Jim:
Permit me to reiterate Bernard's point:
...3 - Conceding the point on VOFDM and rain, it was uniformly expressed that snow, sleet, and sub-zero temperatures were bound to degrade signal quality, and therefore, error-free throughput, in VOFDM.
Below 5 Ghz, signal degradation from rain and atmospheric gasses are almost irrelevant. Above 5 GHz, degradation to signal quality is quantifiable and predictable.
The foregoing is true for any RF transmission technique, i.e., FDMA, CDMA, TDMA, OFDM, COFDM, etc, etc..
As a general rule, no RF transmission technique mitigates the effect of precipitation or atmospheric gasses better than another.
Yet, having conceded that precipitation and atmospheric gasses do impair signal quality, the impairment can be almost totally eliminated as an operational concern by proper RF system design. Anything said to the contrary (Wall Street analysts, reporters and venture capitalists are good examples) is just plain FUD.
...1 - Agreed that VOFDM can overcome line-of-sight (LOS) limitations in a best-case scenario.
2 - It is believed by some that despite that fact, the LOS limitations can only be effectively overcome, in less than ideal situations, by the use of repeater stations, as in a cell network....
I think there is a mixing between applies and oranges here again.
1. Cisco's stated target market for VOFDM is fixed market.
2. To the best of our understanding, VOFDM uses multiple transmitting and receiving antennas.
3. The combination of OFDM modulation with multiple transmit receive antennas (the "V" part for vector) permits adequate and predictable signal strength in less than pure LOS situations.
4. Multiple transmitting/receive antennas are - at this moment in time - impractical for handheld devices typically used by subscribers of mobile networks (handsets, PDAs etc).
The actual received RF signal strength versus the predicted received RF signal strength is a function of frequency, space and time. With that one sentence I've probably lost most people on the thread; I get confused all the time <g>
I sat for a number of hours trying to come up with a pithy way to explain line-of-sight (LOS) issues, and couldn't do it. It depends if a path is mobile or fixed; it depends on frequency; it depends on the air interface (the alphabet soup above), it depends on the environment; concrete canyons (ala Manhatten), surburban or rural environements; it depends on ground cover - desert (Phoenix, Las Vegas) lots of water (lakes, rivers, etc), plains, deciduous tress versus coniferous trees, it depends on time (changes in the earth atmosphere changes how an RF signal propagates.).
Let's put it this way. RF systems designed for mobiles presume that LOS will not be available at every point on the earth from which the user desires to communicate. Consequently, mobile systems have used frequencies where multipath (primarily reflective multipath) acts to increase the overall probability that an acceptable signal can be obtained, and that signal impairments due to multipath conditions are significant design issues but the effects are fleeting due to the "in motion" characteristic of the user. One of Qualcomm's innovations was to make multipath into a patented feature, i.e., rake receivers.
In contrast, because a fixed system is fixed, the complexity of predicting the strength and reliability of the RF signal is reduced greatly. One gradually moves away from statistical probabilities to addition and subtraction exercises on a link budget. Any frequency can and has been used in fixed systems. Traditionally, multipath is considered an impairment (atmospheric multipath and reflective multipath) to be mitigated since fixed systems are generally constructed with LOS between transmitter and receiver. However, as Adaptive mentioned in its paper, ...Given a fixed location scenario, design engineers have very effective, significantly less-expensive options for handling multipath fade and non-LOS locations...
I don't now where to go with this response except to say - one can't mix apples and oranges. Would, I likely choose CDMA if I was constructing a new narrowband mobile network - Yes. Would I use CDMA for a new wireless local loop network - probably no. Does it appear Wi-LAN has some great technology for campus-type LANS - Yes. Would I choose it for an LMDS system - No. Would I choose any of these for a pure point-to-point high capacity link, where the path parameters are under my control or can be predicted with a high degree of confidence - No. Would I choose COFDM over 8VSB - at VHF and UHF frequencies - in a heartbeat.
I could go on - but I'm falling asleep <g>. The why to each of the statements above would likely require a chapter length tome. It is generally safe to avoid the use of the phrase "best technology" when dealing with RF propagation issues.
ww |
