Search This Blog

Thursday, April 27, 2006

Broadband-IP



What is ‘Broadband’?

The definition of what ‘Broadband IP’ constitutes is fairly open to interpretation. While there are numerical definitions of what a ‘Broadband’ connection is, these numbers vary depending on the operator (and what it wants to advertise) and the nature of the service. As an example, if you are in an IM text chat session with your friend and are not exchanging rich media, does it really matter if you are on a 44kbps link or a 3mpbs link ? However, if you were in a Skype audio/video chat with a colleague, you will certainly feel the difference between a 44kbps link (choppy voice/video) and a 3mpbs link (smooth video and audio) . Therefore, the definition of ‘Broadband’ over IP is closely tied to the nature of the service that is to be delivered.
Having said that, for the scope of this paper, let us assune that broadband is any data service that offers a data rate of higher than 56kbps. Practically, most wireline broadband providers start at a minimal ‘broadband’ rate of 128kbs/256kbps. While early versions of second generation wireless technologies such 1xRTT and 1xEVDO offer lower data rates, it is a safe assumption that a majority of ‘broadband’ usage is above the 128kbps data rate.

Growth of Broadband

The success of a service depends largely on two factors: a) The customer need for the service and b) the quality of the service. In some cases, the first factor is so important that users are willing to accept degradation of quality (the 2nd factor). The proliferation of cell-phones worldwide, with dropped calls and jittery voice is a good example of such a service where the uniqueness (mobility) of offering overshadowed the quality concerns. Having said that, cell phone coverages have significantly evolved over the years, since eventually, consumers will expect ‘acceptable’ quality. Therefore, it is important to analyze the growth of Broadband before we can discuss the relevance of the services on top of it. WebsiteOptimization has some interesting data on Broadband Growth, where they show that in the USA, in 2000, the total % of ‘dial-up’ users within the ‘internet connected population’ was 74%, while in 2006, this rate dropped down to 36% (in other words broadband users are now 64% of the total internet population). In terms of the proliferation of Internet in countries, InternetWorldStats reports that the number of Internet users in the USA is currently at 68.6% of the US population while world wide, the Internet users are at 16% of the total world population, up from 5% in 2000.
The relevance of Broadband penetration is a function of the internet penetration of the region in question as well as the market segment within that region. As an example, Asia’s overall internet usage growth since 2000 is over 210%, EU is 180%, North America is 110% (source)
In short, this tells us three things:
  • It is obvious that different regions will have different growth numbers for internet usage (a function of the economy amongst other factors). However, each region has a very high grown rate which implies that the rate at which internet penetration is increasing is very high.
  • As the internet penetration grows, it is expected that the % of broadband users within the Internet users will also substantially grow (as shown above by individual statistics in the US as one region example) due to the fact that broadband rates are driving down and there are new innovations for delivering broadband to the last mile (more on this later)
  • Finally, from an innovative service deployment perspective, it also tells us that the penetration of broadband usage within the healthy internet growth worldwide means the ‘last mile access’ (from the provider to the user) is capable of delivering media rich services. After all, even if the network is connected via high-speed fibers, if the connection to the user is bandwidth restricted, then this severely limits the usability of the internet for high speed data services from a user perspective.
Factors that changed the ‘Communication Landscape’ as we know it


The rapid growth and related excitement about leveraging the Internet for communications can be broadly attributed to some key technologies and market shifts in the past decade:

  • The penetration of the Internet around the world – discussed in detail in previous sections.
  • Voice over IP and Vonage – In the 1990s, using the Internet to deliver voice calls was a much talked about topic. (actually, voice over IP as concept is older and there were several proprietary protocols that dealt with specific requirements, but it took till the late 90’s for standardization momentum to pick up). Competing protocols such as H.323 and SIP were introduced, both of which promised to utlize the internet to deliver global voice calling around the world. Using the internet, which is a de-regulated and distributed packet network to be able to deliver voice calls was exciting to everybody except the incumbent carriers because it allowed them to play in the communications space without heavy investment in infrastructure. Detecting this trend, incumbent carriers also reduced calling rates while trialing VoIP themselves. However, no one really was ready to deploy VoIP for consumers (there were several enterprise VoIP deployments, where QoS is a lesser issue). This was till Vonage took a leap forward and introduced VoIP calling for the masses. For a low monthly rate, they offered a bundle of enhanced services which were charged at a premium from LECs. This effort had a worldwide ripple effect on two fronts. First, consumers and providers realized VoIP as a technology had the potential to work well and threaten to replace PSTN lines (admittedly, there are still issues of emergency calling and overall consistent quality issues which are being addressed.) and secondly, it was clear message to the LECs that this was a disruptive technology that could potentially displace them from a multi-billion dollar market.
  • 3GPP – While Vonage and a slew of ‘me too’ providers penetrated the wireline world as a viable alternative of Broadband IP being used to replace PSTN lines, there was still a large void in the ‘mobile world’. The challenges of a mobile network are much more than in fixed networks (QoS, Roaming agreements, location tracking, Emergency calling, latency, air interface optimization and more). From a technology perspective it was fairly obvious that it was only a matter of time that Broadband IP somehow penetrated into the mobile world and threatened the monopoly of wireless carriers. However, before that happened there needed to be significant investment in preparing the wireless network for the IP infusion. 3GPP took on the responsibility of defining an all IP architecture (in stages) which would provide carriers a migration path from switched to packet based networks and thereby ‘stay in the game’ and be able to retain their customer base as well as fight the challenges posed by alternate technologies that threatened to bypass their network (next point). Interestingly, 3G licenses were put on bid for astronomically high costs for carriers. Once those costs were sunk in, unfortunately, the worldwide market dipped significantly
    during 2001-2003 and revived up again end 2004. Once the market was up again, carriers had already invested too much into 3G licences and it was (and is) in their interest to make it a reality soon. In 2006, there are already several 3GPP trials in the market and we expect to see all-IP 3G deployments by Q2 2007.
  • Distruptive technologies: WiFi, Skype, Peer2Peer and others: While all of this happened, several technology and market innovations took place which drove consumers faster towards ‘using the Internet for communication needs’ than ever before. The first was WiFi – the industry rapidly made progress in WiFi and related technologies where the user could access a high bandwidth internet connection with limited mobility. The mobility range continued to improve with new mobility standards such as 802.11e and beyond. In addition, recognizing the fact that new standards take years to deploy and stabilize, innovations by companies such as Tropos, Cisco and others resulted in new network topologies that increased the range of existing WiFi standards by creating a mesh of interworking routers hosted on city lamp-posts to provide a larger mobility range to users. At the same time, services such as Skype were introduced which provided excellent audio/video quality and easy connectivity with the PSTN which further drove consumers to adopt broadband IP. Finally, solutions such as Skype demonstrated that the Internet was ready for providing VoIP communications in a peer2peer fashion – you don’t need to sign up with some ‘central provider’ to reach another user (as long as the other user is also on the Internet) – the Internet was the ‘provider’.
  • Google – Finally, this has to be said: Google happened. This was not just a demonstration of a ‘search engine’ company. It was the ushering of a new era of communication where google demonstrated how different streams of content – voice, video, data could be effectively grouped together and personalized in such a way that the Internet suddenly became a respository of ‘useful information’ for each individual instead of ‘Raw Data’ using a single interface. In addition, they also demonstrated en-masse how technologies such as AJAX could be used to present attractive User Interfaces which compete with the effectiveness of local desktop applications.

Finally, what are the applications of Broadband-IP ?


We talked about infrastructure & broadband penetration. We talked about disruptive technologies. So how do we tie this end to end ?


Imagine that you are on your desktop PC chatting with your friend. In addition to chatting, you are also sharing photographs from your flickr account for a recent trip you and your friend took. Your friend and you are looking at the photographs together as well as collaboratively editing the photos by adding notes about what you were doing in various photos. Ofcourse, your ‘chat’ is using ‘voice’ (talking), ‘video’ (you have a web-cam) as well as ‘data’ (sharing textual notes). While conversing, you decide to drive down to your friend’s house, so you click on a button and the ‘session’ automatically gets transferred to your cell phone, using VoIP. However, your current network operator does not provide you with enough bandwidth to host a video call, so the session automatically removes the video session when the call transfers to your cell phone. At the same time, you use google maps in your cell phone to map the exact driving directions to your friends house. Your friend in turn can track your current location from home, because you have allowed him to view your location updates. That way, if you are lost, your friend knows where you are without you having to figure out !


Ofcourse, the service could go on an on with the potential combinations that could enhance the definition of what we know communication to be today.
So really, as an answer to the applications of Broadband-IP, the scope is enormous, almost infinite, limited to one’s imagination and fair business reasoning of which service makes sense to deploy and which service would a user be willing to pay for as opposed to being a free service with alternate revenue mechanisms (advertising, as an example).


Crystal Ball

So what’s the next wave ? Well, really, 3G and Fixed Mobile Convergence are too new not to be considered the next wave. But really, beyond that, another area I see that will contribute significantly to the ‘end-end Broadband IP’ dream is that of Home Networking. In 2001, I did quite a bit of work with Telcordia, Columbia University and others to define extensions to SIP for home networking. Unfortunately, the economy tanked then and the timing was just not right. But I am betting this field will be the next to mesh into a common protocol infrastructure in this decade. We shall see.

1 comment:

  1. This topic could reach thousands pages :) Nice, read it all.

    ReplyDelete