Single Chip Holds WiMAX Transceiver

June 13, 2007
The industrys first single-chip WiMAX transceiver supports wireless broadband applications from 2.3 to 2.7 GHz with many cost-saving features and functions.

Broadband services, such as Internet access and streaming video, are coming to a wireless transceiver near you. Many wireless service providers are already offering forms of wireless broadband service through their existing cellular networks. Many are also considering WiMAX equipment as an effective way to augment their networks with broadband-wireless-access (BWA) capabilities. Of course, affordable WiMAX service requires low-cost WiMAX equipment, and the new model MAX2837 wireless broadband RF transceiver integrated circuit (IC) from Maxim Integrated Products is a single-chip solution in that direction. The transceiver operates from 2.3 to 2.7 GHz and supports fixed, portable, and mobile WiMAX operation.

WiMAX, which stands for Worldwide Interoperability for Microwave Access, has been called a wireless replacement for wired digital-subscriber-line (DSL) technology. The wireless-metropolitan-area-network (WirelessMAN) technology provides high-data-rate communications for fixed-station subscribers at distances as great as 30 miles from a base station and for mobile station subscribers at distances as great as 10 miles from a base station. The operating “rules” for WiMAX equipment are defined by several IEEE standards, with IEEE 802.16-2004 detailing fixed WiMAX system requirements and IEEE 802.16e-2005 covering WiMAX requirements for subscribers on the move.

For the IEEE 802.16-2004 fixed standard, licensed frequencies fall around 2.5 and 3.5 GHz while unlicensed frequencies are around 5.2 and 5.8 GHz. Mobile WiMAX, which is formulated to allow roaming of mobile units with handoffs from one base station to another, as defined in IEEE 802.16e, is designed to operate in the 2.3-, 2.5-, 3.3-, and 3.4-to-3.8-GHz frequency bands. Fixed WiMAX is based on orthogonal frequency-division-multiplex (OFDM) technology while mobile WiMAX uses both OFDM and orthogonal frequency-division-multiple-access (OFDMA) technology.

WiMAX employs a number of different modulation formats, including BPSK, QPSK, 16QAM, and 64QAM, switching the modulation format according to the strength of the received signal. By this adaptive modulation approach, transmission rates as fast as 75 Mb/s are possible. WiMAX signals, which have a frame structure with preamble and data sections, are sent in bursts. The noise-like signals are converted from the frequency domain to the time domain by means of Fast Fourier Transforms (FFTs), with as large as 2048-point FFTs used for WiMAX signals.

The MAX2837 single-chip transceiver (Fig. 1) incorporates both the RF-to-baseband receive signal path and the baseband-to-RF transmit signal path. It is based on an advanced SiGe BiCMOS process that features low power consumption but also the low noise figures needed for good receiver sensitivity. The transceiver features a direct-conversion, zero-intermediate-frequency (zero-IF) architecture that includes a fast-switching sigma-delta fractional-N phase-locked-loop (PLL) frequency synthesizer with 20-Hz step size and low phase noise. The zero-IF architecture exhibits extremely low carrier leakage into adjacent channels without the need for frequency correction. In contrast to traditional superheterodyne transceiver architectures that employ an external surface-acoustic-wave (SAW) filter for channel filter, the MAX2837 integrates monolithic baseband filters for its receiver and transmitter sections.

The wireless transceiver IC offers programmable channel bandwidth filters that match the channel-bandwidth requirements of the WiMAX standards, at bandwidths from 1.75 to 28.0 MHz. The on-chip channel filter can be programmed for all of the WiMAX bandwidths: 1.75, 2.5, 3.5, 5.5, 6, 7, 7. 8.75, 10, 14, 15, 20, and 28 MHz. An SPI™-compatible interface enables electronic programming of channel filtering which allows one basic radio design to support virtually all geographic markets where channel bandwidths differ according to each country’s spectrum regulations.

In support of the synthesizer, the well-equipped IC also includes a voltage-controlled oscillator (VCO) and a digitally controlled crystal oscillator (DCXO). The DCXO is designed for use with a low-cost crystal resonator, to save the cost of a temperature-compensated crystal oscillator (TCXO). With the synthesizer, inclusion of the DCXO allows the implementation of digital automatic frequency control (AFC) with the MAX2837 for improved system range while maintaining the minimum bit-error-rate (BER) performance required by the WiMAX standards (10–6 BER).

The on-chip DCXO enables the use of low-cost crystals with ±25 PPM or higher frequency tolerance over temperature. The DCXO’s tuning step is less than 1 PPM and is monotonic in nature, allowing the development of an AFC algorithm that maintains total frequency error within 1 PPM over the full operating temperature range.

The fractional-N PLL synthesizer even enables the use of an inexpensive non-tuned crystal. Instead of digitally tuning the load capacitance of the crystal oscillator, the RF frequency can be adjusted in 25-Hz steps to remove the frequency error. The sampling clock error (and symbol timing error) that is related to the reference clock (the crystal oscillator) can be corrected digitally in the WiMAX modem’s interpolation of the I/Q samples to account for drift in the case of the sampling clock.

The MAX2837, which includes a baseband/control interface, features circuitry for on-chip DC-offset cancellation, in-phase/quadrature (I/Q) error detection, and carrier-leakage detection. Very little additional circuitry is needed for form a complete WiMAX RF transceiver front end, except for an RF bandpass filter, the crystal resonator for the reference oscillator, an RF switch, a power amplifier for the transmitter, and a small number of passive components.

The receiver portion of the MAX2837 features a an integrated low-noise amplifier (LNA) with noise figure of 2.3 dB with input third-order intercept point (IIP3) that is adjustable from –8 to –11 dBm at the maximum gain setting. The peak-to-peak gain variation over the full RF input range is typically 0.8 dB (Fig. 2) while the RF input return loss is typically 13 dB at all setting of the integrated LNA. The total voltage gain is typically 99 dB, which is comprised of adjustable RF and baseband gain. The RF gain steps are 8, 16, and 32 dB while the baseband gain range is typically 62 dB (in 2-dB minimum steps).

Page Title

The settling time for any change in RF or baseband gain to within ±1 dB of the new steady-state value is typically only 0.2 µs. The settling time for any change in RF or baseband gain to within ±0.1 dB of the new steady-state value is typically only 0.2 µs. The double-sideband (DSB) noise figure is a function of the total RF gain. With 50-dB voltage gain and maximum RF gain of 8 dB, the DSB noise figure is typically 5.5 dB. The DSB with 45-dB voltage gain and maximum RF gain of 16 dB is typically 17 dB. The DSB noise figure with 15-dB voltage gain and maximum RF gain of 32 dB is typically 27 dB (Fig. 3). The receiver error-vector-magnitude (EVM) performance is generally better than 2.5 percent for a –35 dBm input signal level with 64QAM (Fig. 4). Measurements were made at 2.5 GHz with a 10-MHz WiMAX (OFDM) signal channel.

The transmitter section of the MAX2837 provides 12 dB typical voltage gain across the 2.3-to-2.7-GHz operating range, with typically 0 dBm output power across any 200-MHz band within the full operating range and typically more than –2 dBm across the full operating range (Fig. 5). The transmitter achieves better than 45 dB sideband suppression. The typical EVM performance while also meeting a –70-dBr spectral mask is better than 1.8 percent for –36-dBm 64QAM signals (Fig. 6). The transmitter phase noise is typically about –90 dBc/Hz offset 1 kHz from the carrier, about –100 dBc/Hz offset 100 kHz from the carrier, and –120 dBc/Hz offset 1 MHz from the carrier. The transmitter circuitry provides a total of 47-dB gain control in 1-dB steps, with 0.5-dB step control available for power-amplifier calibration. The high linearity of the MAX2738’s transmitter output allows the transceiver to be used in WiMAX micro or pico base stations as well as in subscriber customer premises equipment (CPE) and personal-computer (PC) cards for WiMAX.

The single-chip broadband transceiver is suitable for both fixed and mobile WiMAX applications as well as the Korean equivalent of mobile WiMAX, WiBro (for wireless broadband). The transceiver can also be used for proprietary OFDM wireless broadband radio systems operating from 2.3 to 2.7 GHz. The chip is supplied in a 48-pin QFN package that measures just 6 X 6 X 0.8 mm. It is rated for operating temperatures from –40 to +85ºC.

The MAX2837 (Fig. 7) features a power-down mode in which all circuit blocks are powered down except the four-wire serial bus and its internal programming registers. The current draw in this mode is a mere 100 µA. In a standby mode, the frequency synthesizer block is enabled while the rest of the chip is powered down. The single-chip WiMAX transceiver is designed for use with a single +2.7 to +3.6 VDC supply. The current consumption in shutdown mode is only 10 µA. Compared to current WiMAX radio IC solutions, the receiver portion of the MAX2837 uses about one-quarter the power of most existing solutions, while the transmitter uses only about one-third of most other solutions. The receiver current consumption for a –10-dBm input third-order intercept point is 90 mA. The transmit current consumption for a 64QAM signal at 0 dBm is 135 mA. It drops to 125 mA for a 16QAM signal at 0 dBm and 115 mA for a 16QAM signal at –3 dBm.

On the heels of the MAX2837, the company plans to introduce the MAX2838 single-chip, zero-IF transceiver for fixed, portable, and mobile WiMAX applications from 3.3 to 3.9 GHz. As with the MAX2837, the higher-frequency transceiver provides full WiMAX functionality without the need of a SAW filter. In addition, it supports multiple-input, multiple-output (MIMO) functionality in a master/slave configuration. The MIMI approach uses multiple transmit and/or receiver antennas to combat the effects of multipath distortion and signal fading. As with the MAX2837, the model MAX2838 has been designed into a reference circuit, model RD142, for applications at 3.5 GHz.

For evaluation purposes, and to help designers shrink product-development time, the company has developed several reference designs based on the MAX2837, including the model RD087 reference design for WiBro applications and the model RD0117 reference design for dual-mode (fixed OFDM and mobile OFDMA) WiMAX applications. These are full radio designs that can be adapted for other applications and are supported with digital layout files and programming software. P&A: $8.38 and up (1000 qty.) (MAX2837 only). Maxim Integrated Products, 120 San Gabriel Dr., Sunnyvale, CA 94086; (408) 737-7600, Internet:

Sponsored Recommendations

UHF to mmWave Cavity Filter Solutions

April 12, 2024
Cavity filters achieve much higher Q, steeper rejection skirts, and higher power handling than other filter technologies, such as ceramic resonator filters, and are utilized where...

Wideband MMIC Variable Gain Amplifier

April 12, 2024
The PVGA-273+ low noise, variable gain MMIC amplifier features an NF of 2.6 dB, 13.9 dB gain, +15 dBm P1dB, and +29 dBm OIP3. This VGA affords a gain control range of 30 dB with...

Fast-Switching GaAs Switches Are a High-Performance, Low-Cost Alternative to SOI

April 12, 2024
While many MMIC switch designs have gravitated toward Silicon-on-Insulator (SOI) technology due to its ability to achieve fast switching, high power handling and wide bandwidths...

Request a free Micro 3D Printed sample part

April 11, 2024
The best way to understand the part quality we can achieve is by seeing it first-hand. Request a free 3D printed high-precision sample part.