CC Blog Projects Research & Design Hub Tech Trends

Software-Defined Radios

Figure 1 Conventional T&M devices (Source: Keysight)

Their Use in Testing & Measurement Applications

Advances in the radio frequency (RF) industry require extensive and reliable testing to ensure smooth, successful operation. Software-defined radios (SDRs) are integral in testing and measurement operations, and provide a versatile platform for testing multiple applications. In this article, we discuss why SDRs are replacing conventional testing and measurement hardware, and their importance in a variety of applications.

  • How are software-defined radios used in testing and measurement?
  • What are current trends in the radio frequency industry?
  • What are some uses of software-defined radios?
  • Software-defined radios
  • Cyan SDR

In the age of wireless information, new devices are being developed in the radio frequency (RF) industry every day. Testing and measurement (T&M) is imperative, to ensure that this massive wave of new devices and protocols can be properly integrated into network infrastructures, with reliable and effective operation, while also accounting for interoperability. 

T&M is essential in every sector of the RF industry, including antenna design, radar systems, and entire mobile networks. In this context, software-defined radios (SDRs) play a major role in T&M design, because they are flexible, reconfigurable, modular, and capable of being designed for several size, weight, and power (SWaP) requirements. The same SDR can be programmed to generate waveforms, measure a signal spectrum in real time, and simulate entire RF systems–with minimum hardware adaptations. 

In this article, we will discuss how SDRs are implemented in several T&M applications. First, we review how the T&M methods and techniques changed over the years, and the main challenges faced by testing systems based in conventional analog electronics. Second, we will discuss the basic concepts of SDRs and how they can benefit T&M applications, including the integration of high-quality radio front ends (RFEs), digital signal processing (DSP) software suites for various measurement protocols, and continuous integration/continuous deployment (CI/CD) implementation. Finally, we present three use cases of SDRs in T&M: antenna measurements, 5G testbeds, and radar-ranging tests. 

RF TESTING: THEN AND NOW

Electromagnetic signals are invisible to the human eye, so measurement equipment is essential in almost every step of the design chain and product validation. In the electromagnetic systems, spectrum analyzers are always required, with high-frequency signal generators being used to produce controlled signals that emulate several RF conditions, including normal operation, noise, environmental interference, and jamming. Power meters are fundamental to assess signal strength, signal-to-noise ratio (SNR), range, and consumption. Waveform analyzers can evaluate distortion and linearity, which are important parameters for device performance. Auxiliary peripherals required for radio operation, including specialized cables (SMA and BNA, for instance), RF couplers, circulators, and connectors, are also crucial for T&M. Most tests must be conducted under controlled environmental conditions, so anechoic (sound-deadening) chambers and isolated laboratories are the standards for several T&M protocols.

In the RF industry, T&M is mandatory for both design and product certification. The most basic set of T&M requirements concerns electromagnetic interference (EMI) and electromagnetic compatibility (EMC). RF products are strong sources of electromagnetic waves, so it is crucial to ensure that the radiation levels do not interfere with nearby devices. These assessments can be performed using spectrum analyzers, signal analyzers, and vector network analyzers, which evaluate the electromagnetic radiation levels and check if they are within the regulatory standards. EMC tests also require high-frequency signal generators, applied to verify if the device can withstand the environmental EMI levels without failure. Signal generators are also used to simulate RF signals to verify device functionality, including modulation/demodulation schemes, jamming, and network protocols. 

Antennas are the fundamental source and sink of electromagnetic waves; thus, it is crucial to design, select, and test them properly. There are several T&M protocols to assess antenna performance, including near field, far field, and standing-wave ratio measurements. Short wave diathermy is a frequency-dependent parameter. It describes the level of impedance matching between the antenna and the receiver, which defines the amount of signal loss at the cables and connectors. At the system-analysis level, scattering parameters (S- parameters) are convenient metrics to evaluate system behavior, and are especially useful in microwave devices. T&M is also used to detect communication errors, including bit-error-rate analysis, which is necessary to evaluate the reliability of the system at the network level.

With the fast development of novel RF devices, it is becoming more and more difficult for the T&M industry to keep up with so many network protocols, technological upgrades, and communication schemes. Instantaneous bandwidth capturing, for instance, is much more difficult today, as use of higher frequencies entails larger bandwidths. The use of high frequencies also increases the burden over T&M signal generators, which must operate with much faster signal transmission. Another example is dealing with the large amounts of RF data that are constantly being streamed through high-throughput devices, which require state-of-the-art host connection that are difficult to implement using conventional T&M techniques. 

— ADVERTISMENT—

Advertise Here

The most problematic challenge, however, is the lack of flexibility of traditional T&M equipment, which cannot be easily reconfigured to work with the ever-evolving RF standards and protocols of today’s industry. Moreover, each T&M process must be adjusted to match the RF sensitivities of the devices under test, which is difficult in hardware-based equipment. Examples of conventional T&M devices are shown in Figure 1.

Figure 1 
Conventional T&M devices (Source: Keysight)
Figure 1
Conventional T&M devices (Source: Keysight)

With the high levels of system “softwarization”—using a software solution instead of traditional hardware—in modern radio systems, the T&M industry is expected to follow the same trend. Software-defined radios are currently being implemented in a variety of T&M scenarios and applications, ranging from simple antenna testing to complex network verification. SDRs can replace or enhance most of the T&M equipment in the industry, performing several different tasks at the same time, with almost no hardware modification. In the following section we discuss the basic concepts of SDRs and how they can benefit the T&M sector. 

RUNDOWN OF SDR FOR T&M

Software-defined radios were developed to shift most of the radio operations and signal processing load from the analog side to the digital domain. They reduced the amount of hardware required to only the essential functions, including amplification, filtering, mixing, and impedance matching. The typical SDR is divided into two main functional blocks: the radio front-end and the digital back end (Figure 2). The RFE performs all the receive (Rx) and transmit (Tx) functions of the system over a wide tuning range, which is typically between 0-18GHz, but it can reach up to 40GHz in the best-performance SDRs, such as Cyan (Per Vices Corp.). 

Figure 2
SDR block diagram showing the digital back end, RFE comprising of the Rx and Tx, time board, and power board (Source: Per Vices)
Figure 2
SDR block diagram showing the digital back end, RFE comprising of the Rx and Tx, time board, and power board (Source: Per Vices)

Instantaneous bandwidth is a crucial parameter when selecting SDRs for T&M, with the highest-bandwidth SDRs in the market reaching 3GHz with multiple-input, multiple-output (MIMO) RFE channels. The RFE interfaces with the digital back end through independent digital-to-analog convertors and analog-to-digital convertors. The back end is the brain of the SDR, being usually composed of a high-end field programmable gate array (FPGA) with onboard DSP capabilities. It is responsible for all basic radio functions, such as modulation, demodulation, up/down-converting, and data packaging. It also runs more complex algorithms concurrently, including communication protocols, encryption schemes, and artificial intelligence. The main advantage of the FPGA is that it is highly reconfigurable, capable of performing completely different tasks without any need for hardware modification. This feature allows the SDR to integrate several components of the T&M system in software, reducing the total amount of hardware required, and being easy to integrate with legacy equipment. 

One of the most useful features of SDRs is their built-in compatibility with GNU Radio. GNU Radio is an open-source software development toolkit for RF systems. It contains several signal-processing and communication algorithms, making software development for SDRs simple and intuitive. Due to its open-source nature, GNU radio is constantly being upgraded by developers, making it reliable and compatible with most up-to-date technologies. Several T&M packages are available for GNU Radio, including frequency spectrum tools, spectrum waterfall plots, oscilloscope plots, waveform generator, and spurious signal detection. Furthermore, GNU Radio can be used to measure receiver parameters, such as amplitudes, noise floor, dynamic range, and spurious-free dynamic range. Another useful tool is the constellation diagram, which can help visualize IQ phase imbalance, DC offsets, and imbalances in the amplitude. 

SDRs also play a major role in laboratory automation for T&M protocols, because their software-based nature allows storage of recipes and continuous communication with the host infrastructure. For instance, tests can be programmed to complete several tasks in a pre-defined order, while also respecting power thresholds, thus reducing the risk of component damage and human-induced errors. Furthermore, the embedded host connection of SDRs can be used to integrate the T&M system into the lab server, allowing remote control of the tests via TCP/IP. T&M automation is also important in field applications, where the remote installation of the RF components hinder human intervention during the testing process.

Another advantage of using SDRs in T&M is their ability to use a CI/CD paradigm for automated code compilation, deployment, and testing. A CI system automates the merging process of the code of several developers into a common repository. It automatically compiles the main program, runs tests, and provides immediate feedback, which reduces the chance for compatibility errors while maintaining a stable application. CI benefits software development in several ways, such as reducing lead time and mean time to resolution, increasing in the deployment frequency, and reducing the failure rate. Moreover, debugging failures is much more convenient in a CI environment. Following this trend, Per Vices offers the ability to use the CI testing system when purchasing one of their SDR models. Figure 3 shows a basic CI setup applied to T&M.

Figure 3 
 CI system process for testing (Source: Per Vices)
Figure 3
CI system process for testing (Source: Per Vices)

— ADVERTISMENT—

Advertise Here

The flexibility and modularity of SDRs allow the integration of several T&M functions into a single device, reducing the total amount of hardware required for a testing setup. These functions can be programmed on the fly, providing fast reconfiguration to accommodate changing testing conditions and scenarios. Also, the same SDR can provide several different waveforms and transmission schemes simultaneously. MIMO SDRs, for instance, allow the application of several independent channels at the same time, that can be either implemented for different Tx/Rx schemes or act in coordination, which is useful for beam-forming/beam-steering techniques. 

Furthermore, SDRs often provide embedded storage solutions that can be used to record large amounts of information; the highest-throughput SDRs on the market offer state-of-the-art network interfaces that allow the transfer of huge amounts of data to the host system. SDRs are also highly upgradable, and are easily updated to the latest algorithms, protocols, and RF techniques. Modular SDRs provide a high level of device customization, to satisfy any requirements in terms of software, host interface, and SWaP requirements. 

TESTING EXAMPLE #1: ANTENNAS

An antenna is the most basic component in an RF system. It is the interface between the electronic circuit and the electromagnetic waves, converting radio signals into electrical voltages/currents and vice versa. There is a huge variety of antenna designs in the RF industry, with different sizes, shapes, and styles that depend on the intended application. They range from very small micro-strip antennas in PCBs to large parabolic dishes in ground stations. Antennas can be applied in point-to-point communications, broadcasting, satellite communication, radar systems, and radio astronomy. 

One of the most important antenna designs in the industry is the Yagi-Uda antenna—a directional antenna based on one driven element and two or more parallel resonant elements acting as half-wave dipoles. There are two types of resonant elements. Reflectors (usually one) are behind the driven element and opposite the direction of intended transmission. Directors are in front of the driven element and in the intended direction. Yagi-Uda antennas are commonly used with televisions, where they gained their popularity in the 1950s. 

One of the most important T&M assessments in antennas is the measuring the “radiation pattern.” It consists of a diagram describing the strength of the transmitted or received radio signal in each direction, so it is crucial to verify the directivity and function of the antenna. Using the radiation pattern, one can easily evaluate the antenna efficiency at a given direction, which is a great indicator of performance for a given application. 

The radiation pattern is divided into lobes The main lobe is where the antenna radiates the most, defining its directivity and coverage area. The secondary (side) lobes, and back lobe are radiation sections where the antenna is wasting energy, and should be reduced if high directivity is desired. Several common shapes of the radiation pattern are well-understood, due to their large application range. The simplest one is the omni-directional pattern, which sends energy to every direction; but the industry has some other examples, such as the pencil-beam pattern, the fan-beam pattern, and the shaped-beam pattern. Each pattern is best suited to a certain set of applications. 

There are several studies involving antenna assessment and SDR-based T&M, especially for radiation-pattern analysis. For instance, Sreethivya and others [1] proposed a measurement setup to acquire the radiation pattern of a Yagi-Uda antenna using SDRs and GNU Radio. Two whole system consisted of two SDRs and two host computers, along with the antenna. This study is particularly interesting, because it shows that a simple and relatively inexpensive setup can measure the complete H-plane (0-360º) of the Yagi-Uda, and provide accurate results that are comparable with expensive conventional T&M equipment. The use of GNU Radio was key to achieving this level of cost reduction, due to its open-source nature. 

In conventional approaches, a dedicated transmitter and receiver, with fixed behavior defined by hardware, must be implemented to acquire the signal strength at one point. Using SDRs, general-purpose equipment can easily perform the pattern measurement automatically, with practically no human intervention, save the acquired data, and display the results in a friendly graphical user interface through open-source software. The same setup can be reconfigured to work with any other antenna, by simply modifying the software involved.

TESTING EXAMPLE #2: MEASUREMENTS AND CALIBRATION FOR 5G/MOBILE NETWORKS TESTBEDS

Mobile network testbeds are controlled environments that simulate real-world infrastructures. They test novel RF solutions, including radio access network (RAN) components, MIMO systems, software-defined networks, and cloud architectures. Testbeds are research catalyzers, providing a rich environment for collaboration among multiple institutions and companies. Some examples of wireless testbeds are the COSMOS (Rutgers University), the Colosseum (Northeastern University), and the 5GUK (University of Bristol). The design and build process for testbeds is a highly technical problem, often requiring the expertise of several professionals from fields such as electronics, software development, network management, and signal processing. Wireless testbeds must be robust and reliable, so that engineers can trust the results obtained in the performed T&M.   

Calibration algorithms can be implemented in a mobile network testbed. They can be used to calibrate the RFE transceiver in terms of DC offset, filter bandwidth, and imbalances, which further improves the performance of the radio module. They also support algorithms for power management, which are useful to coordinate power-down sequences and control power distribution under different operational modes, while also providing current and voltage limitations to protect the components. 

Furthermore, testbeds can be used to measure the performance of DSP functions of the radio module, evaluating how well the receiver can separate the simultaneous signals coming from the user equipment connected to the network. Finally, testbeds provide means to measure how well data is being transferred, in terms of network latency, data access rates, and information throughput. This is the perfect environment for CI/CD testing, since developers can immediately verify how well a software update will work in a real network.

Another useful application of testbeds is to evaluate the performance of MIMO antenna arrays, in terms of near-field patterns, beam shaping, and polarization measurements. MIMO SDRs are great tools to drive beam-forming antenna arrays, because they provide several Rx/Tx channels that work independently but are synchronized by a common time board (see Figure 4). 

Figure 4
MIMO Beam-forming SDR (Source: Per Vices)
Figure 4
MIMO Beam-forming SDR (Source: Per Vices)

The T&M system proposed by Marinho and others [2] was able to measure the radiation pattern of a beam-forming antenna array, using only one MIMO SDR for both Tx and Rx at 28GHz, and a host computer. In a conventional T&M system, dedicated transmitters and receivers would be required, with specific electronic tuning to accommodate the particularities of the transmission system. Moreover, it would be significantly more difficult to coordinate each RF channel to achieve the same levels of beam forming without software management. The SDR solution allowed fine tuning on software, using mostly commercial off-the-shelf equipment.

Finally, it is almost impossible to discuss 5G testbeds without mentioning the open-RAN (O-RAN) oriented systems. O-RAN is one of the most popular 5G architectures available, focused on the interoperability and standardization of the RAN components, open-source software elements, and network disaggregation. However, several challenges must be solved to bring O-RAN to life. These include the design of robust, data-driven control loops, the reduction of data access overhead to artificial intelligence agents, and the exact requirements of each network element. 

All these issues require very large network testbeds to evaluate possible solutions. One of the biggest examples is the Colosseum testbed in Boston, which is the largest network emulator in the world; it consists of 256 SDRs that can emulate 65,536 radio channels operating at 100MHz. There are several studies about integrating O-RAN solutions in the Colosseum, including the ColO-RAN system proposed by Polese and others [3], which was created to develop apps based on machine learning for O-RAN closed-loop control.

— ADVERTISMENT—

Advertise Here

TESTING EXAMPLE #3: MILITARY TEST RANGES FOR RADAR/SATELLITE

Military technology for both electronic warfare and defense must undergo rigorous and well-developed T&M protocols, to ensure that RF devices work properly in the battlefield. This is particularly important in radar systems and satellite communications, which are fundamental for situation awareness and decision making. To realistically simulate real-life situations for combat training and equipment testing, much of the land owned by the military is dedicated to training and equipment testing, providing a large territory to test range and latency.

Radar cross section (RCS) is an important parameter in radar T&M. It basically describes how well a target can be detected by radar. The RCS is the equivalent area of the target when detected by the radar, so the bigger the RCS of the object, the easier it is to detect. In software-based T&M, SDRs can be connected to the radar antenna, and transmit the radar signal, receive the resulting echo, and process the signal in the host computer using Fast Fourier Transforms and other DSP functions, yielding the RCS for a given target on the fly. This can be used, for instance, to evaluate how well a stealth aircraft can hide from enemy radar, or to calculate the performance of a radar antenna. Figure 5 shows the RCS measurement setup.

Figure 5
Radar cross section (RCS) measurement
Figure 5
Radar cross section (RCS) measurement

Spatial awareness is key to the success of a mission in the battlefield, so testing GPS equipment is crucial in military T&M. GPS field testing is used to analyze receiver functionalities in the tactical environment, evaluate the device’s immunity to adversarial interference, and compare the performance of different equipment in real-life scenarios. SDR-based emulators can be used to simulate GPS signals in a controlled way, which increases the repeatability of the test results under a set of given conditions, ensuring that the results are sound and reliable.

CONCLUSION

The continuous evolution of the RF industry requires flexible, reliable, and reconfigurable testing and measurement solutions in all the processes involved during product design and certification. T&M is crucial in RF development, to ensure that products work properly and are compatible with a certain electromagnetic environment. Conventional T&M techniques based on hardware cannot keep up with the fast development of new RF technologies, especially considering the Internet of Things, 5G, and cloud-based networks. SDRs combine the processing power of DSP algorithms, the flexibility of FPGAs, and the analog performance of high-end RFEs. They provide T&M solutions that not only can work with modern RF systems, but also evolve with them, with minimal hardware modifications. 

Other advantages of using SDRs in T&M include long-term cost reduction, built-in compatibility with open-source software GNU Radio, on-board host interface for high-throughput communications, MIMO channels for beam-forming/beam-steering, reduction in the total equipment count, and compatibility with CI/CD development. Example cases of SDRs being used in high-performance T&M, including antenna measurement, 5G mobile network testbeds, and military testing, indicate that SDRs are already becoming the norm in the T&M industry. 

RESOURCE
Cyan software-defined radio.
https://www.pervices.com/

REFERENCES
[1] Sreethivya, M., Dhanya, M. G., Nimisha, C., Gandhiraj, R., and Soman, K. P. (2014). “Radiation pattern of Yagi-UDA antenna using USRP on GNU radio platform.” International Journal of Research in Engineering and Technology, 03(19), 69–71. https://doi.org/10.15623/ijret.2014.0319014.
[2] Marinho, D., Arruela, R., Varum, T., and Matos, J. N. (2020). “Software-defined radio beamforming system for 5G/radar applications.” Applied Sciences, 10(20), 7187.
https://www.researchgate.net/publication/346245003_Software-Defined_Radio_Beamforming_System_for_5GRadar_Applications
[3] Polese, M., Bonati, L., D’Oro, S., Basagni, S., and Melodia, T. (2021). “ColO-RAN: Developing machine learning-based xApps for open RAN closed-loop control on programmable experimental platforms.” IEEE Transactions on Mobile Computing. (Submitted for publication.)
https://arxiv.org/pdf/2112.09559

PUBLISHED IN CIRCUIT CELLAR MAGAZINE • DECEMBER 2022 #389 – Get a PDF of the issue

Keep up-to-date with our FREE Weekly Newsletter!

Don't miss out on upcoming issues of Circuit Cellar.


Note: We’ve made the Dec 2022 issue of Circuit Cellar available as a free sample issue. In it, you’ll find a rich variety of the kinds of articles and information that exemplify a typical issue of the current magazine.

Would you like to write for Circuit Cellar? We are always accepting articles/posts from the technical community. Get in touch with us and let's discuss your ideas.

Sponsor this Article
+ posts

Brendon McHugh is a Field Application Engineer and Technical Writer at Per Vices. He has a degree in Theoretical Physics from the University of Toronto. He can be reached at solutions@pervices.com.

Kaue Morcelles is an Electronics Engineer and Technical Writer with a Masters degree in Electrical Engineering from Santa Catarina State University in Brazil. He is pursuing his PhD.

Supporting Companies

Upcoming Events


Copyright © KCK Media Corp.
All Rights Reserved

Copyright © 2024 KCK Media Corp.

Software-Defined Radios

by Brendon McHugh and Kaue Morcelles Brendon McHugh and Kaue Morcelles time to read: 15 min