Does W-CDMA use PN offsets?

Yes. It's still CDMA.

 

So, both the good and bad of CDMA will come to the new GSM. Right?

 

With W-CDMA you get the good and bad of CDMA and then the bad of not having all of the CDMA Qualcomm patents.

W-CDMA is the step down from CDMA, that is why the future is really HSDPA in the US and in Europe, China etc..

HSDPA finally make the UMTS system really become 3G.

 

I know alot of people who would disagree with that statement.

 

..........

 

So here is a question. GSM networks were set up not having to worry about Pilot Pollution and were basically set up in a way that's best for GSM, now where you have to suddenly worry about things like Pilot Pollution, etc. have some GSM carriers run into some problems? I assume that there has to be EXTENSIVE testing/tweaking of the network to run a CDMA technology-like network smoothly over what used to be cell sites optimized for GSM.

 

Yeah I would be one of them but they all have there bad and good.

 

I was torn over whether to dive into these questions in this space, as my answer could be a little like your trying to get a drink of water from a fire hydrant.

Original poster > Does W-CDMA use PN offsets?

Garitaar > No. Offsets are a way to create orthogonality with fewer codes. Orthogonality is all that really matters, though the original Qualcomm CDMA solution is exquisitely elegant from a mathematical perspective, and enabled using CPUs with lower MIPS.

TKR > Is pilot pollution a "feature" of W-CDMA as it is with traditional CDMA?

Garitaar > Yes, it is a limitation of the quantity of sectors that can share support of the mobile. This is commonly called the “soft handoff” which is a bit of a misnomer, because it suggests that this is a change of state, but it is actually not transitional. A call can easily begin in soft handoff and remain with the same multi-cell relationships for the entire call. Well-designed CDMA and WCDMA systems will far outperform a well-designed GSM system from the dropped calls perspective for the same reason you tell your daughter to carry the glass with two hands instead of one. TDMA/GSM handoffs are a leap of faith.

Jerro > both the good and bad of CDMA will come to the new GSM…

Garitaar > No Sir. WCDMA has nothing to do with GSM. WCDMA is not the new GSM. Muscle does not turn into fat as they are fundamentally different. GSM cannot be improved; it must be replaced. GSM and WCDMA are not brother and sister, but are rather simply married – and we know how arbitrary that can be.

Rasputin > …W-CDMA is the step down from CDMA… [clip’d]

Garitaar > I love CDMA. I like EV-DO. I was ambivalent about WCDMA until Qualcomm got in to assist, It is quite elegant, now, I think, and does things that could have never been done without the newer CPUs replacing the low-voltage CPUs that were available when the CDMA2000 spec was written. The best thing about CDMA2000 today is that it uses small chunks of bandwidth, and this is handy for things like the implementation of the new internet connected node (base station) for homes.

Andy > …where you have to suddenly worry about things like Pilot Pollution, etc. have some GSM carriers run into some problems?

Garitaar > Of course there have been “problems” – i.e. many opportunities for employment of WCDMA optimization engineers. Not too many operators would think you could take a GSM network and overlay a completely different technology and expect that everything would work. If the operator is motivated to share the antennae among disparate technologies, however, thereby reducing flexibility on beamwidth and azimuth, the resulting network will keep lots of optimization engineers chasing their tails and well employed for years to come.

Garitaar

 

Garitaar, Wow...thank you for your very, very informative post. I really enjoyed reading it.

 

Thanks for the informative post garitaar. I've worked alot with GSM and WCDMA but not CDMA, so I'm always interested to understand that system in comparison to the systems I know.

Regarding PN offsets: I always thought CDMA PN-offsets were pretty much the same thing as WCDMA Scrambling Codes, no? As you said, they are both just forms of orthogonality. The only difference was that in CDMA the SF (Spreading Factor) was "fixed", and in WCDMA the user can take a higher SF and get higher data rates. (Is the term SF used in CDMA also, or just a WCDMA term?).

I somewhat agree. Having soft-handovers in CDMA/WCDMA does give some security, however the tradeoff is double the resources on the radio and transmission networks. And in a radio system that is very sensitive to SIR, that extra traffic can degrade the over-all quality and cause other users to suffer and maybe drop calls. So while your daughter is focusing on the glass she's carrying with both hands, she may be knocking over other peoples glasses and maybe even spilling some of her own

You're right about GSM and WCDMA having nothing in common besides a marriage certificate tho (well, a chip rate too) But in a marriage, somtimes it's good that opposites attract.:
-GSM is very robust system that's very well suited for voice traffic.
-While WCDMA is more sensitive and needs more "attention" (maybe the woman in the marriage? ) however it allows for much higher data rates than GSM could handle (Ah, the magic of a woman ;-). )

So together, the best of both worlds are possible Actually, in the future I think LTE will combine the best features of both of these technologies, so that should be interesting to watch.

 

GSM folks will get over the fear of using more resources to support a call in soft-handoff, especially when their equipment evolves to distributed networks that do not require every node-B packet to go all the way back to a common point for packet selection. GSM engineers will go through some pains learning how to design with spot-light antennae instead of flood-light antennae.

There are no upsides to GSM in any way compared to WCDMA or CDMA, and GSM operators and advocates will eventually get over their self-deception. Here in the US, an operator tried for a while to claim "fewest dropped calls" but they were made to accept the real data - both the metrics and the user reports - showing that the CDMA operators have a small fraction of their GSM drops.

The thing I most look forward to seeing go away is that periodic annoying pulsing & buzzing noise that GSM phones put on other electronic equipment. From an electromagnetic compatibility perspective in many ways, you will learn to appreciate WCDMA. Welcome to low-power, spread spectrum.

OFDMA will be good. It does not have the robustness that comes from CDMA, but there are other upsides that make it interesting.

garitaar

 

On the RAN in CDMA/WCDMA networks the SIR will always increase when a UE is in SHO. That's an inherent in the system. "Cell breathing" doesn't occur in GSM (one advantage of GSM ) For the Iu, it's stll a long time until operators use pseudo-wire techmologies for PS data en masse. Until then ATM link capacity needs to be considered, and UE's in SHO only increase the strain.

The amount of drops an operator has is more in relation to how fine-tuned their network is, rather than it's core technology. I'm sure there's a GSM network somewhere that has less drops than Sprint. But I wouldn't say "that's because GSM is a superior technology", but rather that operator implemented a better network in terms of planning and optimization.

 

If your WCDMA network sees much more than the theoretical minimum in noise-rise or breathing when in SHO, then consider a new analysis of the RF. "Physics Phyrst." GSM can deny the impact of physics to the extent of the capacity afforded by the frequency plan, but in CDMA or WCDMA, there is no hiding from ill-fitting antenna selection and site placement.

While individual operator fine-tuning is obviously the equivalent of "your mileage may vary", there is no doubt that a soft-handoff based system has the inherent ability to have fewer dropped-calls, based purely on statistical analysis. In such analysis, one compares the same system - same locale, sites, heights - and considers not only various architectures, but different vendors of the same architecture. A net present value of every site and network element can be established, as well as a theoretical best performance level based upon the algorithm.

On yet another tangent, I am not sure what you mean by “pseudo-wire”. I was rather describing an all-IP network with distributed rather than centralized controllers and packet selectors. Don’t worry too much about it this now; just know that this is where the operators will seriously improve their backhaul economics.

garitaar

 

I highly doubt that. GSM can make successful HHO's right alongside CDMA/WCDMA's SHO until the cows come home. GSM makes billions of HHO's worldwide daily without any problems. And if the network was heavily loaded, GSM may even perform better, since it's not affected by the overall noise rise like CDMA would be.

GSM's HHO isn't a "leap of faith" so much as you think it is. The network is constantly monitored and analyzed, and GSM will make the HHO at exactly the right time. It's like a sniper who waits for the right moment to pull the trigger. CDMA/WCDMA is a machine-gun that blasts everything. And when you consider bullets as limited network infrastructure, then GSM is much more efficient.

In the end CDMA and GSM are similar, in that they are both successful 2G systems. CDMA just needs more radio and transmission resources to facillitate it's SHO, has to deal with cell shrinkage from "cell breathing", and then has to pay Qualcom a royalty for that I guess it's no wonder the "open source" and more efficient GSM system is dominant all over the world

 

You can "doubt" facts if you wish. I always get a kick out of GSM folks' denial. Good luck.

 

I've worked in GSM and WCDMA, and I really even can't compare the drops or failed HO's in those 2 systems because they are apples and oranges. They work in different ways. There's no counter that says "This call dropped because it was a GSM call". GSM is designed to work fine without SHO's, while (W)CDMA requires them. Anyway, the majority of drops are usually due to something like lack of coverage or neighbor cells not defined correctly....things that can happen regardless of the system.

Anyway, "facts" are always how you percieve them . The glass can be seen as half full, or half empty. GSM has 2.5 Billion users, CDMA has 400 million. I'd see that as the GSM cup 86% full and the CDMA cup 14% full. Nazdarovia! :cheers2:

 

I found this interesting...

Given that GSM (and WCDMA/UMTS 2100) is the *only* network technology used in Europe, it seems to me that their networks are *far* more sound than US GSM networks.

All of the European GSM networks work flawlessly and dropped calls are pretty much unheard of... Coverage is excellent and the sound quality is far better than in the United States.

Any comment? (I'm curious if the European operators are doing something differently from the American ops...)

 

I concur. European GSM networks are like a day and night difference to those of the U.S. I have used GSM networks all across Europe extensively, and loved it. Here in the States, I got fed up with the voice quality and network issues on GSM networks (at least in the Western U.S.).

 

There's 2 main reasons why EU networks are better than US:

1) Common standard

EU has only ONE standard. In the 90's the EU countries agreed to adopt only one (GSM) standard, so that you could make a call in Spain and carry it to Finland without having to worry about any compatibility problems. Call it "Socialism" but it's to everyones advantage to have one common platform.

In the US the "Free Market Economy" was left to decide, so everyone and their brother could invent a standard and build a network with it (as was evident with iDen ). Call it "Freedom" but a splitering of dfferent incompatible technologies isn't good for anyone. Especially confused consumers.

2) Quality Over Quantity

In the US, operators threw up networks as cheap as they could (ie: not tuned well, minimal sites, etc.) so they could sell consumers the cheapest price packages. In the EU operators focused on quality. Maybe it's a difference in culture. But EU consumers paid for that. In the US you could get packages for like $30/month with 500 minutes (or something like that) but you'd spend half your call saying the famous line "Can you hear me now?"..while in EU you're paying $1 per minute (or something like that) but your call is crystal clear.

Now prices in EU are coming down, as operators have well-built networks and don't invest too much more in infrastructure, so altho customers paid heavily early on, they left a legacy of good networks behind them. While in the US prices remain the same, or maybe go up a little as operators are trying to "clean up the mess" of their network that was shoddily rolled out ~5 years ago.

You could also say the EU is physically denser, and doesn't have the wide open areas the US does, but even when visiting NYC/LI area (very dense area) reception is still shoddy and there's no coverage in some major tunnels (at least it was like that 1-2 years ago).

The GSM vs. CDMA debate is kind of childish (but I'll play along sometimes ). I do favor GSM for voice and WCDMA for data, but really all technologies have ups and downs. It really depends more on how well the operator builds their network. ie: if they are willing to invest the time and $ to build a good one or not.

 

Just to add my tuppence worth to the above post, it might also be a factor that networks (in the UK at least) were given targets they had to reach as part of their license for WCDMA i.e. they had to cover a certain percentage of the population by a certain time. If you combine that with the high population density and high usage of mobiles then the networks could be expected to build networks relatively quickly and well because they not only had to provide service to a certain amount of the population in a certain time, but they also had to make sure the service they were providing could cope with relatively high usage.

I find the comments on 'socialism' and 'freedom' quite interesting. Some Americans seem to like the idea of choice even when it is a case of choice for the sake of choice rather than meaningful choice. By all means Americans have more choice than Europeans when it comes to mobile standards; they can choose GSM, WCDMA, CDMA, EV-DO or iDen, but are those meaningful choices? How many American consumers consider the network standard when choosing a new mobile phone? Do most, or even many, consumers compare the differences in standards when buying?

In Europe you don't have much choice of mobile standard, it's either GSM or WCDMA (and with WCDMA you get GSM too of course), but doe that have a negative impact upon consumers or the market? Given the state of the European mobile market I would say not; Europe is, in some ways, a more competitive market for mobiles and mobile services than the U.S. Consumers generally don't care about the technical specifications of the underlying network, rather they care about getting the service they want for a good price.

I feel the U.S. would benefit from a single standard that would allow the U.S. to present a unified market to foreign networks and manufacturers seeking to do business there.