Elevating QoE-centric network performance testing into the 3rd dimension

Traditionally, QoS/QoE centric mobile network performance assessments are conducted on the ground, either with in-car drive test setups like the Benchmarker 3, designed for large-scale benchmarking measurement campaigns, or with portable backpack test systems like the Freerider 4 for indoor or walk tests. However, our innovative customer Wandel Services from Hungary has taken network testing to new heights—literally. By mounting Rohde & Schwarz equipment onto drones, they have pioneered an aerial and autonomous testing approach. After two years of experiments and trials, the solution is now market-ready, with the first commissioned network benchmarking tests already implemented.

In this interview, Zoltan Molnar, the Managing Director of Wandel Services , Honorary Associate Professor at Széchenyi Istvan University in Gyor and the visionary behind this groundbreaking solution, shares his journey in bringing this idea to life.

Note: At the end of the interview, we also made a comparison between other drone-based network testing solutions supported by Rohde & Schwarz products.

What specific challenges in mobile network testing and benchmarking led you to develop a drone-based solution, and how does it address these challenges more effectively than traditional methods?

Zoltan Molnar:

Professionally, I have been involved in mobile network testing for more than 20 years. But already decades ago, before the university, I was a student in the Aircraft Instrumentation Faculty of a secondary school, and even if life drifted me to the field of telecommunications testing devices, mentally, I remained a “flying child”.
With this background, I’ve always kept tabs on various types of flying vehicles, particularly those becoming accessible in everyday life. Not so long ago, the world of drones stepped to a new level, started to spread everywhere, and the flight regulations appeared accordingly. First, it was a pain for both the amateurs and the professionals. Then, with the availability of better and better drones, we got used to it.
In a very short time, drones have been used in almost every part of the industrial and government fields. One day, I found an article mentioning that some drone types had already become available on the market, including the option of using public mobile networks for video transmission. Then, going deeper into the topic, it was predicted that even the controlling of a drone could be done via mobile networks. And now it is not the future.

Dealing with mobile network testing solutions for many years, I imagined the frightened face of the mobile service providers: “Will we have to provide responsive and wideband data networks in the middle of nowhere because of the agricultural and package delivery drones?” “How will we test it above the ground level?”
The answer to the first question is, unfortunately, “yes”. Because agricultural drones are flying in the middle of nowhere. Moreover, newer use cases also appear on ground level, far away from populated areas.
The answer to the second question became obvious: drive test cars and Freerider backpacks cannot fly, so we must mount the mobile network test systems to drones.

Can you elaborate on the integration process with existing measurement equipment, like drive test cars and backpacks? How does this modularity enhance the flexibility and efficiency of network testing?

Zoltan Molnar:

The most important aspect was what we often call the "laziness of the engineers": I did not want to reinvent the wheel. On the other hand, wherever I mentioned this topic, practically everybody wanted to see the solution immediately. Thus, I checked what we had: Drones are available in the necessary size ranges, Benchmarker 3 and Freerider 4 are in our product range. Why don't we put them together?

While the concept seemed straightforward in theory, our journey was far from smooth. Our first hurdle was finding the right drone type. Typically, informing a shop that a product will be immediately disassembled after purchase doesn't sit well with the sellers, often leading to warranty issues. However, our persistence paid off when we discovered ABZ Innovation, a Hungarian agricultural drone manufacturer, who understood our unique requirements and allowed us the freedom to execute our vision.

After removing the spraying system and the pumps and mounting the Benchmarker/Freerider parts (with the active help of SZESAT and DDC, the innovative and enthusiastic organizations and student groups of Széchenyi Istvan University in Gyor), we had the first test flights. And the flying time was dramatically short. We had to "realize" that, unlike chemical spraying, we do not drop our payload to the ground part-by-part (fortunately), so our landing weight equals to the take-off weight, which needs a lot of energy all over the flight.

Confronted with the issue of a short flying time, we took decisive action. We stripped the test system of all non-essential components and, in parallel, the drone manufacturer installed a second flight-battery bay and more effective, new rotors. This problem-solving approach, coupled with our determination, led to a significant increase in the flight time for net measurements, reaching a satisfactory 20…35 minutes, considering the high speed of the drone.

The result is a system that is fully compatible with all existing Rohde & Schwarz benchmarking solutions. The drone has holders for up to 8 QualiPoc devices, the same TSME6 scanner and NCM fixing parts used in Freerider 4, which exactly acts as a Freerider 4 or Benchmarker 3 running SmartBenchmarker (or Romes). The only difference is that it is flying. This compatibility reassures stakeholders and highlights our project's integration capabilities.

What are some of the most significant benefits that telecom operators or regulatory authorities can expect from using this drone-based benchmarking system in their network performance management?

Zoltan Molnar:

Initial responses from the market revealed a significant misunderstanding of the solution's intended purpose. Mobile network operators queried, "Will future benchmarking be drone-based and automatic?" Meanwhile, drive testing companies expressed concern, "Will this innovation render drive test services obsolete?"
Absolutely "no" concerning both questions.
Drones (especially in this size and weight range) cannot legally fly above populated areas according to the actual flight regulations. Their flight is even more risky and prohibited at the "altitude" of the head of a human. The fundamental problem (mobile network-based drone operation) is irrelevant if there are no drones. From a classic drive testing point of view, it does not make sense to use drones at all. Where people are, there are roads — or at least pavements, so the drive test cars and Freerider backpacks can easily be used.

The challenge comes when we go to the "at this moment unusual" cases. In agricultural fields, where a drive test car cannot even use its wheels, more and more modern agricultural machines need mobile network coverage. Industrial sites, which cannot be approached by a drive test car (e.g. huge railway cargo sites, chemical industry places, mines, etc... ), and we are still on ground level. Then comes the drone industry in all its civil and non-civil ranges, from ground level up to 120 meter AGL (Above Ground Level) altitude.

For regulators and the QoS responsible departments of telecom operators, the solution to these challenges lies in the form of 'benchmarker drones '. These drones, specifically designed for benchmarking, can effectively address the unique testing needs of the telecom industry.
The more painful topics are just starting: To what degree is a mobile operator responsible for data coverage at 120m altitude, far away from any city or village? Will private networks solve this problem? How do the authorities have to regulate this? Our main point is that accurate measurements are now possible and available for these hard discussions.

Which customer groups can you address, which test use cases can you cover and which insights can you generate for your customers?

Zoltan Molnar:

The previous point answered the main points of this question. The system addresses all mobile network providers, drive test companies, and regulators, and due to its technical setup, all of them can get every kind of test result they could measure with their existing or future-purchased benchmarker devices and probes.
However, there are two extremely important aspects which we were keeping in mind during the complete process of experiments and fine tunings, in order to make this test dimension extension smooth and budget friendly to all existing and future customers. Our system is designed to be cost-effective, providing financial benefits to our customers and instilling confidence in the value of our product.

• No new test system parts (except the drone itself) must be implemented and learned. It is even possible that a customer who already has a Benchmarker 3 or a Freerider 4 goes to the field with an existing drive test car (which can carry the drone, too, in folded position), removes the necessary amount of QualiPoc phones, a scanner and an NCM from his system, clicks them to the drone, and starts flying. The complete test setup can be controlled and programmed from the existing Benchmarker system, and the results can also go to the existing post-processing part. Based on an EU-wide reseller contract with the drone manufacturer, Wandel-Services can provide either a "drone only" package or the complete system, depending on the actual needs.
• The system is in the C4 UAS category, which is very important: It can be registered in the given EU country's flight authority accordingly and operated in the Open flight category with a simple A3 drone pilot license, such as a commercial hobby drone. (Of course, these categories' rules must be kept during the flight.) We took extreme care about the weight parameters, so even if 8 QualiPoc devices, a TSME6, and an NCM are on the board, the MTOM (maximum take-off mass) is under 25kg. (In fact, 24.8kg.) Why is it important? This way, only the basic drone pilot license is needed. No special flight operation permissions and complicated airspace reservations are necessary inside the EU in the above-mentioned use cases, so no special-skilled workforce or dedicated employees are required from the customers.

Can you briefly explain the technical setup how you installed all the required R&S equipment onto a drone?

Zoltan Molnar:

As seen in the system topology diagram, the logic of the Freerider 4 or Benchmarker 3 is untouched. When the system is mounted to the "ready to fly" state, it acts like Freerider 4.
In principle, we could have fixed the backpack itself to the drone, but then we would have exceeded the targeted weight and decreased the flight time to a useless range. Thus, we removed everything that was not critically necessary. Of course, commercially, we are not selling a Freerider backpack for the drone, only the required number of QualiPoc devices, the scanner, and the NCM computer with the related software options.

Since we are working with a flying vehicle, we are using phone holders and TSME6/NCM trays to protect the devices against falling down and against springing particles (e.g. small stones in the rotors' wind) while their cooling is still fine. To the lower tray, we mounted a lightweight but system-compatible USB hub and the power supply of the test system. Everything is connected to the necessary length of cables (not longer, due to the weight), except the GNSS and scanner antenna cables, because in the first versions, we want to keep the original RF status.
Powering the test system is still improving. Now, we are using a lightweight airplane model LiPo battery in a hot-swap connection and a power stabilizer. With this, the test system works much longer than the flight time with two flight batteries. Experiments are in progress to change this to a small inverter to power the test system directly from the drone's main flight batteries. By further decreasing the payload weight, we have longer flight time than when we use an extra 12V/10A from the main battery.

Besides these, we use the well-known Freerider 4 tablet to reach the NCM, but its control can be done from an average laptop or the drive test car's controller (if the customer has them).
When the drone system reached its recent status, we conducted a high-precision RF emission test in a shielded chamber to ensure that the drone control and high-power electronics did not cause disturbing signals that could ruin the test results.
The drone's default accessories and its "Mission Planner" are excellent for controlling it. Theoretically, running the drone's mission control from the test system's controlling laptop is possible, but that is not recommended for safety reasons.

Could you share insights from any early adopters or test cases, such as the collaboration with Magyar Telekom and others, that demonstrate the system’s impact on network benchmarking?

Zoltan Molnar:

Yes, there are surprisingly more and more. Right after the first successful test flights, we started to discuss the possibilities and demo measurements with Magyar Telekom (part of DT) because they received some questions regarding the mobile network behavior at higher altitudes. These tests were made in the early phase of our drone project and gave us good results. At that time, MT’s drive test system parts were mounted to the drone, and the interesting fact was that the vehicle was controlled on-site while the onboard Benchmarker was configured and controlled from a 120km distance from the Telekom HQ.

A significant step was the test series in cooperation with SZESAT (Szechenyi Istvan University Satellite Group), where we compared the drone possibilities and capabilities with Zeppelin-like blimp-based measurement cases. This alone could occupy pages. In addition to their valuable results, we collected many parameters to modify the payload and optimize the drone. That was when ABZ Innovation also modified this drone type and started manufacturing it in a version without the chemical spraying system.
Not so long ago, we received the first RFQ from a Hungarian drive test company. It initiated our reseller contract with ABZ Innovation for the EU territory related to all their products.
However, these are all “classic” benchmarking test cases in the air; the results are well-known KPIs, except that some 3D presentations were required.

The most unique, and from a mobile network testing point of view pretty much unusual test case that might transform the whole mobile communication field a little bit is just running:
The customer is HungaroControl (the Hungarian Air Navigation Service provider). This project is a mobile network availability/benchmark test campaign in the Hungarian airspace below 120m in the whole country to check if the operation of SIM-based drone transponders are reliable.
Although the „basic” drone identification electronics (Built-in or Hook-on devices, as the shops call them) have been mandatory in the EU since January 2024 to ensure the correct authority control of the UAV systems and their operators, at this moment, most of them are WiFi/Bluetooth-based and not really suitable to act as real transponders for effective air traffic control.

To ensure the safe operation of all flying vehicles, organizations in the aviation industry worldwide (ICAO, FAA, NASA, EASA, European Commission, vendors etc.)* are intensively working on a standardized „unmanned aircraft system traffic management” (UTM) system.
It is easier to imagine this as a kind of Flightradar24 system for unmanned vehicles and real-time operation. Some transponder devices are already on the market, which are lightweight, SIM-based Network Remote ID circuits that continuously transmit the vehicle parameters, position data, etc. required by the UTM system.
Today, these devices mainly use 4G mobile data networks of public mobile service providers, but ITU and 3GPP are also involved in the discussions, so newer mobile technologies in UTM systems are highly expected.
It is self-evident that these devices work only under proper mobile data network coverage. What is not always obvious (at least has not yet been tested) is the altitude range of the UAV/drone systems.
Thus, this project aims to check how Hungary's territory fits these requirements in the 0m…120m AGL (Above Ground Level) altitude layer.
Needless to say, this topic may soon generate interesting tasks for mobile network service providers and telecommunication authorities.

*Further references on drones resp. unmanned aircraft systems:

• Overview unmanned aircraft system traffic management (UTM)
• International Civil Aviation Organization (ICAO) UTM core principles
• European Commission, Unmanned aircraft
• European Union Aviation Safety Agency (EASA), Easy Access Rules for Unmanned Aircraft Systems

Comparison of drone-based network testing solutions powered with Rohde & Schwarz technology

In addition to the above discussed drone solution from Wandel Services, Rohde & Schwarz technology is also integrated in another drone-based network testing solution from Nokia Drone Networks. In this section we want to give a brief overview about these two solutions and explain their key customer use cases.

Note: Rohde & Schwarz does not market any drone-based solutions. The listed solutions are available from the mentioned companies.