Software Defined Radio SDR Architecture Informational

What is a software defined radio and how does it differ from a traditional hardware radio?

A software defined radio (SDR) is a radio communication system where components traditionally implemented in hardware (such as filters, modulators, demodulators, amplifiers, and detectors) are instead implemented as software algorithms running on a general-purpose processor, digital signal processor (DSP), or field-programmable gate array (FPGA). In an ideal SDR, the antenna connects directly to a high-speed analog-to-digital converter (ADC) that digitizes the entire RF spectrum of interest, and all subsequent signal processing is performed digitally. In practice, an analog RF front end (LNA, bandpass filter, and possibly a frequency downconverter) is still required before the ADC due to dynamic range and sampling rate limitations. The fundamental difference from a traditional hardware radio is flexibility: a hardware radio's operating frequency, bandwidth, modulation format, and protocol are fixed by its physical circuit design, while an SDR can change all of these parameters through software reconfiguration without any hardware modification. This means a single SDR platform can be reconfigured to receive AM, FM, GSM, LTE, Wi-Fi, GPS, ADS-B, or any other signal format simply by loading different software. SDR also enables rapid development and prototyping of new waveforms, cognitive radio capabilities (dynamic spectrum access), and simultaneous multi-mode operation.

Category: Software Defined Radio

Updated: April 2026

Product Tie-In: SDR Platforms, ADCs, FPGAs

Software Defined Radio Architecture and Principles

SDR represents a paradigm shift in radio design, moving intelligence from fixed hardware to flexible software. This shift has been enabled by advances in high-speed ADCs, FPGAs, and general-purpose processors that can handle the processing demands of real-time RF signal processing.

SDR Architecture Tiers

Ideal SDR: ADC directly at the antenna. All processing in software. Not yet practical for most applications due to ADC limitations
Direct sampling SDR: ADC samples at RF or IF frequency using undersampling or direct conversion. Minimal analog hardware. Examples: Ettus USRP B200 (direct conversion), KiwiSDR (direct sampling of HF)
Superheterodyne SDR: Traditional analog front end with frequency conversion to a lower IF, then ADC. Provides best dynamic range and selectivity. Used in high-performance applications
Digital IF SDR: Analog downconversion to a moderate IF (10-200 MHz), then high-speed ADC. Digital down-conversion (DDC) in FPGA extracts the desired channel. Common architecture for wideband receivers

Key Advantages Over Hardware Radio

Reconfigurability (change waveform, frequency, bandwidth via software), multi-mode operation (process multiple signals simultaneously), rapid prototyping (new modulation schemes tested in software), future-proofing (hardware stays the same, capabilities added via software updates), and cognitive radio capability (adapt transmission parameters based on spectrum sensing).

Processing Chain

A typical SDR receive chain: antenna, LNA, bandpass filter, mixer/ADC, digital down-conversion (DDC) in FPGA, channelization, demodulation in DSP/CPU, and application-layer processing. Each stage that would be hardware in a traditional radio is replaced by a software algorithm, with the division between FPGA and CPU processing depending on latency and throughput requirements.

SDR Sampling and Processing Requirements

ADC sampling requirement (Nyquist): f_s > 2 x BW_signal
Direct sampling of 6 GHz band: f_s > 12 GSa/s (challenging)
Digital IF at 100 MHz: f_s > 200 MSa/s (practical with modern ADCs)
Processing load: MFLOPS ~ BW_signal x bits_per_sample x operations_per_sample

Common Questions

Frequently Asked Questions

Is SDR the same as a ham radio scanner?

No, although there is overlap. SDR is a fundamental architecture approach where signal processing is done in software rather than hardware. Ham radio scanners and receivers that use SDR technology (like RTL-SDR dongles) are specific consumer products. Professional SDR platforms (Ettus USRP, Analog Devices ADALM-Pluto, National Instruments) are used for everything from 5G research to military communications to radio astronomy.

Can an SDR transmit as well as receive?

Yes. Full-duplex SDR platforms include both ADC (receive) and DAC (transmit) paths, allowing the SDR to generate arbitrary waveforms for transmission. The transmit chain mirrors the receive chain in reverse: software generates the digital baseband signal, the DAC converts it to analog, and an RF front end upconverts and amplifies it for transmission. Regulatory licensing is required for transmission.

What limits the performance of an SDR compared to a purpose-built radio?

SDR limitations include: ADC dynamic range (14-16 bits provides 80-96 dB SFDR, less than the best analog receivers), processing latency (software processing adds delay compared to dedicated hardware), power consumption (general-purpose processors consume more power than dedicated ASICs for the same function), and maximum instantaneous bandwidth (limited by ADC sample rate and processing throughput). Purpose-built radios can be optimized for a specific application to exceed SDR performance in that specific application.

Keep Reading