With 3G, TDMA providers will have more options to help reduce blocked calls. But it's still possible to wring more capacity out of their 2G networks.

Ask most industry experts about how the major IS-136 service providers in the United States plan to improve capacity in the future, and most will throw out solutions such as EDGE or W-CDMA. Although those 3G upgrades will boost capacity, the fact remains that many of these providers are facing capacity issues in their core areas right now, and the nature of the TDMA format makes an immediate remedy a serious challenge.

Unlike CDMA, IS-136 was designed primarily to fit easily into the existing AMPS networks. It uses 30kHz channels, each of which can accommodate up to three users, who share timeslots. One obvious method of immediately increasing capacity would be to increase the amount of timeslots in each 30kHz channel. This method is used in the similar GSM system used extensively in Europe and in PCS bands in the Americas. In IS-136, however, the effect on voice quality that comes with moving to a half-rate vocoder would seem to outweigh the capacity gain. As a result, there has been little implementation.

Although three users share a common 30kHz channel, the obvious problem is that there are still only so many channels available. Worse, if there isn't enough geographic spacing between two channels that reuse the same frequency, the result is co-channel interference.

Frequency-reuse patterns were developed in order to limit this problem. The industry norm is a reuse pattern of seven, although some providers do implement variations. If a provider is looking to increase its capacity, a natural step is to attempt to decrease this reuse pattern. Some providers currently are looking at methods that allow reuse patterns of four.

A Tight Fit
Several concerns arise when lowering the reuse pattern. The greatest is co-channel interference. Michael Elford, CenturyTel director of engineering and construction, said, "While we do have serious concerns about capacity in core areas, and we are actively looking at methods of increasing capacity, there is no way we would even minimally risk reducing voice quality in order to accomplish such a capacity increase."

Lowering the reuse pattern without hurting voice quality often is a balancing act.

"The primary challenge in moving to a lower reuse is getting the handsets to power down while keeping bit-error rates below an acceptable level and at the same time minimizing any interference sources" said Eric Parker, BellSouth performance manager, Atlanta market.

Interference itself is another capacity robber. Many urban locations face tremendous out-of-band interference, which often causes intermodulation products that can seriously diminish voice-channel performance to the point of rendering them useless. Increasing the base station's sensitivity in order to lower the handsets' transmit power in a high-interference environment can lead to more problems than are solved.

"One possible solution (is) superconducting filters and low-noise amplifiers on the base-station receivers, which can offer improved sensitivity while dramatically limiting the interference in the urban environment," Parker said.

Although tower-mounted amplifiers also could be deployed to increase the base-station's sensitivity, there's the possibility that the urban environment's high interference could have an even more dramatic effect. Another important consideration is that many receivers' front ends will suffer degraded performance along the band edges, sometimes called rolloff.

Another potential solution is dynamic channel allocation. Although each infrastructure vendor can take a different approach, the technique essentially involves the network deciding which channels to allocate where, depending on interference and usage. As a result, the network automatically optimizes itself for the best capacity and voice quality. A central processing center can make decisions based on base-station field measurements about whether to use certain channels. The end result should be more channels used - but only if voice quality for the call will be above an acceptable level.

The Digital Dilemma
Shifting as much traffic as possible from analog to digital is one obvious way to increase capacity. Because three callers can share a single 30kHz channel, a provider can get a threefold increase in capacity as traffic moves to digital.

"Because we have a lot of rural coverage, many of our markets still have high analog use, and many of our customers still use high-power analog phones," said CenturyTel's Elford. "This, of course leads to two issues: We want to move our customers to digital to increase features and quality as well as increase our capacity, and it also means we have customers making calls on 3W mobile phones."

Rural networks originally were designed for 3W bag phones, which have more power and thus a greater range than modern 0.6W portables. Some subscribers still use older 3W phones, but in urban areas, their high power often causes interference.

The bottom line is that although providers are focusing on next-generation solutions, most will have to look at how they can increase their capacity today. For TDMA/FDMA networks, the answer probably will involve a mix of several methods: lowering the frequency-reuse pattern, decreasing the effect of interference, increasing digital traffic and implementing dynamic channel-allocation systems.

Miceli (amiceli@suptech.com) is Superconductor Technologies regional sales manager.