(FORTUNE Magazine) – After four years of bankruptcies, accounting scandals, and general investor malaise, telecom is having a strong 2004. As of late September, phone company stocks were up 6% for the year, and shares of their gear suppliers had soared 30%. But to hear the analysts who follow the sector, you'd never know things were improving. "There's nothing exciting there," grumbles Susan Kalla, a telecom equipment analyst for Friedman Billings & Ramsey. "It's like watching grass grow." Ditto, says Jeff Halpern, who follows phone companies for Sanford Bernstein: "It is hard to see a growth story."

And indeed, investors have good reasons to be cautious. New technologies and deep-pocketed competitors are flooding the industry at an alarming rate, and all the players are scrambling to adjust. AT&T and MCI continue to face intense pricing pressure in their long-distance businesses. Plus, the government plans to take away a regulatory benefit that allowed them to buy wholesale local phone service at deep discounts, cutting off a potential source of growth. Nor should investors get hung up on the Baby Bells. Yes, they've performed well over the past 12 months--partly because they benefit from the same ruling that hurts AT&T--but their core local phone operations are under attack from all sides. Cable companies and startups alike are using Internet networks to deliver cheap calls; wireless operators are pushing cellphones as a substitute for traditional home phones; and soon new categories of "fixed wireless" services such as Wi-Fi and Wi-Max will be able to deliver phone calls just as readily as the Baby Bells.

Fine, you say, I'll invest in the attackers. They must surely be growing. Well, put those plans on hold. The cable companies have their own challenges. Part of the reason they are pushing into the phone business is that they are losing market share to scrappy satellite operators. What about wireless? Sorry, wrong number: Cellphone companies are still on the rise, but most of them are units of the traditional telcos. And one of the few wireless pure plays, onetime Wall Street darling Nextel, suddenly has its skeptics. (For more, please see following story.) As for Wi-Fi, everyone loves the technology, but it is not clear how a company makes money selling the service. (Cometa, a private company that aimed to sell Wi-Fi service wholesale, closed its doors earlier this year despite backing from Intel and IBM.)

The good news for consumers is twofold: Lots of nifty new services are on the way, and all the competition is driving down prices. And the upshot for investors? Service providers of all stripes will have to spend like mad to keep pace with the competition.

That's why most of the money managers and analysts we talked to favor a handful of equipment makers--the companies arming the attackers and defenders alike with the tools to do battle. Yes, investors have heard this mantra before ("No matter how you cut it, you've got to own Cisco," FORTUNE wrote in 2000, when shares were trading at $60, vs. $18 today), and there are plenty of suppliers that investors may wish to avoid. (Those under investigation by securities regulators come to mind.) Heck, given the high valuations and record of volatility, cautious investors may want to shun the sector altogether. But we fearlessly set out to find good companies focused on the growth areas of telecom--wireless, broadband, and VOIP--and identified a handful of would-be winners.

Investors who want to take advantage of growth in wireless should consider Qualcomm (QCOM, $39), the maker of software and chips for cellular networks. Wireless providers are in the process of upgrading their narrowband voice networks to broadband networks, capable of transmitting digital photos, web pages, even movies, at fast speeds. To do so they all need to incorporate some form of Qualcomm technology, a wireless standard known as CDMA.

The big risk associated with buying Qualcomm is its already dizzying stock price. After an 86% gain in the past year, it trades at 40 times the previous 12 months' earnings. Nevertheless, Albert Lin of American Technology Research says it isn't too late to jump in. In 2005, Qualcomm is expected to increase revenues at almost two to three times the rate of other semiconductor companies, says Lin. His favorite statistic, though, is that Qualcomm sells more per employee than any other publicly traded company--some $690,000 per worker. "Managing your labor force is getting more expensive and complicated," Lin says. Qualcomm's ability to generate strong sales from a relatively small employee base "justifies a significant premium above its peer group."

Folks looking to double down on the future of wireless might like Alvarion, (ALVR, $13), a maker of antennas and other equipment for Wi-Max. (Think of Wi-Max as a brawnier version of Wi-Fi, capable of transmitting data and voice calls over distances up to 30 miles.) Wi-Max is in its infancy right now. U.S. regulators think it might be a good way to deliver broadband in rural areas, for example. But John Freeman, an analyst with Precursor, an independent research shop specializing in technology, believes Wi-Max will be huge overseas, in markets that lack traditional phone or Internet services. The tiny Israeli company (2003 sales totaled $127 million) already does business in 130 countries, and it has contracts or trials with all four Chinese phone companies. "China and India alone could drive this company," Freeman says. The stock is trading at a vertigo-inducing 550 times trailing earnings, but just 30 times 2005 profit estimates--underscoring how fast the company is growing.

Investors have been hearing a lot about Voice Over Internet Protocol, a technology that translates phone calls into the language of the Internet and delivers them over data networks. It is probably too soon to tell which phone companies will make money deploying VOIP, but Lucent spinoff Avaya (AV, $14) seems poised to do well in the transition. Avaya makes corporate phone systems, and increasingly that means it sells gear that aggregates voice calls on corporate data networks. It has a tough rival in Cisco, which leads with about 34% of that market, according to Synergy Research. But Avaya is coming on strong: It now commands roughly 24% of the market, up from 10% at the end of 2002. S&P Equity Research analyst Ari Bensinger likes Avaya's improving margins. Operating margins climbed to 9% of sales in the second quarter, up from breakeven a year earlier. At 25 times trailing earnings, Avaya trades at a discount to peers such as Nortel. Bensinger--who rates only one stock, Avaya, a buy--suggests it should be valued at closer to $22.

One company positioned to take advantage of both the VOIP and wireless trends is Tekalec (TKLC, $17). Tekalec makes signaling systems that help phone companies coordinate and control traffic on their wired and wireless networks. It also acquired two companies that make switching equipment for VOIP networks. Erik Zamkoff, a telecom analyst with Independent Research Group, says both businesses are humming right now, yet the market isn't assigning any value to Tekalec's VOIP switching business, which he thinks is worth another $3 to $4 a share. "It's like you're getting the VOIP business for free," he says. The way things are going in telecom these days, buying stocks that give you something for nothing sounds like a smart call.