Moxa Americas, Inc.'s Charles Chen ponders the way forward for transportation communications networks in the US
The communications market, like any other, has its strengths, weaknesses, opportunities and threats. What distinguishes it is perhaps the sheer pace of technological development - something which makes it especially difficult to gain a perspective on future directions or the long-term validity of any investment.Charles Chen, embedded business development manager with
"A decade ago, we were still using ISDN and POTS, people hadn't even heard of UMTS and ADSL - 3G was barely even on the horizon. Now we're looking towards 4G and beginning to speculate over the capabilities of 5 and even 6G.
"In the US and Europe, we already own more mobile telephones than there are landlines and it'll be user behaviour that will influence technological development. More and more people are using mobile internet services but the existing network data rates are too slow to host many of the applications people want - they can't get the response times they need. So there's a big demand for speed improvements: that's Opportunity Number One.
Opportunity Number Two is network security, which is going to become a much bigger issue. Again, that'll be driven by user demand. Whereas 10 years ago most of what we communicated was over leased lines, now much of what we do is over the air. 'Security' comes in two parts, authentication and data encryption, and I think we're going to see both network algorithms and protocols improve dramatically in the near future. It's certainly an area where the IEEE is placing a lot of emphasis. Network security is a major feature of the New York City Wireless Network [NYCWin] project. NYCWin provides the city with a wireless environment in which officials can securely access data at a rate 50 times faster than over current networks. Moxa partnered with Northrop Grumman, which is making something of a speciality of secure networks, on this project.
"Opportunity Number Three is the diversity of technology available for access to hardware, so in plainer terms that means that CPU and IC chip prices continue to fall, as does the cost of memory. Again, user demand is driving the development of lower-cost solutions with more reliability."
Threats
Although technology continues to develop at a rapid pace, one of the greatest challenges, perversely, is the rate of replacement - although this sets up certain dynamics.Chen: "Developing countries are at an advantage here, in that they can leapfrog the legacy developments that we find in the US and Europe. The US for example has a very widely deployed ISDN/POTS, or copper-based, network. This is older, analogue-based stuff but it is very widespread. Replacing it will take years, and a lot of money. Countries such as China and India can go straight to fibre and learn from our experiences. That's Threat Number One.
"We also face a situation where the different types of technologies don't work with each other. WiFi, for instance, can't communicate with GSM/GPRS. 4, 5 or 6G might well take care of that and again it's an issue being addressed within IEEE. But for now we're stuck with a lot of isolation. That's perhaps Threat Number Two.
"We have to realise and accept that we're in the 'middle age' of cellular. It's neither new, nor is it fully matured in many respects. We already have GSM/GPRS. We talk about 3G, CDMA and so on but the reality is that different countries' regulatory environments still encourage and result in different technologies. We need to drive their better, more rapid combination. Call that Threat Number Three.
"However it perhaps doesn't benefit us to make firm decisions as yet, precisely because the technology is still developing and changing. Ethernet, TCP/IP, 802.3, we can consider that mature enough already. But 802.11 and 802.15, the standards for wireless local area networks and personal area networks respectively, aren't really there yet. That makes it hard for us to decide on the right combinations at present."
Future-proofing
That causes customers something of an issue but Chen says that future-proofing a network isn't as onerous as might be imagined."That's a question that customers ask me a lot: 'How can I future-proof my network if technology and standards are still in a state of flux?'. The answer is to base things on the hardware. Manufacturers have worked hard to find common communications interfaces. Take, for example, the PCI slots on a PC: they're standard and have been for perhaps 20 years. They'll probably never change. Nor will USBs, PCIEs, PENs or PC/104s. So you can control the computer hardware side. You want to use 3G now? Fine, you can. But tomorrow you'll want 4G. How do you prepare for that? Simple: reserve some of the common interfaces in your carrier boxes and then when the time comes remove and replace some of the cards. You don't need to change the main hardware, just the peripherals. That's the key, and we implement it in a lot of our products."
The expected and the unexpected
Such an approach is necessary when looking to manage the sheer pace of development, he continues."The background to all of this is the increasing capability of communications devices. Three years ago, cellphones made basic calls and had a few other features such as a fairly useable camera or MP3 capability. They might also have been able to access the internet but things were slow - after all, you we're talking about a CPU with only about 50Mb. Now, we have CPUs running at 1GHz. That's 20 times more computing speed in just three years. What was considered sophisticated and probably only existed in the hands of the military and a very few top-end users is being commoditised, getting more robust and less expensive all the time. Cellular technologies will merge in the future - 4G will probably bring CDMA and TDMA together. And don't forget that people will continue to change their phones annually."
Standards drivers and market factors
Whereas in some countries such as Japan where the government is the prime driver of standards work, in the US it is the network owners who have the biggest say. Manufacturers and end-users actually have little say and tend to have to follow the big players' (for which read AT&T's and"I think they're on broadly the right track but they're slow," says Chen. "They feel the pressures brought to bear by end-users but they still tend to dominate. They realise that they can earn more from the changeover from 3G to 4G; they already charge lots more for data than voice, for instance.
"The US is still the world's biggest economy but as things are it risks losing its technological lead to countries such as China and Germany. The US always seems to be the last to adopt new technology - there's not enough awareness of that.
"Other countries have clear competition between vendors but the duopoly we have in the US makes it very hard for other companies to break into the market. I pay almost twice as much a month as I did a couple of years ago for a phone from T-Mobile. How can that be, when the trend everywhere is for communication costs to fall? It's because T-Mobile has to lease infrastructure from its direct competitors and it highlights the problem."
Wireless versus hard-wired
"Wireless has the advantage over fixed networks of being easier to install and replace. It'll take us years to replace the old, copper-based network in the US whereas wireless can be installed in months - the NYCWin project, for example, implemented 400 new stations inside a year."It doesn't hold all the cards, though. There's no doubt that hard-wired TCP/IP networks are more secure and have greater bandwidth. That said, I see the balance shifting significantly in wireless's favour in the coming years.
"For developing countries, the solution is much more straightforward: install and commission a high-speed fibre-based network which forms the communications backbone and add a wireless overlay for specific needs and situations.
"Developed countries, with their significant sunk investments in older-generation hard-wired networks, need to place much more emphasis on wireless - and they need to look at improving the robustness and speed of wireless solutions. That's far easier and cheaper to do than ripping out and replacing what's already there. That fact's already being grasped in some parts of the transportation sector. On the train networks in Europe, for example, rolling stock is increasingly being outfitted with wireless systems because they are significantly simpler, in all respects, to upgrade or replace.
"That gives us Opportunity Number Four, although there's good reason, perhaps, to elevate it to Number One: aggressively closing the gap in performance between wireless and fixed networks and systems. In the US specifically we're currently seeing much greater implantation of secure wireless networks to support emergency management and military capabilities but the transportation sector still has to appreciate and grasp the opportunity."
