The Happy Medium

Narrowband PCS (N-PCS) has breathed new life into the paging industry. A 1998 study by Frost & Sullivan estimated that by 2004, half of all paging subscribers could be using N-PCS. By 2000, the study projected, N-PCS revenues would approach $2.5 billion, up from roughly $48 million in 1997, the year that N-PCS made its mainstream debut.

But N-PCS' advanced features also mean new optimization challenges. As in racing, higher speeds mean less tolerance for error.

"It's much tighter criteria when you're running at the higher speeds, such as 6.4kb/s," said Sue French, Glenayre senior RF-systems engineer. "As the paging world has gone from POCSAG to Flex, we've kept turning up the speed. You don't get something for nothing. To turn it up, we have to watch (for) the possible increase of interference."

One place where interference can take its toll is the return path. In the past, paging largely was about blasting a signal into as many places as possible. That's only half the battle with 2-way, where path balance is a new concern. A site might be able to put a solid signal inside a building, but that doesn't guarantee that the pager will be able to muster a signal strong enough to reply or send a message of its own.

"Not only do you think in terms of broadcast, but now you also think in terms of return," said John Beletic, PageMart CEO. "You need to construct receiver patterns and receiver technology that allow you to provide return paths in the same way that you provide an outbound path."

Two-way has the advantage of store-and-forward, where the pager periodically tries to send the message until its signal reaches the network. But that approach isn't perfect because repeated attempts sap battery power. Increasing the sensitivity of site receivers is another option as long as the site isn't plagued by interference. Noise can be a headache in a 2-way system because it's one more thing that the site receiver has to sift through to hear the pager. As in cellular and broadband PCS, there's no shortage of interference sources, including spurs.

"It's noise many times from paging as well as cellular," French said. "Just because the SMR might be 800MHz, they still may have some emissions in the 900MHz band that could affect our receivers."

Tracking down and mitigating interference isn't necessarily difficult, even for 1-way providers still in the process of learning the nuances of 2-way.

"It hasn't been too much trouble (for) the providers to identify who the noisy transmitter is," said Bill Durno, Glenayre senior manager, N-PCS systems. "You can typically talk to the service provider, and they can filter their sites and have a cleaner signal. If everybody plays within the rules, then the site's not going to be too noisy."

A CELLULAR APPROACH TO PAGINGChoosing as noise-free a site as possible is key because a bad site can come back to haunt just as much as a bad design. But with co-location the de facto law of the land, the number of available sites usually is limited. One-way and 2-way systems typically achieve seamless coverage by using a small number of powerful sites located high above terrain. That's a different approach from cellular and broadband PCS, where the trend now is toward putting more sites with lower power sometimes as little as 20 feet above ground.

"Street level is a disadvantage in a simulcast environment," said PageMart's Beletic. "You want to bathe the whole area. The higher up, the bigger area you bathe."

But in some cases, higher isn't always better. One unwelcome byproduct is simulcast delay spread (SDS), or simulcast misalign, which occurs when the pager is in an area where multiple sites' coverage overlaps. Each signal arrives at the pager at slightly different times because of propagation delays.

"Whenever you have that overlap, there's going to be an area that could have a high delay," said Jamie Marraccine, Grayson Wireless field-sales engineer. "One signal could be delayed much more than another."

At 2-way's high data rates, SDS can make a difference.

"The relative delays to the same point is high," said Glenayre's French. "The definition of high keeps shrinking as the speed gets faster. You used to be able to tolerate a big delay. Now you can't tolerate as big a delay."

One way to identify areas plagued by SDS is by looking at the "eye pattern." If the eye is closed, SDS is present. Drive-test tools, such as Grayson Wireless' PageTracker Elite, log eye patterns, which can be exported to mapping software to identify areas plagued by SDS. Such tools also help track down the source of a problem.

"Plotting this openness metric right next to a bit error rate (BER) would allow you to determine whether or not this BER is caused by a simulcast misalign," Marraccine said. "Before, you would put BER with signal strength, and if signal strength was weak, you'd assume that the BER was due to the weak signal. But if you had high BER and good signal strength, you're at a loss (to determine) whether it was adjacent-channel interference or simulcast misalign."

For the subscriber, BER can be annoying. "When the BER is high enough to where the built-in error correction doesn't result in an ASCII character, you get a blinking character on your pager," Marraccine said. If the carrier provides automatic correction of garbled messages, the pager alerts the system, which resends the message. That's a feasible solution for mitigating SDS, at least until usage increases to the point that resending messages saps too much capacity.

One way to combat SDS is to use lower-powered transmitters and build out in a more cellular fashion by increasing the number of sites while reducing each site's coverage area.

"Sites up in high locations, which may have worked for lower-speed paging systems, might not be usable for the higher-speed paging systems," Durno said. "Some sites would have to be moved off of the high locations to reduce the SDS to a tolerable level."

SDS is one example why covering a market no longer means just putting up a couple of high-powered transmitters on the tallest buildings available. Network design is changing as carriers find that 2-way's unique needs require approaches different from those that worked in 1-way.

"Some paging operators are taking their slower-speed networks and just turning up the speed with their existing sites and seeing some increase in their SDS interference," French said. "There are ways around that. You can modify your sites, change your antenna configuration (or) change some parameters and launch delays to clear up the interference. Sometimes, you'll have to relocate sites.

"You can still have a lot of capacity without having really tiny sites, like in cellular. But the sites can't be quite as large as the old-time POCSAG sites. We aimed for the more balanced system, but it's actually more in between traditional 1-way and cellular."

RF design isn't the only area where many of the old rules no longer apply. PageMart, for example, built an IP-based network to each of its sites across the United States.

"For us as a paging company, that's a whole new world going from the very simple paging protocols to IP networks," Beletic said. "There was a whole learning curve associated with that."