The Evolving Network

One thing that emerged from Mobile World Congress in Barcelona this year was the certainty that the network of tomorrow won’t look much like the network of yesterday. The ongoing tidal wave of more and MORE and MORE capacity continues to drive the adoption of higher capacity IP-based systems. Those who do not adapt are rapidly being left behind. In fact, recent market surveys have shown that the hybrid microwave systems, which have been the majority of the systems sold in the past two years, are beginning to decline in terms of market share and the IP-based systems are projected to grow to dominate the microwave market in the next few years. However, the capacity growth is only part of the story. There were at least two other trends that are relevant to the middle mile.

The first trend is driven by the capacity growth and it is the desire to offload the low-value traffic from the 4G packet switched network to the best effort internet as soon as possible. The presumption here being that the 4G network, with its strict SLAs, is inherently more expensive per bit than the best effort Internet. In order to achieve this, however, we need to place more intelligence closer to the edge of the network to allow the traffic to be classified and appropriately routed. This increases the cost of the edge devices and that cost increase must be offset by reduced transport and core switching costs. Further, this implies a rich interconnection between the 4G packet switched network and the best effort Internet network with all the implications that has on security, transport, switching and billing. Although most market analysts agree that 4G offload is a significant and growing portion of the market in tomorrow’s network, the implications of that on the ecosystem are far from clear at this point in time.

The second trend is the move to smaller cell sizes. With the announcement of a variety of small cell approaches ranging from Ericson AIR (Antenna Integrated Radio) to Alcatel Lucent’s (News - Alert) lightRadio architecture, which shrinks the radio access radio and remotes the base band processing to the cloud, to a large variety of smaller players pushing solutions originally targeted at femtocells to pico/metro cell designs for larger number of users and indoor as well as outdoor form factors.

The main driver for this migration to smaller cells is, of course, still the capacity growth and the desire to make more efficient use of the Radio Access Network resources and reduce the number of users per base station to deliver more capacity per user. While the winners in this space are far from clear, the potential performance advantage that this approach offers seems to guarantee that a significant number of the new base stations in tomorrow’s network will not be the traditional macro cells of yesterday’s network. Much of the focus has been on the base station design and form factor and to date very little attention has been paid to what will happen to all of the other elements of the traditional macro base station – i.e. power conditioning, battery back-up, environmental protection, switching and transport. Most of these functions are still required for the new microcells and they will need to shrink in size and cost just as much as the base station.

Fiber-based backhaul at the lamp-post is much less likely than at the macro cell site tower. Further, the wholesale backhaul model that depends on multiple tenants per tower does not work in the new world either. At the same time, conventional microwave backhaul, with large parabolic antennas, is not a viable solution here either. Finally the drive to offload the 4G network introduces additional complexities and demands additional functionality that must be factored in as well.

In order for the small cell revolution to be successful we will require innovative solutions that provide all of these functions in a well integrated, small, cost effective package. Point solutions for cell site routers, transport, power, etc., will not be economically competitive or will they be acceptable to the city planners for deployment at street level.

On top of these trends, the service set that the users are demanding continues to evolve. This drives changes in the required SLAs in terms of capacity, peak vs. average, committed information rate, latency, downlink to uplink ratio and many other factors. It is no wonder that tomorrow’s network will not look much like yesterday’s network. In order to survive and flourish, the middle mile providers (and it is becoming much more than simple backhaul) will need to continue to innovate and adapt. The bad news is that last year’s products won’t be optimized for tomorrow’s networks. The good news is that there continues to be opportunity to deliver significant value to our customers through our innovation. As in the Chinese proverb, “we live in interesting times.”