As the market awaits the latest 3G phones, it's clear that consumers are eagerly anticipating new mobile-entertainment experiences. Mobile digital-broadcast TV is promising to deliver the big bang that will excite consumers worldwide. It marries two of history's hottest technological innovations: television and wireless telephones. The task that confronts wireless OEMs is to develop TV-enabled handsets that users will like.

Limited offerings of analog broadcast television are now available, but mostly in Japan and Korea. In addition, the U.S. has already seen early trials of DVB-H technology. These early mobile-TV offerings have shown that consumers want high-quality mobile TV. Yet they've also established that analog picture quality simply isn't good enough. A viewer who's watching a 2-in.² screen may not be able to detect minor losses in video quality. But that viewer certainly will notice the pops and fades and signal drops that are apparently inevitable in mobile analog TV. Consequently, digital-television reception will be essential as OEMs and service providers seek to win the hearts and minds of television viewers on the go. Only with digital TV can reception be bright, clear, and reliable enough to please huge numbers of users.

Most of the capabilities that are needed to reproduce TV picture and sound are already built into 3G handsets. For example, phones based on Texas Instruments' OMAP platform already provide all of the necessary back-end processing, H.264 video decoding, and formatting color display. They also include the user-interface capabilities that are required for digital television. Add an appropriate antenna, tuner, and demodulator/decoder and you quickly have digital television on the wireless handset.

Although these additions aren't simple, they would be readily do-able if consumer preferences weren't paramount considerations. Putting the components together in an appliance that consumers will like presents several significant manufacturing and design challenges. In some cases, the technology needed to meet these challenges isn't yet available to OEMs (FIG. 1).

As has been the case with almost every advance in handsets, limiting power consumption will be the biggest challenge. Wireless OEMs already began to reach beyond simple telephony to add such features as color display, MP3-player capability, video games, and now digital television. In the meantime, consumers have refused to sacrifice talk and standby time even as they clamor for the latest handheld-device innovation.

It's safe to assume that the broad marketplace acceptance of mobile-television technology will depend on the industry's ability to receive, process, and present TV pictures and sound without significantly cutting battery life. Unfortunately, the power consumption of a typical television tuner makes this difficult. In most existing configurations, a tuner consumes 2 to 3 W. It roughly equals the combined power consumption of the other components of a wireless phone. In other words, simply adding a tuner to the phone could reduce battery life by as much as half. Bolting a tuner onto the front of the phone also could cause the device to heat up as it is being used. While this aspect might present some interesting marketing opportunities in cold climates, most consumers would not welcome such a thermal effect.

To overcome these difficulties, mobile-television standards provide for energy reduction within signal transmission and reception. DVB-H, for example, includes a time-slicing scheme. This scheme allows the handset to receive data in bursts. It also can switch off the tuner and demodulator/decoder between bursts. Other standards, such as ISDB-T, truncate video bandwidth. As a result, they limit the amount of data that's required for TV on a handheld device. These schemes can reduce power consumption by as much as a factor of 10. Yet both of them somewhat degrade picture quality. The fact is that displays on wireless handsets are so small and have such low resolutions that the human eye cannot detect such picture-quality reductions.

Advances in semiconductor-process technology promise to further improve the power economy of wireless handsets. By integrating the tuner, demodulator/decoder, and signal processor into a single submicron CMOS device, designers can help to extend wireless battery life. They will effectively reduce overall system power requirements. In addition, submicron CMOS allows the radio-frequency tuner to consume a fraction of the power of conventional tuners. Bringing the tuner on chip with the digital signal processing also allows the tuner to be precisely controlled by the system logic. This feature further extends performance while reducing power. Of course, a well-architected digital-signal-processing system will always consume minimal power in a deep-submicron CMOS process that's been optimized for mobile-device applications.

As digital-television capabilities are integrated into the handheld entertainment center, form factor is another major challenge that must be faced. Consumers have grown accustomed to extremely compact devices that fit easily in a shirt pocket or purse. They are unlikely to accept anything larger. Yet a standard television tuner can be quite large. Plus, a tuner and demodulator/decoder together may require several semiconductor chips.

Mounting all of this hardware on a board would eat up more real estate than a wireless handset designer can spare. Instead, the handset calls for a tuner and demodulator/decoder combination that occupies no more than a few square centimeters of board space. CMOS technology can meet this challenge by providing the necessary integration within acceptable real-estate limits.

Another test will be the task of adding functionality without greatly increasing system cost. A conventional tuner can be quite expensive. But market studies suggest that consumers aren't willing to pay much more for a wireless handset than the prices that they see today. As a result, the additional components that are required for handheld television reception must cost very little compared to their conventional cousins.

Most needed functionality is already built into the phone, which will help OEMs control cost. Semiconductor makers also can do a lot to address cost concerns as they integrate the tuner, demodulator/decoder, and processor. In the race to hold the line on handheld-wireless-system prices while adding television-viewing capabilities, extremely high levels of silicon integration will be critical. Such integration must be based on the most cost-effective process technologies.