SDR Architecture: Understanding Software Defined Radio

This article delves into the architecture of Software Defined Radio (SDR), outlining its elements and the benefits it offers. SDR employs software-configured transmitters and receivers, marking a significant shift in radio technology.

Introduction to Software Defined Radio (SDR)

As the name implies, Software Defined Radio is a wireless communication system that operates at radio frequencies and is highly configurable through software. Instead of relying heavily on dedicated hardware components, SDR leverages software to define and control many aspects of the radio’s functionality.

In SDR, hardware components traditionally found in analog radios are often replaced or augmented by software implementations. This software is typically deployed on Field Programmable Gate Arrays (FPGAs) or Digital Signal Processors (DSPs) within the wireless communication chain.

The following components are often configured in SDR:

• Mixers (used in upconversion and downconversion)
• Amplifiers
• Modulator and demodulator modules
• RF detector
• Filters (BPF, LPF) using FIR filter concepts

Modern SDR implementations often extend this concept to encompass the entire PHY layer and RF chain within the software.

SDR-based designs are increasingly prevalent in existing and next-generation wireless technologies and standards, including WLAN, Mobile-WiMAX, 4G LTE, LTE-Advanced, and 5G NR (New Radio).

SDR Transmitter Architecture

Figure 1: Transmitter part of SDR architecture

Figure 1 illustrates the transmitter portion of a typical SDR architecture. It comprises a DSP, a Digital Up Converter (DUC), a D/A converter, an analog RF Up Converter, and a power amplifier.

The digital baseband processing is implemented within the DSP, generating I/Q data according to the specific transmitter requirements. This data is then digitally up-converted using the DUC, employing a digital Local Oscillator (LO) and a digital mixer.

The resulting digital IF (Intermediate Frequency) samples are converted to an analog IF signal using a D/A converter. This analog IF signal is then up-converted to an analog RF (Radio Frequency) signal via the RF up-converter.

Finally, the RF signal is amplified by a power amplifier before being transmitted over the air using an appropriate antenna, selected based on the desired system operating frequency.

SDR Receiver Architecture: A Comparison

Let’s contrast the architectures of traditional analog radio receivers with those of software-defined radio receivers.

Figure 2: Analog radio receiver

As shown in Figure 2, in a traditional analog radio receiver, the RF signal is first amplified using an RF Amplifier. The amplified signal is then fed to an RF mixer for RF down-conversion. This is accomplished by mixing the input amplified RF signal with a locally generated LO signal.

An IF signal is extracted and amplified at the output of the RF mixer. Typical IF center frequencies used are 455 kHz for AM and 10.7 MHz for FM broadcasting receivers. This amplified IF signal is then demodulated and passed to an audio amplifier.

An envelope detector is used for amplitude demodulation, while a frequency discriminator is used for frequency demodulation.

The RF mixer performs the critical function of converting the RF signal into the IF signal.

Figure 3: SDR Receiver architecture

Figure 3 depicts the architecture of an SDR receiver. The first module is an RF tuner, which converts the RF signal into an amplified IF signal. It essentially replaces the first three modules (RF amplifier, mixer, and IF amplifier) found in the traditional analog receiver shown in Figure 2.

The following modules are implemented after the RF tuner module:

• A/D converter: Converts the analog IF signal into digital IF samples.
• Digital Down Conversion (DDC): The digital samples are passed to the DDC, which converts the digital IF samples into digital baseband samples (referred to as I/Q data). The DDC consists of a digital mixer, a digital Local Oscillator (LO), and a low-pass FIR filter.
• DSP Chip: The digital baseband samples are then processed by a DSP chip. Algorithms are implemented on this chip to perform tasks such as demodulation, decoding, and any other required signal processing.

This digital implementation forms the core of SDR. An FPGA is often used instead of a DSP in these architectures to enable faster signal processing algorithms.

The flexibility afforded by implementing the baseband processing chain in software on a DSP/FPGA allows for real-time correction of baseband and RF impairments present in the I/Q data through the use of sophisticated algorithms.

Typically, algorithms such as DC offset correction, I/Q gain and phase imbalance correction, and time, frequency, and channel impairment correction are implemented in an SDR receiver.

Benefits of Software Defined Radio Architecture

The SDR architecture offers several key advantages:

• Easy Upgradability: SDR enables seamless upgrades to different software versions during the evolution of standards like LTE or LTE-Advanced. This allows product manufacturers to meet time-to-market requirements and client needs more effectively.
• Cost Savings: Significant cost savings are achieved due to the reduced development time compared to traditional analog systems. Changes required during testing can be readily implemented in software, eliminating the need for extensive hardware modifications.
• Flexibility for Experimentation: SDR facilitates easy experimentation with new ideas even while the system is operational, promoting innovation and rapid prototyping.