Ruthless competition in the handset market continues to drive manufacturers to search for new opportunities to drive down cost, printed circuit board (PCB) area and power consumption. Simultaneously, the rollout of third-generation (3G) networks has opened the door to a variety of new multimedia and data-based applications, from wireless Internet access and mobile video to text messaging and mobile TV.

As demand for these new applications rises and the market becomes increasingly globalized, handset makers face a quandary. How can they support the increasing number of frequency bands to support global platforms and the multiple high-bandwidth technologies needed to deliver these new revenue-enhancing services without violating the market's strict cost, footprint and power constraints? The number of frequencies supported by the latest 3G partnership program (3GPP) standard has increased from three to 10 and is set to continue to expand. The current frequency bands and their associated bandwidths are shown in Figure 1.

One thing seems clear; to succeed in today's markets, handset designers need to deliver multiband, multimode capability. While existing 2G GSM/GPRS networks continue to thrive and represent the largest percentage of networks today, shipments of handsets based on EDGE technology, which boosts data rates by introducing a second modulation format into the GSM system, are rapidly growing. At the same time, network operators are continuing to roll out 3G wideband CDMA (WCDMA) networks. Based on the universal mobile telecommunications system (UMTS) network topology, this new technology is rapidly becoming the leading global mobile-broadband solution. Industry analysts predict WCDMA and EDGE will represent the two fastest-growing segments of the handset market over the next few years. Moreover, to meet demand for IP-based services, a growing number of UMTS operators worldwide are deploying high-speed downlink packet access (HSDPA) capability. High-speed uplink packet access (HSUPA) is ready to follow in the near future. Figure 2 offers an overview of each cellular standard and associated up and downlink data rate.

At the same time carriers and service providers believe the time has come to accelerate the WCDMA evolution path toward the 3GPP's long-term evolution (LTE) initiative. LTE is emerging as the leading technology for next-generation wireless broadband networks. It will deliver data rates up to 100 Mbps for downlink and 50 Mbps for uplink and improve network coverage and efficiency by using orthogonal frequency-division multiplexing (OFDM) transmission with multiple-input, multiple-output (MIMO) smart antenna technologies.

While LTE will lay the groundwork for 4G technologies, it requires network operators to support yet another modulation scheme. To capitalize on the capabilities these new network topologies offer, network operators must address two formidable obstacles: higher costs and higher power consumption. The BOM costs for WCDMA handsets are double that for EDGE handsets and nearly triple those for GSM/GPRS devices. At the same time, GSM handsets offer twice the typical talk time of WCDMA phones, a key factor in customers' perception of the quality of a phone.

Those distinctions are largely attributable to the higher complexity of WCDMA front-end architectures. WCDMA is a spread-spectrum technology that spreads its transmissions across a wide 5 MHz carrier. Since WCDMA uses full-duplex communication, the receive and transmit functions operate simultaneously. This requires front-end electronics that attenuate the transmitter's wideband noise to prevent degradation of the receiver's sensitivity. Typically, this is accomplished using a duplexer along with additional bandpass filters in the transmit and receive paths. Moreover, designers commonly use external LNAs as well. The additional component count and area contribute to WCDMA's higher cost relative to GSM/GPRS and EDGE alternatives.

Power efficiency is also a challenge. The output power amplification stage typically consumes a large percentage of the battery capacity in any wireless device. Unlike the power amplifiers (Pas) in GSM handsets, which are used in saturated mode, Pas in WCDMA systems operate in linear mode. The use of complex quaternary phase shift keying (QPSK) modulation techniques requires the PA stage to be highly linear so as not to degrade the quality of the signal or to smear it into adjacent channels. Accordingly, WCDMA designers constantly face a trade off between the high linearity needed to ensure excellent WCDMA performance and the high levels of power efficiency required for longer battery life in handset designs.

Traditionally, handset designers seeking to support multiple air interface standards in the same device have resorted to stacked radio architectures with separate radio transceivers for different standards. While the industry has made major advances over the past few years in packaging technology and designers have made significant progress integrating portions of the receive path of a multiband, multimode phone, the same cannot be said of the transmit side. Typically, the use of multi-air interfaces has a major impact on handset component count because it requires the use of multiple surface-acoustic-wave (SAW) filters, oscillators, filters, and specialized mixers. Needless to say, a large component count poses a major liability for designers competing in the cost- and power-conscious wireless handset space. Moreover, this duplication of functionality conflicts directly with the need to minimize product PCB area. A typical seven-band WEDGE radio block diagram is shown in Figure 3. It shows that currently four Pas, 10 SAW filters, three LNAs, three duplexers and a nine-throw switch are required to realize this front-end function.

Clearly, designers building handsets for worldwide markets need an architecture, which eliminates the redundancies inherent in the stacked radio approaches used today. A single, common transmit path could maximize on-chip circuit re-use and reduce system BOM costs. It would also save PCB area and simplify the design of the handset's front-end. Moreover, since linear Pas consume a large proportion of the battery capacity in handsets, a single transmit path capable of using non-linear power amplifiers (Pas) could dramatically reduce power consumption and extend handset battery life.

1 2 Next

RSS    Save to Del.icio.us  Digg This