Thumbnail: 

CSL CTO Christian Daigneault tells Telecom Asia group editor Joseph Waring why its new HSPA+ network is a key differentiator and why its has invested to be the fastest

Joseph Waring: You switched over to your all-IP HSPA+ network in April. Can you give us an update on that move‾

Christian Daigneault: There is lots of interest in the shops. This is not the handset market so we're not expecting the same numbers for dongles. We cannot talk about the numbers as this is very strategic. But, in terms of traffic, we saw data traffic double in the first month compared to that on our 3G network. Users are consuming a hundred times more data than the regular users, so it doesn't take many users to make up the traffic.

Our 3G network was not really being used for data, although we saw data traffic double each year. But that was nothing. That was the traffic generated by handsets, downloading music and things like that. But once we offered dongles, it complete changes user behavior. In just one month we saw traffic double, so it remains to be seen what it will be after 12 months.

What are the implications for your network if traffic continues to increase at this rate‾

First of all, this network has been built with huge capacity. A typical network in Hong Kong is using one frequency, although we have the possibility of three. HSPA 21 Mbps gave us more efficiency so we can consume more data because we have efficient modulation.

We have built this network for 40 times the capacity of our existing network on day one. So there is no issue for this year. But if the success continues the way we see it now, that would be a good problem, and we'll need to expand further the capacity.

With many people now looking at the infrastructure more as a commodity, which can easily be outsourced, how much of a differentiator in the market is an operator's network today‾

We think it's still a huge differentiator. All the operators have access to the same technology because technology is a world standard. But how much money they put in the network is the choice of each operator. We believe we are investing higher, and the Next G network is a full-blown replacement network.

All our promotions are based on having the highest speed and the widest coverage in Hong Kong, so network differentiation is really what we are promoting. We see from tests that the network is very different, and once users try it and experience it they will see the difference.

What happens when everyone offers 21 Mbps‾

The challenge is that it is becoming very technical and more difficult for the customer to be able to understand. You can have a very good device, like the iPhone, but if your network is congested, then you work at 50-kbps max because you have so many users. Although every operator is claiming they have 14.4 or 21 Mbps, their ability to support that speed at peak times is really a function of how much they invested in the network capacity.

If you look at the fixed broadband example, there was a lot of talk about speeds in the past, but now speed is rarely something that comes up.

Will that happen in mobile broadband‾

I think it will. Right now speed on the fixed network is becoming a commodity - it's something that assumed to be there. For wireless it will take some time. Yes, in five years from now, potentially you will not differentiate with speed. But you still have a long way to go. The reason for that is that the investment is significant to be able to bring that capacity.

Do you see high-speed mobile networks as a viable replacement for fixed broadband‾

Yes, in the long run. We are targeting to be a fixed broadband replacement, which is why we think we'll need capacity and technology like LTE on top of what we have now. Mobile couldn't be a 100% broadband replacement at this stage. The capacity would be enormous, but there will be some level of replacement for people who require mobility.

I think the trigger will be mobility, and then people can ask if they still need fixed broadband at home. We will need to come up with new products because the issue with home broadband is normally many users share the connection. We are looking at products, like mobile routers. But then the consumption becomes very high, and we need some experience in the capacity requirements before we can move ahead on this.

What is your device strategy going forward‾ Do you see dongles having less impact a year or two from now as handset user interfaces improve‾

Dongles will be the main driver of network traffic and to a lesser extent our TV Studio on Demand and Musicholic applications, which were recently launched. We're introducing the HTC Touch HD and Diamond2 so users can maximize the network.

But handsets are not stretching the capacity of the network. Even when we talk about the implications of apps like Studio on Demand, they consume only a fraction of the bandwidth of PCs. Most people would watch Studio on Demand for five to ten minutes, so even after some time you use only 30 megabytes. When you use the dongle, you can very quickly go to 1 gigabyte in one day.

The iPhone is the best mobile device for using the internet. But when you look at the statistics, the traffic level is far from the level of dongles. I would say the dongle gets ten times the volume. But if handsets like the Touch HD and Diamond2 were to become mass market, then they would have a big impact, but that is not the case.

What about other formats - MIDs and netbooks‾

The PCs we're selling in our shops now are the mini PCs, which are the same price as high-end handsets. You have to think more and more people will use their handsets to access the internet, but the limitation of the screen size means that with mini laptops you will still always consume more [bandwidth].

What happened to Wimax in Hong Kong‾

I believe it's dead. I would be surprised if it comes back. We were surprised that no one came in and proposed Wimax, although from a technology standpoint, we always thought that there was no place for Wimax in Hong Kong. But this is our belief.

We thought others had the belief that they could make Wimax work.

Already HSPA+ has exceeded the throughput of Wimax, which is in the 6-Mbps range. There was a window of opportunity probably two years ago or last year, but now it doesn't have any benefit over LTE. The biggest challenge for Wimax is that LTE is backward compatible with HSPA, so we can decide to deploy LTE only in locations where we really need the capacity. In terms of coverage we can do it much faster. If you deploy Wimax and want to achieve the same coverage, the investment would be huge in Hong Kong.

I think that Wimax been a success in some developing countries as a fixed replacement where there was no fiber in rural areas.

The downturn has forced many operators to focus more on opex. What moves have you taken to run your operations more efficiently‾

We were running two networks after we merged two years back with New World Mobility. The networks were running separately with five switching centers, which wasn't efficient. With the new network we have two switching centers and a fully integrated network, which brings economies upscale. We now have more efficient backhaul, operations and maintenance.

Another area, and this is common to all operators, is we are renegotiating the costs of site rentals. The price of real estate has dropped some 20% on average but not for the operators. It is a bit more challenging because it's not the same as people going from one apartment to another. We are relocating a high percentage of sites because we are negotiating with the building owners and they are still asking in many cases for increases of 20-30%, which we cannot accept. Relocating increases our capital costs, but its largely compensated by bringing down the price of our site rentals.