Traditional wireless communication and radar systems suffer from a fundamental architectural bottleneck: their signal processing algorithms are tightly coupled to specific, fixed-function hardware components like ASICs and FPGAs. As noted by Khaled Ben Letaief et al. in The Roadmap to 6G: AI Empowered Wireless Networks, this tight coupling forces costly, global system redesigns every time a hardware component is upgraded or a new wireless standard emerges. This lack of abstraction limits modularity, making it difficult to deploy complex radio functions across diverse hardware platforms without extensive redesign (Jiaxing Wang et al. in Parallel FM Signal Demodulation System based on Software Radio Platform).
To break this rigidity, a major research direction focuses on decoupling signal processing from fixed hardware by abstracting the physical (PHY) and media access control (MAC) layers into modular software components (Mao Yang et al. in Software-Defined and Virtualized Future Mobile and Wireless Networks: A Survey; Mao Yang et al. in Cross-Layer Software-Defined 5G Network). By defining low-level radio functions entirely in software, the physical layer transitions into a generic, programmable resource that can dynamically adapt to real-time hardware capabilities and evolving protocols without physical hardware replacement.
Overcoming the Throughput Bottleneck
Historically, physical layer tasks were relegated to dedicated hardware because general-purpose processors struggled with the high-throughput, serial nature of operations like channel decoding (Jianwen Chen et al. in Open Wireless System Cloud: An Architecture for Future Wireless Communications System). Software-defined architectures solve this by shifting computational tasks to highly parallel, general-purpose hardware environments, such as consumer-grade GPUs, Vector DSPs, and scalable many-core processors (Rüdiger Göbl et al. in SUPRA: open-source software-defined ultrasound processing for real-time applications; Chuan Zhang et al. in Guest Editorial Circuits, Systems, and Algorithms for Beyond 5G and Toward 6G; Marco Bertuletti et al. in Efficient Parallelization of 5G-PUSCH on a Scalable RISC-V Many-core Processor).
This software-defined acceleration environment allows systems to dynamically balance latency and throughput. For example, Chuan Zhang et al. demonstrate that utilizing GPUs and vector digital signal processors for channel estimation and error-correction decoding provides the high throughput required for next-generation standards while retaining the flexibility to update algorithms via software. Similarly, Qimin Chen et al. in PMF-FFT Spread Spectrum Signal Acquisition Algorithm Based on GPU Parallelization Design utilize GPU parallelization to handle high-rate spread-spectrum signal acquisition in real time, enabling immediate software adjustments to parameters like FFT points and integration periods.
Mechanisms of Runtime Reconfigurability
When signal processing logic is decoupled from the underlying hardware via a standardized software architecture, radio functions can be managed as independent software applications (Jiaxing Wang et al. in Parallel FM Signal Demodulation System based on Software Radio Platform). This abstraction enables several key capabilities:
- Arbitrary Waveform Generation: Instead of relying on fixed-function direct digital synthesis chips, transitioning to an FPGA-driven digital-to-analog converter architecture allows software to generate arbitrary waveforms (Guobin Yang et al. in Design of the transceiver for a new type HF/VHF ionospheric sounding system). This allows researchers to implement complex pulse shaping (such as Gaussian or phase-coded sequences) to optimize spectral efficiency.
- Software-Based Modulation: Modern general-purpose processors can effectively execute complex modulation schemes, such as Gaussian Minimum Shift Keying, entirely in software (as demonstrated by the authors of Implementation of Cognitive Radio Network Testbed for Multimedia Communications). This allows for rapid reconfiguration of a radio's operational parameters on commodity hardware without custom integrated circuits (Muneer M. Al-Zu'bi in Experimental Outdoor Performance Evaluation of TVWS Narrowband Data Communication using SDR Platform).
- Dynamic Protocol Switching: Transceivers can dynamically switch between entirely different operational modes mid-mission. Malong Ke et al. in Next-Generation URLLC with Massive Devices: A Unified Semi-Blind Detection Framework for Sourced and Unsourced Random Access present a unified framework that switches between sourced and unsourced random access modes via software, while Marco Maso et al. in Reconfigurable cognitive transceiver for opportunistic networks show how a software-defined radio can dynamically switch to an interference-aligned transmission mode to exploit spectrum opportunities.
Cross-Domain Applications
The utility of programmable physical layers spans multiple domains, from deep space to medical imaging. In satellite communications, Radhika Radhakrishnan et al. in Survey of Inter-satellite Communication for Small Satellite Systems: Physical Layer to Network Layer View and Steven Adams et al. in SSETI ESEO New Zealand Ground Station Project highlight how software-defined radios allow satellite swarms and ground stations to adjust frequencies and protocols via software updates, ensuring interoperability across heterogeneous missions.
In radar and medical imaging, Youngseok Bae et al. in Field Experiment of Photonic Radar for Low-RCS Target Detection and High-Resolution Image Acquisition and Rüdiger Göbl et al. in SUPRA: open-source software-defined ultrasound processing for real-time applications demonstrate that moving signal parameters and beamforming out of proprietary hardware into software-defined pipelines enables real-time operational versatility and enhances research reproducibility.
Even at the hardware boundary, Chao Ma in Current State and Advanced Architectures of Doherty Power Amplifiers shows that transitioning from analog to digital control architectures allows software to precisely manage amplifier turn-on points and gain, reducing the reliance on frequency-specific physical components. Ultimately, by decomposing monolithic radio stacks into modular, software-defined components, a single hardware device can support multiple protocols and adapt to changing environments on the fly (as detailed in Enabling Virtual Radio Functions on Software Defined Radio for Future Wireless Networks).
Key Foundational Work
Mapping a Sensor Interface and a Reconfigurable Communication System to an FPGA Core (2005) — contributed 2 concept elements to this direction.
Addressing the System-on-a-Chip Interconnect Woes Through Communication-Based Design (2001) — contributed 2 concept elements to this direction.