they are to be used for firmware
distribution, because of their low effective bandwidth compared to the other
models. If each firmware load is broadcast to all meters, regardless of whether
the household served contains a device to receive it, then perhaps RF Mesh AMIs

5


can be used for this purpose. However, using connection-oriented download
strategies to target individual devices within the household, and to control the
order in which devices are updated, is much more problematic, because it would
be easy to overrun the network. A carefully chosen distribution strategy taking
advantage of detailed knowledge of the network topology would be required.

Comparing AMI Network Types
Network
Architecture Characteristics Vendors

• Most common in the USA • Itron
• Lowest effective bandwidth of • Landis+Gyr
any AMI architecture • Silver Springs
• 900 Mhz RF band • Trilliant
• One “concentrator” tower for • Elster
every 50-5000 meters
RF Mesh
• Maximum raw bandwidth
meter to meter is about 20
kbps, or only 2000 characters
per second
• Effective bandwidth is much
less in a busy network

• More popular in Europe • Echelon
• Some RF Mesh solutions in • Smart Synch
Cellular the US use Cellular backhaul
• 3G data rates exceeding
36MB/sec

• Bandwidth can be comparable • Echelon
to broadband internet • Enel
Power Line • Expensive devices to
(PLC) modulate signal on power line
• Can cause RF interference
• Good for rural areas

Typically, to deploy an AMI network, a utility will collaborate with two or more
equipment vendors. The utility usually standardizes on one meter manufacturer,
such as Itron or Landis+Gyr. Some meter vendors also sell and manage the
network infrastructure, while others do not. Some of the infrastructure vendors
also produce devices that can convert first generation intelligent meters to two-
way AMI Communications.

To date, utilities have not attempted to develop their own software to manage the
data collected by the AMI infrastructure, whether or not it includes HAN
extensions. Commercial entities that do this in partnership with utilities are

6


termed Energy Service Providers (ESPs). Some AMI networking vendors also
provide the energy management software services, and so are also ESPs. Other
ESPs, like Tendril, do not manufacture or deploy AMI networks or devices, but
work in partnership with the hardware vendors. Regardless of the supply model, it
is the ESP which ultimately has the closest working relationship with the utility.

Data collected from HAN devices is transmitted over the AMI network, which is
connected to the Internet backbone by means of some sort of concentrator. The
concentrator may reach the Internet either via a cellular wireless connection or
have a direct Ethernet connection to the fibre backbone. The data is collected at
a Network Operations Center run by the ESP. The ESP carries out energy
management policies set on the one hand by utilities (rates, demand management
advisories) and on the other hand by consumers (home energy policies such as
acceptable temperature ranges for the home, times of day when the utility may
shut off some devices such as HVAC, or goals for energy savings). Because of
this, the ESP is emerging as the likeliest candidate for controlling firmware
distribution to intelligent energy-management devices in the home.

Consider the logistical advantages that the ESP has over other possible candidates
for managing software distribution:

• The ESP’s NOC is an appropriate repository for firmware images ready to
be distributed.

• The ESP must already have good information about what devices are
present in each household it services.

• The ESP already has relationships with the primary device vendors.

• The ESP can serve as “gatekeeper” to decide whether to admit third-party
devices into the energy-management HAN.

• Because the ESP serves multiple utilities, the effort of interoperability
testing, making deployment decisions, etc. is reusable. It would be too
much effort for each utility to perform this testing independently.

• Because ESPs are participants in standards organizations, they receive
early warning about changes and can benefit from the interoperability
testing performed by standards bodies, vendors, and industry associations.

Conclusions
Because all AMI/HAN deployments by public utilities are in their infancy, the
firmware distribution problem has not yet been well tested. In the rush to be first
to market, this problem, which inherently arises in the second generation, has not

7


been the top priority for ESPs or their utility partners. Today, however, as
important regulatory agencies such as NIST, FERC and the Department of
Homeland Security are focusing on AMI deployments and insisting that strong
security be built into these networks from the beginning, the firmware distribution
issue is moving up the list. A secure network cannot be maintained without a
secure reliable method of updating the nodes in the network. The energy industry
can profit from the experience gained in the 1980s and 1990s with wide area
networks. The costs of failing to build security into any networking solution from
the beginning are now well understood.

As the Smart Grid leaves the prototype era and becomes a firm reality, software
architecture and the ability to develop software to exacting compliance with
published standards will become increasingly important to new energy startups, to
utilities, and to the large traditional players in the energy management space.
Magpie has been solving problems of this nature in the communications arena
since 2001. Our growing Smart Energy practice applies this critical know-how to
this exciting new field of endeavor. To date, we have provided software services
and consulting to HAN device manufacturers, ESPs, business-area-network
providers, and industry/academic consortia.

Abbreviations and Acronyms
The primary acronyms and abbreviations used in this publication are listed here:

Acronym Definition

AMI Advanced Metering Infrastructure

ESP Energy Service Provider

HAN Home Area Network

NEMA National Electrical Manufacturers’ Association

NOC Network Operations Center

PLC Power Line Carrier

SEP Smart Energy Profile

8


About The Author
Marcia Martin Marcia is a systems analyst and architect whose focus
is applying the Smart Energy Profile into smart grid
business and systems applications at Magpie.
Previously, she was a StorageTek Fellow where she
produced numerous publications and earned eight
individual patents in the Data Storage domain. Marcia
has also been recognized by the Women in Technology
International. She holds a Master of Science degree in
Computer Science from Washington University's Sever
Institute of Technology.

9


Magpie — Engineered Software That Works™

Magpie is a valued partner in delivering engineered software that works™. The
company specializes in solving tough technology problems for communication–
enabled applications, business support systems, interactive Internet applications,
and network protocol integration. The Magpie team is well known for its Agile,
iterative, and open software development process. With customers spanning
North America, Magpie has consistently delivered on–time, on–budget results for
the telecom, health information, and emerging smart–grid energy industries. The
company warranties all its work and also has unique expertise in providing
systems architecture, automated testing, systems integration, and technology
consulting professional services. Make Magpie your choice when your business
depends on software™. For more information, visit Magpie on the web at
www.magpieti.com.

Copyright © 2009 Magpie Telecom Insiders, Inc.
All Rights Reserved. This publication, in whole or in part, and the associated electronic file(s) may
not be reproduced, transmitted, or distributed in any form (print, electronic, or other media)
without the prior written consent of Magpie Telecom Insiders, Inc. Information in this publication is
subject to change without notice. Any statements or opinions expressed in this publication are
provided without warranty, either expressed or implied, and without representation for their
suitability or applicability in any business or personal use. Magpie Telecom Insiders, Inc., their
employees, agents, and partners assume no responsibility for any errors that may appear in this
publication nor any liability for any decisions or conclusions any person or business may make as a
result of these statements and opinions.

Magpie and the Magpie logo are trademarks or registered trademarks of Magpie Telecom Insiders,
Inc. Third party trademarks, trade names, logos, and product names referenced in this publication
may be registered trademarks or trademarks of their respective owners.

Contact Us
Worldwide Headquarters
Magpie Telecom Insiders, Inc.
12050 N. Pecos St. Suite 210
Denver, Colorado 80234

Roxanna Podlogar, SVP Smart Energy Practice
303–453–8380
magpie@MagpieTI.com
www.MagpieTI.com/smart-energy-practice

10

Microsoft Word Solving The Han Firmware Distribution Problem

More Related Content

What's hot (14)

Viewers also liked (6)

Similar to Microsoft Word Solving The Han Firmware Distribution Problem (20)

Microsoft Word Solving The Han Firmware Distribution Problem