News and New Products
Voices: Intellon’s Mark Hazen on the HomePlug AV powerline-networking alternative
How will today's diversity of powerline networking technologies sort out over time, how will the market potential be impacted until then, and how does powerline compete with, coexist with, and cooperate with other LAN and WAN interconnect approaches? Intellon's Mark Hazen fields these and other probing questions.
By Brian Dipert, Senior Technical Editor -- EDN, 5/15/2008
Last fall, powerline-networking provider DS2 weighed in with its thoughts on the technology's strengths, shortcomings and future directions (see "Voices: Chano Gómez on powerline networking's 'universal' hope"). In this follow-up interview, Mark E. Hazen, senior marketing communications engineer with HomePlug Powerline Alliance representative Intellon, DS2's primary competitor, shares his perspectives on a similar set of questions.
OVERVIEW
I'd like to begin by asking you to provide an introductory summary of the historical development and current status of HomePlug technology, as it applies to LANs, to broadband Internet access service distribution (WANs), and to other past and present applications (such as power-meter monitoring).
The emergence of the HomePlug Powerline Alliance began in 2000, followed by the release of the HomePlug 1.0 open standard in 2001. HomePlug 1.0 was designed to be a broadband-powerline-communications technology with very robust noise immunity capabilities and a 14-Mbps maximum PHY throughput rate. It could supplant or complement Wi-Fi (IEEE 802.11b), which was starting to see significant market acceptance in the US and abroad.
Initial market reaction to HomePlug 1.0 was somewhat mixed, for two main reasons:
- Concern that it could not perform over AC wiring and
- The allure of the wireless, untethered experience. Many people associated HomePlug 1.0 powerline communications (PLC) technology with early powerline-based command and control technologies that were unreliable, especially across phases in a home. Even now, there are a few people who still believe that the PLC signal cannot jump phases in a home (which of course it can).
Intellon moved the HomePlug 1.0 standard forward with development of a higher-speed version named "Turbo", which retained full backward compatibility with HomePlug 1.0. Turbo reaches PHY [physical layer] speeds of 85 Mbps, making it faster than IEEE 802.11g Wi-Fi.
During the early years of 2001 through 2004, consumers began to realize that Wi-Fi is great for portable applications as long as the signal can be maintained. Like any wireless technology, Wi-Fi signals experience attenuation by distance and by physical obstructions such as walls and furniture. This degradation is particularly true in larger homes, multistory homes or homes built with steel, stone, and mortar, as is often the case in Europe. Adoption rates for PLC began to accelerate as consumers realized that HomePlug PLC is a reliable alternative to Wi-Fi or, as is often the case, a complement to Wi-Fi, by providing a backbone for home networks to which consumers can connect additional access points in order to create complete wireless coverage in any home, regardless of size or construction.
As HomePlug products became increasingly successful, the global membership of the HomePlug Alliance was aware that still-faster PLC technology was needed to support the convergence of the entertainment and PC worlds. Thus, the HomePlug AV 200-Mbps standard was ratified in 2005. Intellon contributed heavily to this standard and immediately went to work to bring to market the world's first 200-Mbps HomePlug AV-based ICs and firmware, releasing the INT6000 IC MAC/PHY IC in 2006 and the INT6300 MAC/PHY/AFE IC in 2007.
All along the way, consumers' hunger for more bandwidth and content drove both Wi-Fi and PLC speeds higher, along with cable and DSL access speeds. An ultrabroadband superhighway from content providers directly to consumers is essential to enable video (especially HD video) content storage and distribution. Even the latest IEEE 802.11n Wi-Fi technology does not have the speed or coverage that new in-home applications demand. Consequently, HomePlug AV has met with tremendous market success by moving the consumer into the future.
The HomePlug Powerline Alliance is also addressing the Broadband Power Line (BPL) access markets, the Smart Grid markets, and the in-home Command and Control (C&C) markets. Smart Grid enables power companies to utilize PLC technology in monitoring and controlling the power-distribution grid to prevent blackouts through load redistribution and demand management.
Remote metering and commercial in-home C&C is an extension of Smart Grid. Remote metering allows for a fully automated data and billing ecosystem. Commercial C&C enables the power company to "reach" through the power meter to heavy appliances within homes and businesses, for example to temporarily turn off water heaters, air conditioners and other major electrical loads during periods of peak demand. Consumer C&C allows home owners to monitor and control the home environment, either locally or remotely via the Internet. These markets are now just starting to emerge with the HomePlug PLC and C&C standards in place.
The HomePlug Alliance now has over 75 members worldwide, a membership that includes the "who's who" of CE manufacturers, OEMs, and service providers.
In your mind, how do HomePlug technology's attributes enable it to coexist, supplement, and/or supplant other traditional data-distribution technologies, for both LAN (CAT5, HomePNA, MoCA, UWB, various 802.11 flavors, etc) and WAN (DSL, cable, fiber, cellular, fixed wireless, WiMAX, etc) applications, and both today and in the future?
Great question. HomePlug technology is playing a strong support role both inside (LAN) and outside (WAN) of the home. Because of HomePlug's attributes, such as whole-house coverage, noise immunity, a standards-based approach, secure communications, and multivendor support, operators view it as a network "backbone" technology, not supplanting other forms of home networking but instead providing a means to work in concert with other networking technologies such as Wi-Fi.
In fact, multiple HomePlug-Wi-Fi bridge products are currently in the market that allow consumers to expand their coverage simply by plugging in a bridge anywhere in their house where Wi-Fi coverage is weak or nonexistent. It is the value of HomePlug as this "bridgeable technology" that contributes to its growing popularity.
In addition to the support role for wireless, including UWB and cellular, we believe HomePlug AV is currently the only technology that provides reliable whole-house HD-video-streaming capability, with an emphasis on "whole-house," which leads me to HPNA and MoCA. These are clean-line technologies that work well over a limited number of connections available in the home. The average American home has roughly 43 AC outlets, compared to only a few telephone and cable jacks. HomePlug PLC technologies, of course, also work well over clean-line mediums.
HomePlug PLC LAN technology supports any to-the-home WAN technology and actually provides much higher speeds throughout the home than most services to the home. As a WAN technology, HomePlug is starting to see some growth, especially in areas of the world that are underserved by xDSL or cable Internet. Broadband-Over-Powerline Access (BPL Access) can provide typical broadband service to-the-home through the medium voltage (MV) lines across the neighborhood and into the low-voltage (LV) service lines feeding the home. In this way, a utility can participate and compete in the broadband-services industry, the same way a telco, MSO, satellite, or WiMAX provider might.
Smart Grid is another form of BPL. With HomePlug providing a very robust and stable communications channel, we are seeing indications that the utilities are recognizing HomePlug's capabilities for applications such as automated meter reading (AMR) and Demand/Response (DR) monitoring, coupled with load shedding. HomePlug may be a natural way to exercise self-awareness in the power grid external to the home. Utilities understand that controlling and monitoring their assets in the most efficient and effective way possible benefits everyone in the ecosystem: utilities, consumers, and the world's natural energy resources.
TECHNOLOGY SPECIFICS
How does latest-generation HomePlug AV modulate data on the AC power signal, and how does it handle detection, correction, and/or retransmission of errors in that data? How does HomePlug AV compensate for varying noise levels on the power grid, caused by fluorescent lights, motors (vacuum cleaners, hair dryers, heater and air-conditioning fans, etc)? And how do you educate consumers on the potential need to install noise filters on the power inputs of these interference sources, in order to ensure reliable powerline network operation?
HomePlug AV uses a multidimensional approach to the powerline noise problem. Among the techniques it employs are time domain mitigation, frequency domain mitigation, dynamic channel estimation, and a very efficient forward-error-correction method based on Turbo Codes and developed by France Telecom.
In the time domain, transmissions synchronize with the zero-crossing of the AC line cycle. Much of the noise on household AC wiring is time-related. In other words, noise events often repeate at the same locations of the AC cycle. Once recognized, it is a simple matter to time transmissions around the noise.
In the frequency domain, domestic AC wiring noise is not evenly spread across the HomePlug AV frequency spectrum (approximately 2 to 30 MHz). Noise appears at separated frequency points across the band. HomePlug AV employs orthogonal frequency division multiplexing, composed of some 1155 carriers spaced 24 kHz apart (only 917 of which actually find use with HomePlug AV, in consideration of the Amateur Radio community). Each carrier, when modulated, creates analog symbols that contain as many as 10 data bits. For carriers that frequency-cohabitate with noise, adjusted "bit loading" optimizes performance under that challenge.
Dynamic channel estimation is the intelligence behind noise mitigation. This channel-estimation process frequently and repeatedly takes place between every two nodes in the home PLC network, optimizing timing and carrier bit loading for each powerline-network path.
Forward error correction adds bit overhead, but it is essential for robust communications over any medium (wireless or wired). In many cases, FEC eliminates the need to resend a packet or block of packets when data is lost. FEC has the ability to reconstruct the data on the receive end in real time. France Telecom's Turbo Convolution Code technology achieves exceptional performance and realizes greater throughput in the presence of noise, allowing HomePlug AV home networks to operate at a lower SNR. This advanced FEC coding has power efficiencies approaching (within 1 dB of) the theoretical Shannon limit and is well proven, having been widely adopted in harsh communications environments such as cellular telecommunications (CDMA2000 and W-CDMA), satellite (DVB-RCS, CCSD), and broadband wireless (802.16, WiMAX).
How does HomePlug AV handle the reality that two power outlets (either in close proximity or not) that a consumer may want to interconnect via HomePlug may be fed by different circuit breakers, and (even more challenging) may derive from opposite phases of the 220V (US) source feed?
This question has its roots in users' experiences with low-frequency, low-bit-rate technologies such as the venerable X-10. HomePlug PLC technologies are vastly different in both frequency and speed. While X-10 operates at 120 kHz, HomePlug technology operates in the HF frequency spectrum, roughly 2 to 30 MHz. In this frequency range, capacitive coupling between phase wires is very strong. Think here of the twisted-pair wires from the transformer to the house, and more importantly, within the home going to the hot-water heater, clothes dryer, HVAC system, etc. The net result is that the bus bars in the breaker panel look like a low-impedance crossover point for the signals. The loss here is only a few dB.
Most circuit breakers present no problem to the PLC signals. Again, the amount of attenuation that circuit breakers contribute is very small compared to the dynamic range of the technology.
In all three of the above operating characteristics areas, how has HomePlug AV improved on its HomePlug 1.0 and HomePlug 1.0 Turbo predecessors? And why was HomePlug AV defined to be noninteroperable (albeit coexistence-capable) with HomePlug 1.0 and HomePlug 1.0 Turbo?
HomePlug AV is a 200-Mbps-class PLC technology, versus 85 Mbps for HomePlug 1.0 with Turbo and 14 Mbps for HP 1.0. HomePlug AV is designed to operate in a class of its own, with a stronger feature set for noise mitigation, and to serve the needs of IPTV [Internet Protocol television], VOD [video on demand], and streaming HD. Though it is possible to build a backward-interoperable device, the complexity would be prohibitively expensive to our customers. Such a device would have to include the ability to switch between FEC techniques and many other features, which places cost at an unworkable level. Our customers need optimized (performance, power, and cost) products. Often, much more than they want backward compatibility, they want optimization for targeted applications. With the HomePlug AV platform, Intellon sees a clear path to forward compatibility for future platforms, such as our 400-Mbps extension and beyond.
HomePlug AV is specified as a "200 Mbps" technology, but testing suggests that it delivers only a limited percentage of that speed in real-life usage environments. Why was the "200-Mbps" peak PHY rate chosen as the technology designator? What range of TCP (transmission control protocol) and UDP (user datagram protocol) speeds do you believe most consumers will experience in real-life settings, and what kinds of applications are supportable (and conversely unsupportable) by those speeds, both in single- and multiple-coincident-data-stream situations? And do you address potential consumer confusion and frustration, when they don't get the performance results that the "200-Mbps" stamp on the outside of the product box might otherwise suggest they'll achieve?
This is another great question. There is some confusion about this in all data-communications communities, including wireless ones. Take, for example, IEEE 802.11g Wi-Fi, which is known as a 54-Mbps technology. 54 Mbps is the PHY rate minus forward error correction bits. The standards bodies, during technology and specification development, decide which type of PHY rate they will associate with their standard. In the case of 802.11g, 54 Mbps is the coded PHY rate, as opposed to the higher raw PHY rate (a little over 70 Mbps). Few consumers are aware of this distinction. Even so, it allows the Wi-Fi standards people to churn out different standards classes with the same apples-for-apples PHY rating.
Regarding HomePlug AV, 200 Mbps is the raw PHY rate, which includes forward error correction bits. This approach follows that of the wireless technologies—associating a PHY rate with each wireless standard. Some of our customers who market in retail state on their boxes what UDP and TCP rates consumers should expect.
The user-datagram protocol is the fastest medium access control (MAC), protocol because it does not require acknowledgements to each packet sent. It finds use for streaming music and video, and for VOIP telephony. Our tests have shown that HomePlug AV typically delivers a UDP rate across the home in the 50- to 90-Mbps range, depending on noise and attenuation.
The transmission control protocol is slower than UDP, but is more careful, checking to make sure that each packet was successfully received and, if necessary, repeating the transmission. HomePlug AV typically delivers a TCP rate across the home in the 30- to >60-Mbps range, depending on noise and attenuation.
A protocol that many use to judge the performance of a digital technology is Windows File Transfer. This protocol runs under TCP and is therefore even slower than TCP. The peak Windows File Transfer throughput for HomePlug AV is around 36 Mbps.
It is important to understand that whether the technology is wired or wireless, this hierarchy of protocol performance is always true. We need more editors to understand this to help us educate the public. Thanks for asking the question in this forum.
Each successive iteration of HomePlug technology seems from my standpoint to focus on increasingly improving UDP performance, versus TCP. Assuming you agree with my perspective, why and how was UDP prioritized in the definition and implementation of the technology? And how does HomePlug AV stack up against other powerline technologies that enable consumers to customize TCP-versus-UDP prioritization, and conversely against other powerline technologies that offer no TCP-versus-UDP optimization capabilities?
The answer to this question crosses over into the quality-of-service arena, in which streaming or live content is always given priority over data. Therefore, UDP streams by default are "at the front of the line" for bandwidth allocation. The OEM or service provider can customize Intellon's firmware in order to redefine traffic priority.
Intellon has improved HomePlug AV performance over the past several years in all protocol areas, including UDP, TCP, and Windows File Transfer. Optimizing performances in all areas is our constant goal. It is interesting to note here that IPTV, which is a major focus for Intellon, is multicast—that is, essentially UDP.
HomePlug AV allows for QOS (quality of service) prioritization of particular types of data streams. How do you balance the desire for a robust consumer experience from a QOS (or other) standpoint against consumers' desire for a robust out-of-box experience (one that doesn't require tedious and confusing calibration and customization of QOS and other technology parameters)?
HomePlug AV is Ethernet-based. That fact means that Ethernet QOS gets mapped onto HomePlug's QOS. However, we have given our customers the ability to map QOS for specific applications. Service providers, of course, are interested in this. For consumer applications, the entire question of QOS is a moot point because the default QOS mapping follows Ethernet QOS, with priority for streaming media. Keep in mind that this does not mean that the consumer can only handle one type of traffic at a time. On the contrary, users can enjoy HD streaming in one room, IPTV in another, streaming music in another and general Internet activities at will. Our products recognize the content type and automatically set the priority. Therefore, the consumer has no need to understand or care about QOS.
In my hands-on experience, powerline often exhibits extended latencies as compared with other networking technologies. What (if any) applications are therefore not candidates for using powerline as their transport scheme? And how can applications (and the operating systems they run on top of) compensate for powerline networks' extended latencies?
This is an interesting question because what you most likely have experienced is a time- and situation-dependent event that is difficult to nail down. All digital communications technologies have latencies, which usually only become important in VOIP applications where buffering is not acceptable (as it is acceptable when streaming music and video).
Naturally, I too have HomePlug AV technology in my home. I have really been enjoying streaming full episodes of my favorite programs over the Internet, then through my AV devices to my entertainment center, as well as streaming video content stored on my PC. Even programs in HD flawlessly stream. However, I recently purchased a Digital Media Adapter (DMA) and began using it right away. I noticed that some videos stored on my PC would frequently and repeatedly pause and then continue playing—very annoying. Further investigation found that the DMA was at fault, not the HomePlug AV adapters. Some of today's DMAs will exhibit glitches even when directly connected to a router via an Ethernet cable.
Again in my hands-on experience, initial HomePlug 1.0 Turbo firmware revisions did not fully support UPnP and other network discovery/identification schemes, and end-user upgrades of adapter firmware weren't practically feasible. I'm glad to see that HomePlug AV appears to be more consumer upgrade-friendly, but perhaps obviously, a no-upgrade-required scenario is even more preferable. What did (and does) the HomePlug standards body do to ensure robust out-of-box technology support both for various network protocols and for interoperability of various manufacturers' HomePlug-cognizent equipment?
Firmware upgrades for any technology are a desirable thing. Pretty much everything you buy these days has a phone or Ethernet jack, in part so that the equipment can receive firmware updates. Sometimes they're to correct bugs and other times they're simply to add features and improve performance. I've noticed that DVRs often have menu items reserved for future upgrades.
Most of our OEM customers in the retail space offer a graphical interface to assist users with update downloads and installation, thereby simplifying the process. However, if the user decides not to upgrade, performance should still be satisfying. Intellon has a program of in-house and partner testing before the firmware gets to the customers.
How much of a concern is data security across a shared power-distribution topology, both in a multiresident neighborhood environment and in a multiapartment single-premises setting, and how do you consequently educate consumers on the potential need to change the default encryption password and make other security adjustments? What encryption scheme(s) does HomePlug AV use, is encryption enabled by default (and at what performance impact versus an encryption-disabled alternative configuration), and can the encryption protocol be upgraded or otherwise enhanced on a situation-by-situation basis?
This is always a good question for any digital-communications technology. Intellon has worked to build a simple connection scheme into the HomePlug AV standard. It enables pushbutton pairing and network-joining, along with (of course) comprehending the corresponding encryption key. Our customers are starting to offer this pushbutton feature, which makes pairing very easy; push the button on the first device, push the button on the second device, and you are done. Adding more network nodes occurs in similar manner. For products that do not yet have the pushbutton feature, a software disk provides customization, just as with Wi-Fi products.
Some of our customers are shipping devices with randomized passwords, which require the user to use the pushbuttons for pairing. Others are shipping pre-mated packaged adapter pairs, for secure out-of-the-box use. These are just a few of the available strategies that move users away from universal passwords that could compromise security.
The amateur (Ham) radio community has been quite vocal with its concerns regarding potential interference in the presence of an active powerline network, due to radiated powerline noise and consequent inductive coupling to the amateur-radio setup's broadcast and reception antenna. Are the concerns valid, and if so how has HomePlug technology been architected to mitigate these concerns (via notch filters and other schemes)...and at what potential impact to powerline performance and other robustness measures? What other potential destructive interference scenarios (wireless keyboards and mice, for example) exist?
The Amateur Radio Relay League (ARRL) has endorsed HomePlug PLC technology, which means that they have performed tests that confirm that HomePlug PLC does not interfere with Ham activities. Early on, Intellon worked to ensure that HomePlug technology would be a good neighbor to the Ham community. Even though HomePlug AV has 1155 carriers available for modulation, only 917 actually find use. The forfeited carriers are those in the Ham bands. If HomePlug AV used all available carriers, HomePlug AV would have a raw PHY rate of nearly 250 Mbps. Also, the signal strength of HomePlug AV on the AC wiring is extremely low, with very little radiated power, which otherwise would become a concern of the FCC.
Unfortunately, other non-HomePlug-based PLC devices exist that are not Ham-friendly. That's another reason why looking for the HomePlug logo on the product package is so important.
Finally, as I understand it, powerline technology is unable to work "through" a surge protector or battery-backed UPS (uninterruptable power supply). How do you educate consumers on dealing with this issue? And how do you deal with the fact that surge protection filters are increasingly being built directly into AC outlets?
Most of our customers include this information in their user booklets, which ship with the product. Naturally, OEMs such as Linksys, Netgear, and D-link want their customers to have a good experience with their HomePlug PLC products. Also, not all surge protectors are a problem. More than anything, it is best not to plug any PLC device into a power strip with other appliances. However, we have customers, such as Monster, that are offering entertainment-center power centers that actually have HomePlug AV embedded inside, cleanly filtered from other appliances.
COMPETITION, COMPATIBILITY, AND STANDARDIZATION
Does the HomePlug AV specification provide room for proprietary (and backward-compatible) standards-based enhancements, as was the case with HomePlug 1.0 Turbo versus HomePlug 1.0? If so, how does the standards body plan to handle cases of companies that decide to implement such enhancements?
HomePlug 1.0 with Turbo is proprietary to Intellon and is backward interoperable with HomePlug 1.0. Turbo is fully compliant with HP 1.0. There is no HomePlug standard for Turbo. Enhancements are permitted within the HomePlug specification, as long as the resultant product still fully complies with the HomePlug standard. The HomePlug Alliance only certifies products for specific standards compliance; they do not certify enhancements.
Your two primary competitors, as I view the marketplace, are the DS2-championed UPA powerline technology and the Panasonic-developed HD-PLC approach. Do you agree, or are there other powerline schemes that I've overlooked? How do you position yourself, as a technology and as a company creating products based on that technology, against your competitors? And how do you see the competitive market both today (worldwide and geography-specific) and as it will evolve over time in the future?
Yes, you are correct in terms of PLC competitors outside of the HomePlug Alliance. According to a 2007 In-Stat report, "Powerline Home Networking 2007 Update: Gaining Power in the Global Market" (page 64, Table 12), the amount of HomePlug powerline equipment shipped in 2007 was over twice that of non-HomePlug equipment. In-Stat's projections through 2011 show double the growth rate for HomePlug versus non-HomePlug PLC equipment. These data are consistent with the feedback we are receiving from our customers, who tell us that Intellon's HomePlug AV solutions are more robust and reliable in real-world use cases than any of the other PLC technologies. Our customers recognize the wealth of knowledge and experience that we bring to their solutions.
Currently, as I understand it (please confirm), multiple coincidently operating "200-Mbps" powerline technologies will actually degrade each other—far from coexisting (or, ideally, interoperating). However, there's hope, at least on a conceptual basis: I'm aware of some interesting coexistence and interoperability work going on in the IEEE specifically involving HomePlug AV and HD-PLC. Please describe the activities, their goal, and an anticipated timeframe. What cost and performance impacts will there be to coexistence and/or interoperability? And what will it take (on the part of HomePlug, HD-PLC, the IEEE, and DS2) for UPA to be included in this effort?
As you're likely aware, the IEEE P1901 Work Group is developing worldwide standards for powerline networks, covering both MAC and PHY specifications. The IEEE efforts are focused on three specification clusters:
- access (or "to the home"),
- in-home, and
- coexistence.
There has been some confusion on the most recent voting results in the IEEE P1901 Work Group. In an effort to clarify the facts, we are offering the following information. We also encourage you to visit the IEEE P1901 Work Group website for confirmation of these results.
During the period of October 15 to 18, 2007, the P1901 Work Group conducted one round of elimination voting (down-selection) on the Access and In-Home clusters. The Work Group's approved down-selection process requires that the proposal that receives the least number of votes be removed from consideration for the next step in the standard-development process. Members can vote for more than one proposal. There were 38 voting entities present at the Boston meeting.
For the Access Cluster, the HomePlug/Panasonic merged proposal (document P1901_0337_r0) received 22 votes. The UPA/OPERA/Mitsubishi merger proposal (document P1901_0345_r0) received 17 votes. As a result, the HomePlug/Panasonic merged proposal remains as the single candidate for the confirmation vote, and the UPA/OPERA/Mitsubishi proposal was eliminated.
For the In-home Cluster, the Panasonic/HomePlug/HiSilicon merger proposal (document P1901_0339_r0) received 28 votes. The UPA proposal (document P1901_0343_r0) received 13 votes. As a result, the Panasonic/HomePlug/HiSilicon merged proposal remains as the single candidate for the confirmation vote, with the UPA proposal being eliminated.
The down-selection on the Coexistence Cluster was postponed to a future meeting.
Over the next several months, the IEEE P1901 Work Group will be refining the detailed specifications in the two remaining proposals. IEEE P1901 is a very important worldwide standard that we expect will help increase the rate of adoption of broadband powerline-networking products. It is important to note that the goal of this standardization effort is to ensure that legacy products using HomePlug AV will be interoperable with 1901-compliant devices that implement the HomePlug AV PHY.
HomePlug technology has been in the marketplace for nearly eight years and has experienced significant growth, even in the face of other proprietary technologies in the market. We believe that the work being done by the IEEE in terms of setting a standard will be beneficial for the industry and have an accelerating effect on the growth of the powerline industry as a whole. We fully support the efforts by the IEEE in terms of driving for such an industry standard. We also believe that one of the reasons why HomePlug was involved in the proposal submitted to the IEEE P1901 work group (and is part of the one proposal left standing) is due to its performance advantages and significant market adoption to date.
The number of operators and equipment manufactures that have adopted HomePlug as the powerline connectivity technology of choice is a testament to its strong innovative engineering pedigree arising from the collaboration of multiple silicon and equipment providers in the industry. We fully expect this engineering innovation to continue as next generation HomePlug technology is engineered and brought to market in the years ahead.
The current "let the market decide" powerline landscape is not exactly consumer-friendly: multiple incompatible "200 Mbps" products sit side-by-side on retailers' shelves, and some networking vendors even offer incompatible powerline technologies within the same product line! How much consumer confusion currently exists, and how are you working with your customers (the networking vendors, along with their customers, the retailers) to minimize it...considering that support calls, product returns and long-term brand backlash are costly expenses for all of you to bear? Until either a standards body such as the IEEE mandates interoperability, or might-makes-right market pressures cull out technology alternatives, won't consumer frustration inevitably grow as the powerline-networking market grows? And won't this incompatibility frustration put an unfortunate "cap" on the powerline-networking market's growth potential?
We are also concerned about the reputation of PLC in general, as unwary consumers buy less-capable PLC products. The HomePlug Alliance has therefore developed a logo that manufacturers can put on their boxes as a "safety" signal to the consumer. Like many well-known brands of products, the HomePlug logo is a symbol of quality that helps consumers make the best choice. Consumers are also savvy to online product reviews and often read them before making purchases.
FUTURE DIRECTIONS
What, in your mind, are current shortcomings of HomePlug technology that the standards body (and/or proprietary HomePlug-based enhancements) plan to address in the future, and what are the timeframes for these enhancements?
Currently, we believe that a 200-Mbps standard is enough to meet today's needs, along with the most visible emerging market needs. The standard itself is quite solid after recent improvements made to enhance ease-of-use, such as the pushbutton-encryption feature. Beyond the standard are customizations that Intellon has made to meet specific market requirements. Some, like pushbutton, become part of the HomePlug AV standard while others will always remain Intellon IP. This approach is how we bring a great deal of added value to our customers while still complying with the standard.
At least one competitor publicly demonstrated next-generation "400-Mbps" performance-boost technology last fall. How much reality do you believe there is behind the claims, and how do you plan to respond to them?
Looking forward, there is much talk about a 400-Mbps PLC solution. One of our partners, devolo, has just demonstrated the world's first 400-Mbps HomePlug PLC proof-of-concept at CeBIT in Hanover, Germany. This innovation is fully HomePlug AV backward-compatible and could conceivably be ready for marketing in the second half of 2009.
As you know, I'm working on an ongoing basis on a home-automation project, the first fruits of which were published as a design feature (with online addendums) in EDN last fall (see "Homeland security: monitoring and manipulating remote residences"). Current powerline home-control technologies such as X10 and INSTEON have numerous shortcomings, I've discovered. The recently ratified HomePlug Command and Control 1.0 specification is therefore of great interest to me (especially since I already have an operational HomePlug AV data network in the residence I'm attempting to automate). Please describe the HomePlug Powerline Alliance's work here in more detail, including anticipated product availability timeframes.
Command and Control (C&C) will operate in the HF spectrum, well above the 120-kHz frequency of X10 and other home-automation approaches. HomePlug C&C is much faster and more robust than the old technologies and is not affected by cross-phasing in homes.
The ability to plug a piece of equipment into an AC outlet, as a consumer has been doing for decades, and have it immediately be network-accessible is the fundamental scenario that's always piqued my interest in powerline networking. Currently, powerline transceivers locate external to system power supplies, but AMD and Intel have both demonstrated systems containing powerline networking-cognizant power supplies. When will PSUs with integrated powerline networking support be widely available, both for personal computer and other applications? And should EDN's readers anticipate cost savings, and/or other benefits, resulting from this integration?
I cannot give you a specific timetable for this application. It does show that these well-known entities see HomePlug AV technology as reliable and viable for this use. Any integration of HomePlug AV into a product represents a savings to the consumer when the alternative cost of external solutions is considered. The benefit of powerline-networking integration is that the entire home automatically becomes an Ethernet broadband highway when the PC is plugged in. In addition, the PC is automatically connected to consumer gear that also has HomePlug AV embedded, such as ZyXEL's new DMA1100P and Monster's new Entertainment Power appliances.
HomePlug AV devices are being viewed by consumers as connectivity devices, without having to call them network devices. "Network" is often a scary word to consumers, so Intellon's engineering teams (both hardware and software) have worked hard to make the technology just-plug-it-in simple. Now our customers (OEMs and ODMs) are making our technology disappear from view by embedding it into consumer appliances. Frankly, there is no value in the customer seeing the technology. They just want it to work, and it does.
WRAPUP
Thanks for your time. In closing, what topics haven't we yet covered in the above questions, that you'd like to briefly comment on before we wrap up?
I just want to emphasize that Intellon is far more than just a chip designer. Intellon is a solutions design house. That description includes ICs, firmware, and reference designs. With a long track record of experience and innovation that has led to today's HomePlug standards and standard enhancements, Intellon brings much added value to its customers with further enhancements and innovation. While HomePlug AV (for example) is an open standard and there are competitors within the HomePlug camp coming online, there is no substitute for this kind of hardware/firmware IP and customer interaction over time.
Thanks, Brian, for this time of informative interaction with your readers.


