Here's what NEC said about their WCDMA
americasnetwork.com
NEC’s road to 3G
The telecommunications industry is focused squarely on Tokyo and the imminent launch of the world’s first 3G network by NTT DoCoMo in May
By Tony Chan
IMT-2000 will offer more bandwidth, richer user experiences, and bring in a new range of services but, as much as anything else, 3G is a tremendous opportunity for vendors. For those that missed out on the GSM boom, 3G is a chance to catch up on their European rivals. Chief in this category are the Japanese suppliers and operators.
3G:america’s Tony Chan paid a visit to NEC headquarters to talk with Toshiyuki Takenaka, associate senior vice president and executive general manager of NEC’s Mobile/Wireless business unit and Takakuni Kuki, general manager of NEC’s mobile division (infrastructure).
3G:america: Which part of DoCoMo’s 3G network are you supplying?
Toshiyuki Takenaka: The base station controllers and the core network are all supplied by our division.
Takakuni Kuki: For the antennas, DoCoMo has chosen another company. But the key parts of DoCoMo’s 3G network is being supplied by NEC.
Are all these components ready?
TK: Yes. For example, the core network, the radio part, the radio network controller, we have already delivered in the middle of this year. The rest of equipment was delivered at the end of August. We are now helping DoCoMo install the network. Of course, most of the parts are already installed and they are now making test of the equipment and network.
From your experience in supplying the 3G network to DoCoMo, what is the main difference between 2G and 3G? Do you need more base stations?
TK: In that sense, the number of base stations are about the same, but the traffic capacity itself is much larger.
How much larger?
TK: Everyone asks that, but it is not easy to answer. The radio transmission network is improved, in that sense, a couple of times (compared to 2G networks). The main expected traffic will be packet transmission, in that case, many subscribers can connect to the network, but actually they share the network. Transmission is intelligent, so in that sense, a huge number of subscribers can connect at the same time to the network. That is quite different from the 2G networks.
With packets, are you talking about data traffic, or voice as well?
TK: At the beginning, voice will be treated as circuit-switched traffic.
NEC has not been a big player in second-generation GSM and CDMA. Is 3G a big opportunity?
TT: For GSM, your are right. But for CDMA, we are not the largest, but a major supplier. We have supplied networks to operators in Brazil and Russia. 3G is of course a big opportunity for us, as it is for everybody. For example, in Japan, we have been supplying PDC networks, but that is only for Japan. With 3G, it is for the worldwide market so it is a major business opportunity.
Big opportunity, but also many competitors. How will NEC compete?
TK: NEC has a long experience in narrowband CDMA. 3G is based on CDMA technology, so that is an advantage for us. In the second generation, TDMA-based technology was the most popular, but CDMA and TDMA are quite different technologies. For example, GSM has eight time slots, and it is almost impossible to add a ninth time slot. In the CDMA case, for example, 10 conversations are being accomodated by the infrastructure, the eleventh one is easy to put on. But if you put too many, the quality of the transmission degrades, therefore it is very different the limit of the capacity for CDMA. It depends on the traffic condition, the propregation condition and so on. So, in order to achieve the optimum capacity for CDMA, long years of experience is needed, which NEC has. This is our main advantage.
DoCoMo has many suppliers for the 3G network, do you have to work closely with them?
TK: Yes, but DoCoMo has chosen many suppliers, but not all of them have delivered the equipment, only a few. Those that have delivered their products, we are working with them, but for those that have yet to deliver any equipment, there is no way to work with them.
What has been the hardest part in designing a 3G system?
TK: To make it more stable, in terms of software. Of course, the network we supplied to DoCoMo has been tested, but only in very limited conditions, for example, only a few subscribers communicating with each other. But with a 100, 1,000, or 100,000 users, there could be some hidden bugs. Now, we are in the process of debugging the system. It is not difficult, it is normal precedure.
Are there any major differences between a W-CDMA network and a second-generation PDC or CDMA network?
TK: Of course, there are some differences in the core network. As we pointed out, NEC has few experience in the GSM market and the W-CDMA network is based on the GSM core network, so we have had to develop new software for that. For PDC, we have mature software already inside the infrastructure, now we have to figure out the GSM evolved network. Our development is still in the early stage, that’s where we have a lot to work on, to make it mature. There is a lot of software from the ANSI-41 network developed for use with IS-95 CDMA networks which cannot be reused for the GSM MAP network used by W-CDMA.
Do you have other 3G contracts besides the one with DoCoMo?
TT: We have signed agreements with some operators for trials of our 3G network, pre-commercial trials. Officially, that is not a contract. It is not a binding agreement, they can still interview another supplier for the supply contract.
How many trials of 3G networks does NEC have?
TT: We have trials in the UK, Germany, Singapore, Korea and China.
Can you describe a trial network?
TK: It depends on the market. In the Telecom Italia case, it was a complete network from the core network to the radio interface, but in some countries, it is just the radio technology with just the base stations and radio interface.
How are you competing against the major suppliers of second-generation GSM networks?
TK: In the beginning, the Nordic companies, we were not really competing. Especially Nokia, their company strategy was to keep the GPRS infrastructure longer, so 3G is just a way to offer international coverage or more capacity into the network. At the time, NEC was promoting 3G as a complete all-in-one network solution, not just an extension of GSM. With the auctions in Europe, there are now new operator[s] in each market. They will have to build out their 3G networks earlier. It seems that every company [equipment vendors] have changed their position and is now promoting 3G as a whole system. Now, we are promoting 3G together and competing against each other.
Since much of the GSM market is already controlled by two or three companies, do you see any challenges in selling a 3G system into the market since most of the operators are already customers of one of your competitors?
TK: It is difficult, because a lot of the existing equipment can be upgraded to 3G by adding new components or products into it, at least that is what the promotional material says. They are right, that’s why we have an alliance with Siemens, which have been supplying core networks to GSM operators. They are not the strongest supplier, but they are also working with the TDD-version of CDMA.
You mean China’s TD-SCDMA? What do you think of it?
TK: Yes. Right now, the planning state of TDD version of CDMA is not clear. In Japan, the license for TD-SCDMA is not there, only FDD spectrum is applied. In China, they are waiting for the majority of TD-SCDMA spectrum, so licenses in this year cannot be expected. In Europe, FDD CDMA is already happening, so everyone is focused on that. TDD CDMA deployment in Europe is very difficult to picture.
Do you see TD-SCDMA as good enough to be just like W-CDMA?
TK: TDD and FDD versions of CDMA are not competing technologies, they are complementary. For wide coverage, FDD is the cheaper solution, but for indoor coverage with high data traffic, TDD is the cheaper solution.
Is it easy to make the two technologies work together since they have different carrier sizes?
TK: With dual-mode handsets; yes. TDD and FDD use different frequencies. They are very close, but different. Dual-mode handset is needed to accomodate that, but can be done. A GSM/W-CDMA dual-mode handset is not easy, but it will be done. TDD/FDD dual-mode is easier.
In terms of pricing, is a W-CDMA network more expensive than a 2G network?
TK: For voice application, it is almost the same. But of course, 3G is not just for voice. So, if you expect heavy data traffic, then more investment may be required. At this moment, no one knows the distribution of the traffic.
What about some of the services that are talked about for 3G?
TK: There are many demonstrations of video telephones because it is easy to understand and easy to show, but nobody believes that that will be the major application for 3G. Actually, e-commerce or such kind will be the major applications for 3G, not really video telephony.
So, people will be buying things with their 3G phones?
TK: Yes. But if people want to buy something from their phones, of course, they will also want to see the product as well. So, in that case, video or pictures should be downloaded. That is why a picture will be transmitted. That will take up the majority of picture transmissions on 3G networks, not video telephony. Some people will like video telephony and use it, but it will not be the majority of the traffic.
Right now, 3G seems to be a very separate environment. There is still a circuit-switched network for voice and a packet-switched network for data, will they be integrated in the future?
TK: Even from the beginning, they will be integrated. That is the difference between Europe and Japan. In Europe, most of the operators will already have GSM switches, and they add, for example, GPRS, to get packet-switching. When they move to 3G, they will still be using the GSM core network, so the network is still separate. In Japan, the second-generation network is PDC so there is no GSM core network and the system cannot be modified to support 3G. So, we need to introduce a new core network. We can then choose a better implementation of the core network with only one ATM switch which handles both circuit and packet traffic.
So, it is a cheaper solution than upgrading an existing network?
TK: If most of the traffic is circuit-switch, or most of the traffic is packet-switch, this is not a cheap solution. But if they are changing day-by-day, or increasing month-by-month, then it is a cheaper solution.
In other words, for networks upgrading from a GSM network, there is a fixed capacity for circuit-switched and a fixed capacity for packet-switched traffic?
TK: It is like a GSM/GPRS network. Circuit-switched mode is limited to the capacity of the GSM network, while packet-switched traffic is limited to the capacity of the GPRS network. What the ATM-based system offers is the total capacity of the core network for both circuit-switched and packet-switched traffic.
Is this NEC’s solution — an integrated core network? Is there anyone else doing this?
TK: Most of the Japanese vendors are proposing this solution. The European vendors can also do this, but they have not started to develop this solution.
What is your perspective on the American nature of CDMA?
TK: Qualcomm is the most famous company for CDMA. It is their concept. Their biggest contribution is to open the door. Before their proposal, nobody believed CDMA can be practical. [It] can only be used as deep space transmission for very limited applications. So, Qualcomm is the company that invented the concept, but there were many engineers and technicians who implemented the concept. NEC was one of the companies to implement CDMA. CDMA was born in the US, but actually, it is achieved in Korea and Japan.
What is NEC’s contribution to 3G?
TK: NEC was one of the companies who proposed a technology for 3G. Of course, we proposed wideband CDMA. Narrowband CDMA is a good technology, but it is not sufficient. Using wideband, it could be a next-generation technology, so we proposed it. We first proposed it to Japan’s ARIB, and then it was submitted to the ITU. We contributed mainly to the air-interface and Europe’s ETSI contributed mainly to the core network and core network interface based on GSM technology (for W-CDMA). NEC continues to contribute to the progress. We have committed to send one chairman to the 3GPP’s specification group for Radio Access Networks.
What about Qualcomm’s IPR claims? Many vendors don’t want to pay IPR to Qualcomm.
TK: Nobody wants to pay. But we recognized their IPR. When they proposed narrowband CDMA, they already have the IPR to the fundamental technology. W-CDMA is still based on that. They will hold the essential IPR for CDMA.
Does NEC have any IPR for W-CDMA?
TK: Yes.
So, you pay less royalties to Qualcomm for 3G than 2G because you also own IPR?
TK: Less or more, I don’t know. But Qualcomm is an IPR company, they don’t need to cross-license any IPR. For example, someone like Motorola is also manufacturer, so they are interested in using NEC’s IPR. However, Qualcomm doesn’t make anything, so they are not interested.
What about cdma2000, are you developing a system for that?
TT: NEC has now developed a core network for cdma2000, but not the radio network. We think cdma2000 will get some market share, but we feel W-CDMA be much larger. Our resources are limited and right now, we are concentrating on W-CDMA. If we had spare resources, we would do cdma2000, but we don’t at the moment.
What do you like about W-CDMA that is better than cdma2000?
TK: Not necessarily better. W-CDMA is based on a GSM core network. cdma2000 is based on the ANSI-41 network. At this point, the GSM core network has reached maturity and is global. The ANSI-41 network is limited to North and South America, plus Korea. In that sense, W-CDMA is much bigger, so it is a better solution.
Are you supplying KDDI’s CDMA network and the upgrade to cdma2000?
TK: We are proposing the core network only. We have not yet started the development of radio equipment for 1X CDMA.
How do you see the global rush to 3G? In Japan, DoCoMo is running out of capacity, so they need a new network, but in other markets, they don’t even have GPRS. Do they really need another network?
TK: In the GSM world, GSM is a good system. Many years ago, that was the only 2G system, GSM, then TDMA, CDMA and then the Japanese PDC systems were invented. The US and the Japanese learned GSM technology, but found that it was too old, that is why they invented new systems. They could have used GSM, but it was too old. Even if the new packetized GPRS systems are put in place, it is still based on the old technology. So, it is good timing to go to 3G.
Do you think that GSM will go away then and that operators will be replacing their existing 2G networks on existing spectrum with a W-CDMA network?
TK: The GSM core network will stay. For the radio network, some operators with 1.8 GHz spectrum are now thinking of replacing the GSM radio network with a W-CDMA network. For operators with 800 MHz and 900 MHz spectrum, at the moment, no one is thinking of that, but in the future, if 3G is very popular, they will probably want to upgrade to W-CDMA.
Are there any difficulties in implementing a 3G system in lower spectrum?
TK: If the spectrum is clear, it is no problem. But most of the spectrum is occupied now, so migrating users from existing networks to a new W-CDMA network might be difficult.
In this sense, analog operators with 800 MHz spectrum could just go straight to a W-CDMA system?
TK: If they have the spectrum, yes.
Are you worried that so many operators have paid so much money for their spectrum that they won’t have enough money to buy a network?
TK: Definitely. If it is only their trouble, then we have no problem, but they are making it our trouble.
So, many of the operators are asking for vendor financing? Do you think it is risky?
TK:Yes. It is very hard. 3G is a very promising market, but if you have to choose whether one particular player will be successful or not, then it is very hard. |
