POST OF THE DAY
Qualcomm
FLO Technologies: Beyond 3G CDMA

Format for Printing

Format for printing

Request Reprints

Reuse/Reprint

By BRational
June 8, 2005

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

Several interesting white papers (and assorted brochures and data sheets) have appeared in the past month on the MediaFLO corner of Qualcomm's web site. I was particularly interested in the technology-oriented white papers (see references and links below), though there is also a good concept overview white paper that provides the rationale from a user perspective of the underlying technology, and also a brief report on "data rights management", meaning security and preventing illegal copying and distribution.

I had not paid too much attention previously to the fact that MediaFLO is essentially an entirely new wireless transmission technology� I had assumed it was some form of optimized-for-multicasting variant of EV-DO, but it is in many ways a post-3G technology� do we dare say 4G?

The basic idea is quite simple� one can deliver essentially the same content to many different handsets much more efficiently than delivering entirely different content to those same handsets. You would only need to carry it once through the network instead of carrying N different copies of it. What makes multicast different than, say, broadcast, is that the latter would need to be received identically by all at any given time, whereas multicast allows customizing the bundle delivered to each handset from the available selection of "common" items, and possibly combining with purely "local" content (that a local wireless carrier might supply).

What makes the delivery of this common content efficient is that it goes only one way, in a so-called Forward Link Only (hence the FLO acronym) fashion. This means you can only download content, but cannot send anything, at least not through the FLO network. Interactivity is envisioned to be provided in combination with the existing 3G wide area network (operated by the wireless carrier).

This forward-link-only network is envisioned built entirely from scratch. There is a good section on "network economics" in the Technology Brief white paper. To understand it, it is useful to understand the air interface technology, and the trade-offs it entails.

To achieve its efficiency, FLO uses OFDM; it does NOT use CDMA to transmit signals. The second technology white paper has more about OFDM principles. In many ways, it uses ideas from CDMA, in that it chops up data packets and reassembles them at the destination, but does so more efficiently than CDMA because it adds a so-called "guard" to each data packet that provides the orthogonality, meaning it essentially eliminates interference from other signals. While not directly addressed in the reports, nobody appears to own the rights to OFDM�the knowledge is essentially in the open scientific domain.

To get additional capacity out a given frequency band, FLO uses Time Division Multiplexing (TDM). Specifically, it divides the available band into three separate channels, with shoulders on the side. Think of it as dividing a freeway into three lanes, with shoulders on both sides. In each lane, traffic is run at very high capacity using OFDM. In many ways, the concept is reminiscent of TD-SCDMA, whereby the total band is subdivided into a number of channels, just like TDMA (and GSM do), but then what flows in each lane uses Synchronous CDMA. With FLO, we have a TDM division into "permanent" lanes, with each lane delivering content using OFDM protocols. MediaFLO is envisioned running within a 6MHz-wide band, though one could use narrower bands as well.

Why not use CDMA? Simply put, for this type of application, OFDM is more spectrally efficient. It does not have to negotiate sharing the right of way with uplink traffic, and does not have to deal with many different types of unique packets going to different places. Interestingly, Toshiba is promoting a proprietary mobile multicast multimedia standard called S-DMB that uses CDMA. The technology overview white paper contains towards the end a comparison between FLO and other competing technologies. I found it very helpful, though I'm sure the competition will argue that Qualcomm may not be exactly disinterested and fair in this comparison. The technologies considered are (Table 2, page 14, of the technology report white paper):

Format Description
ISDB-T
Origin: DTV packet data technology (Japan)
Modulation/Coding: OFDM, convolutional, Reed-Solomon

T-DMB
Origin: Derivative from European DAB, modified for multimedia (Korea)
Modulation/Coding: OFDM, convolutional, Reed-Solomon

S-DMB
Origin: Proprietary format, primarily from Toshiba (Japan)
Modulation/Coding: CDM, convolutional, Reed-Solomon

DVB-H
Origin: Derivative from DVB-T (Europe)
Modulation/Coding: OFDM, convolutional, Reed-Solomon

FLO
Origin: QUALCOMM packet data technology (USA)
Modulation/Coding: OFDM, turbo, Reed-Solomon

Many here may have read Eric (InCards') post on Tinker's Board questioning Qualcomm's attempt to promote its own standard when Nokia has decided to push for its own DVB-H. Apparently the basic underlying technologies are quite similar, though Qualcomm claims that its implementation is superior and delivers higher quality of service from a user perspective. Tables 3 and 4 on page 15 in the white paper present technical performance parameters and user experience parameters, respectively, for the five competing technologies. No prizes for guessing which one comes out dominating the others. Could more holy wars be far behind?

The network economics discussion explains that the spectral efficiency is such that coverage for a metropolitan area could be achieved with as few as three cell towers, assuming operation in a 700 MHz spectrum. Operating in the low 700's turns out to be near-optimum for this technology. Table 1 on page 13 in the technology overview report lists the typical coverage areas associated with different frequency bands, showing the rapid degradation as one moves to higher frequencies. This explains why Qualcomm is lobbying to auction the 700 MHz spectrum (See article linked by Slacker on the SI Moderated thread.)

This is only a brief synopsis of my readings in the MediaFLO corner. I found the white papers to be quite readable, reasonably well balanced and generally free from blatant hype. It also made me realize that this is indeed a real notion that is moving along towards deployment. It also made me realize that we're not nearly there yet, and won't be for quite some time. Given the "from scratch" nature of the technology, there is still a need to make the chipsets for both base stations and handsets. If the latter need to also be able to speak 3G in at least two jargons in order to have an uplink capability, then the challenges for the chipset and handset designers seem non-trivial. I also suspect getting the interfacing right with the local carrier services will require considerable field engineering talent.

While I am yet to be convinced about the value of mobile small-screen based TV as a business proposition, I find the rather innovative technological aspects and the possibility of building out entire networks of post-3G technologies to be certainly intriguing propositions.

Comments, corrections and thoughts would be appreciated.

BR


Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.