Front End Fits SDR Picocells

July 16, 2010
This compact RF front end meets the needs of multi-band, multimode transmit and receive requirements in software-defined-radio (SDR) picocell stations.

Software-defined-radio (SDR) approaches are being used in compact cellular base stations, including in picocellular systems serving small areas such as onboard commercial aircraft. Wireless communications systems onboard an airplane must provide high-quality audio, video, and data services conveniently but without interference with the aircraft's navigation systems. As a result, a picocell solution must perform with sufficient suppression of out-ofband spurious signal products along with the bandwidth needed for multiband, multimode operation. The RF front-end system developed here is compact but supports channel bandwidths to 20 MHz. It provides multiple wireless access services for passengers onboard an airplane. Experimental results show excellent linearity and phase-noise results with outstanding error-vectormagnitude (EVM) performance (less than 1 percent) when operating under orthogonal-frequency-division-multipled (OFDM) 64-state quadratureamplitude- modulation (64QAM) signal conditions and a 20-MHz channel bandwidth.

Picocells are similar to W-Fi access points. As part of a cellular wireless communications network, a picocell base station is typically a small, simple, low-cost unit that connects to a base station controller.1 It allows direct connection to the Internet without a base station controller (BSC) infrastructure. In this report, the picocell subsystem is based on a softwaredefined- radio (SDR) configuration with 50-MHz-wide digital intermediate- frequency (IF) architecture. It is designed for simplicity and reliability with the capability of handling broadband frequency ranges and multiple wireless communication standards.2,3Figure 1 shows a basic block diagram of the picocell transceiver subsystem, with five basic components (frontend module, receiver, transmitter, local oscillator, and interface circuitry) that combine to achieve system requirements of 60-dB dynamic range with 90-dB receiver gain (and 30-dB receiver dynamic range) and 30-dB transmitter gain.

A complex programmable logic device (CPLD) chip is used as the major component of the interface. By invoking corrected order and timing, the CPLD coordinates a number of related devices for functions such as automatic power control (APC), automatic- gain-control (AGC), automatic frequency control (AFC), and switches to accomplish required system service requests. An additional microcontroller unit (MCU) is used to access control signals for the phase lock loop (PLL). Table 1 shows the timing diagram and bit map for the baseband control signals.

The front-end module provides selection between transmit and receive functions as well as 2.4- and 5.8-GHz Industrial-Scientific-Medical (ISM) bands. To minimize the effects of path losses and multipath in order to process low signal levels at the receive antenna, a series of low-noise amplifiers (LNAs), preselector filters, and a single- pole, double-throw (SPDT) switch are implemented to raise the level of the received signal strength (RSS) and to separate desired RF signals from uncorrelated received signals.

The IF portion of the receiver is comprised of IF amplifiers, a frequency downconverter, an IF surface- acoustic-wave (SAW) filter and a pair of AGC circuits controlled by a baseband processing unit, a model A2A1A0(000)D6-D0 from Maxim Integrated Products, to achieve a dynamic range of 60 dB. The SAW filter features a 20-MHz bandwidth from 364 to 384 MHz with 50-dB adjacent-channel suppression.

In the local oscillator (LO) subsystem, which is based on an SDR architecture, frequency synthesizers are used to implement the frequency bands required for demodulating and modulating signals for the receiver and transmitter, respectively. Controlled by the model A2A1A0(110)D7-D0 baseband processing unit, two synthesizers make use of a fractional-N structure for low in-band phase noise with high-resolution channel spacing in accordance with the required wireless channel configurations.

Time-division-duplex (TDD) operation makes use of different time slots to transfer different channels of information with reduced crosstalk between transmit and receive channels. With a digital IF architecture, it is possible to correct in-phase/quadrature (I/Q) gain and phase unbalance typical of analog component manufacturing tolerances in the digital domain. The SDR picocell transceiver was evaluated by means of a model E4445A spectrum analyzer, model E4438C signal generator, and model 89600 software from Agilent Technologies. The SDR picocell transceiver was evaluated by testing the front-end module with a data link formed of the receiver/transmitter/LO setup.

The EVM of the front-end module was measured for different transmit frequencies under standard IEEE 802.11g transmit signal conditions. With a standard IEEE 802.11g waveform based on OFDM and 64QAM and data rate of 54 Mb/s, the EVM performance is 2 to 4 percent with gain of 24 to 26 dB. For IEEE 802.11a/b transmit conditions, the EVM is about 4 to 6 percent and the gain is 23 to 25 dB. For IEEE 802.11g receive conditions, the insertion gain is 8 dB and the noise figure is 4.9 dB. For IEEE 802.11a/b receive conditions, the insertion gain is 2.3 dB and the noise figure is 2.3 dB.

A typical wireless communication system requires phase noise ranging from -80 to -120 dBc/Hz.4-6Table 2 shows that the phase noise of the LO subsystem used in the SDR picocell system ranges from -80 to -110 dBc/ Hz with sufficient power to drive the mixers of the transmit/receive system.

The transmit/receive subsystem for the SDR picocell was also evaluated for amplitude flatness and output power. The amplitude flatness was better than 0.8 dB in receive mode, with sufficient IF output power (to 0 dBm) to meet baseband requirements. In transmit mode, the amplitude flatness was better than 0.6 dB with more than +20 dBm output power, meeting the needs of the picocell subsystem. According to experiments, the dynamic range, which defines the range of the input RF power from the sensitivity threshold to the maximum detected signal, can reach 60 dB in receive mode and 30 dB in transmit mode.

EVM is a standard measure of performance for many wireless communications systems. It captures the phase and amplitude imbalance, gain compression, phase noise, and I/Q mismatch. Signal distortion caused by these factors can be easily demonstrated though the EVM constellation diagrams. The relationship between EVM and signal-to-noise ratio (SNR) can be expressed by Eq. 1:

EVM = 1/L(SNR)0.5 (1)

In modern wireless communication systems, quadrature-phase-shiftkeying (QPSK) modulation is typically used for digital-video-broadcast (DVB) satellite transmission systems, while QAM is often used for cabletelevision systems and coded orthogonal frequency division multiplex (COFDM) modulation has commonly been used for terrestrial transmission systems.1,7 The EVM of the transmit/ receive picocell subsystem was evaluated with different modulation formats, such as QPSK and single-carrier 64-state QAM (64QAM) as shown in Figure 3 and Figure 4. The EVM performance of the subsystem is also shown in Table 3 for OFDM-64QAM signals at receiver input power levels from -90 to -30 dBm and at 15- and 20-MHz bandwidths.

The measured results show that the picocellular transmit/receive subsystem can support different channel bandwidths (5, 10, and 20 MHz) as well as multiple modes of operation. Using SDR technology, the channel bandwidth of the system can be varied for use with different wireless access systems. As shown by the constellation diagrams for the transmit/receive subsystem in Fig. 5, it displays outstanding EVM performance of better than 1 percent for OFDM-64QAM signals with a 20-MHz channel bandwidth. The specification SNR calculated by Eq. 1 is better than 40 dB. A prototype of the multimode, multiband RF transceiver was fabricated on six-layer printed-circuit boards (PCBs) measuring just 220 x 110 x 1.5 mm (Fig. 6).

ACKNOWLEDGMENTS
The authors would like to acknowledge that this work was supported in part by the Natural Science Foundation of China (NSFC) under Grant 60621002 and 60702027, by the National High-Tech Project under Grant numbers 2008AA01Z223 and 2009AA011503, and by Boeing Phantom Works.

REFERENCES
1. Ulrich L. Rohde and David P. Newkirk, RF Microwave Circuit Design for Wireless Applications, Wiley, New York, 2000.
2. Wei Hong, Zhengbo Jiang, Jin Liu, and Ling Tian, "TDSCDMA Receiver Captures Multiple Channels," Microwaves & RF, July 2009.
3. K. Chang, Microwave Solid-State Circuits and Applications, Wiley, New York, 1994.
4. P. Vizmuller, RF Design Guide, Artech House, Boston, 1995.
5. U. L. Rhode, Microwave and Wireless Synthesizers, Wiley, New York, 1997.
6. Kai Chang, RF and Microwave Wireless Systems, Wiley, New York, 2000.
7. Ian Oppermann, Matti Ha'ma'la'inen, and Jari Iinatti, UWB Theory and Applications, Wiley, New York, 2004.

Sponsored Recommendations

What is a Vector Network Analyzer?

Feb. 19, 2024
For engineers that are outside of the RF and Microwave industry and work in industries like the healthcare and medical fields, food and beverage, agriculture, and beyond, a vector...

Time Domain Gating Measurements with a VNA Webinar

Feb. 19, 2024
Time domain gating is a very useful tool in vector network analysis. If a measurement can be made over sufficient bandwidth, there will be ample resolution in the time domain ...

Webinar: Introduction to OTA Measurement for mmWave and Sub-THz

Feb. 19, 2024
Join Jeanmarc Laurent, a leading expert from MilliBox, for an exclusive live webinar showcasing a complete Over-the-Air (OTA) testing system setup. In this immersive session, ...

Using a CMT VNA with Socket Server

Feb. 19, 2024
This application note describes use of a software application CMT Socket Server which is distributed and supported by Aphena Ltd. Please email [email protected] regarding purchase...