In recent years, the rapid expansion of the civil unmanned aerial vehicle industry, especially the widespread application of micro and small drones in various domains, has brought about increasingly prominent flight safety issues. Incidents where drone flights affect the takeoff and landing safety of civil aviation aircraft or threaten the security of important areas and facilities occur from time to time. With the growing number of drones held by the public, the problem of uncontrolled flight poses severe challenges to drone regulation and the healthy development of the drone industry. From my perspective, addressing these challenges requires a fundamental rethinking of how we monitor and manage drone operations. In this article, I will analyze the core difficulties in drone regulation from technical and economic standpoints, and propose a practical and cost-effective solution inspired by existing aviation surveillance technologies.
1. Core Challenges in Drone Regulation
The current difficulties in drone regulation stem from two primary technical aspects, in addition to policy issues: 1) the difficulty of detection, and 2) the high cost of monitoring and control. Let me elaborate on each.
1.1 Difficulty of Detection
Civil drones, especially micro and small ones, exhibit the “low, small, slow” characteristic. They typically fly at low altitudes, move at slow speeds, have a small radar cross-section, and follow irregular flight paths. Consequently, when a micro drone is airborne, it is often very hard to detect in a timely manner, making it difficult to apply appropriate management measures. The current mainstream detection technologies include:
- Conventional Radar: Near the ground, radar faces strong ground clutter noise. The radar cross-section of a micro drone is extremely small, so the reflected signal is easily buried in noise. Moreover, drones often hover or move very slowly, depriving radar of the velocity discrimination feature. Radar can only “see” echoes, not distinguish target identity.
- Conventional Electro-Optical (EO) Systems: Due to the small size of micro drones, their optical and infrared signatures are weak. Without prior guidance, searching for a drone with EO equipment is like looking for a needle in a haystack. Weather conditions such as rain, fog, poor illumination, or low visibility severely degrade performance. The effective range of EO systems is typically much shorter than that of radar or radio detection.
- Radio Frequency (RF) Detection: RF detection monitors the drone’s data link or control link as a signal source. However, with the development of frequency hopping, spread spectrum, and OFDM technologies, the RF power is distributed over a wide spectrum, making each spectral line very weak. This makes narrowband direction-finding increasingly difficult. Additionally, most consumer drones use the shared 2.4 GHz and 5.8 GHz ISM bands, which are crowded with diverse wireless services and complex electromagnetic environments, introducing many uncertainties. Especially for non-preset sites and non-preset drone models, the probability of intercept is very low.
The following table summarizes the detection limitations:
| Technology | Key Limitation | Effective Range | Cost Level |
|---|---|---|---|
| Conventional Radar | High ground clutter; small RCS; slow speed/hovering | Moderate (3‑8 km) | High (₤ several million) |
| Electro-Optical/IR | Weak signature; weather dependent; short range | Short (<2 km typically) | High |
| RF Detection | Spread spectrum/hop; crowded ISM bands; low intercept probability | Moderate (3‑8 km) | High |
1.2 High Cost of Regulation
Each of the three detection systems mentioned above, if purchased individually at a “usable” performance level, costs on the order of several million yuan. Integrating all three into a basic detection system typically exceeds ten million yuan. Moreover, these monitoring devices have limited effective ranges—usually only 3 to 8 km. To improve identification and detection capabilities to a “good” level, one would need high-performance phased-array radar, fast-scanning laser/IR EO systems, and wideband high-sensitivity beamforming RF detection equipment. This dramatically increases the cost. If we need to expand the coverage area, a network of multiple devices and systems working together must be deployed, pushing costs even higher. For nationwide drone regulation, such costs are simply unaffordable.
This cost challenge can be expressed in a simple formula. Let Ctotal be the total cost of a wide-area drone monitoring system based on traditional detection methods:
$$
C_{\text{total}} = N \times (C_{\text{radar}} + C_{\text{EO}} + C_{\text{RF}} + C_{\text{integration}})
$$
where N is the number of sensor nodes required to cover the target area. For a typical city (radius ~20 km), even with a sparse deployment of three nodes, the total cost easily exceeds ¥30 million. This is clearly unsustainable for mass-scale drone regulation.
2. A New Approach: Learning from Civil Aviation Surveillance
The fundamental problem is that the airspace is “blind” to drones—we cannot see them in real time. Existing technical solutions are prohibitively expensive. However, we already have a proven real-time surveillance system for manned aircraft: the ADS‑B (Automatic Dependent Surveillance–Broadcast) system, mandated by ICAO. ADS‑B uses GPS and air‑ground/air‑air data links to broadcast aircraft identity, position, altitude, velocity, and other parameters automatically and autonomously. Ground stations and other aircraft receive these broadcasts, enabling full situational awareness. This system has greatly enhanced the performance of legacy radar, expanded coverage, and improved accuracy. Today, most manned commercial and general aviation aircraft, as well as large civil drones, are equipped with ADS‑B.
Could ADS‑B be directly used for consumer drone regulation? Technically, yes—it is more than capable. However, the sheer number of consumer drones, which can reach extremely high densities in some local airspace, especially in swarm operations, would clutter the ADS‑B display and interfere with normal civil aviation monitoring.
Therefore, I propose developing a simplified, dedicated system tailored for consumer drones, which I call APS‑B (Automatic Position Surveillance–Broadcast). This would be a lightweight version of ADS‑B, specifically designed for drone regulation. Let me examine its feasibility from both technical and economic perspectives.
2.1 Technical Feasibility
To achieve automatic position broadcasting, we first need basic parameters: a unique drone identity, real-time latitude, longitude, and altitude. Modern consumer drones already generate these data for their flight controllers. The only additional requirements are: (a) standardizing the drone identification code format, and (b) encoding the identification and position information into a broadcast message format. The wireless modulation and broadcasting techniques for such messages are mature—they are simple, low-bandwidth, and do not require sync‑channel establishment. We can either reuse ADS‑B data link technology or adopt another simple method.
2.2 Economic Feasibility
The cost added to each drone for APS‑B is minimal. The identity and position data already exist internally; the only extra cost is for encoding software and a small radio transmitter module. Using off‑the‑shelf components, the added bill‑of‑materials is around tens of yuan per drone (less than US$10). For mass production, this can be even lower. On the ground, an APS‑B receiver plus display software costs about a few thousand yuan. Compared to the multi‑million‑yuan systems that can only monitor a few kilometers, APS‑B offers orders‑of‑magnitude better cost‑effectiveness. Moreover, as long as receivers are deployed, the monitoring range can be extended almost indefinitely by networking.
The cost comparison is summarized in Table 2:
| Item | Traditional Integrated Detection System | APS‑B Based System |
|---|---|---|
| Per‑node cost (ground) | ¥10–30 million | ¥3,000–10,000 |
| Per‑drone added cost | N/A (external detection only) | ¥30–100 |
| Coverage per node | 3–8 km | Up to 10 km (TX power dependent) |
| Scalability for nationwide use | Prohibitively expensive | Feasible with low deployment cost |
This dramatic cost reduction is the key enabler for wide‑area drone regulation.
3. The Proposed Drone Regulation Solution
My proposed solution consists of two pillars: (1) the APS‑B system to make drones “visible and identifiable,” and (2) a wireless‑based electronic fence technology to enable flexible regional control.
3.1 APS‑B System – Technical Specifications
I recommend the following technical parameters and standards for APS‑B:
| Parameter | Description |
|---|---|
| Unique Identification Code | Uniformly formatted: manufacturer ID + model ID + serial number |
| Position Data | GPS/BeiDou latitude, longitude, altitude; optional barometric altitude or relative height |
| Data Format | Compact binary format to minimize transmission time and support high‑density airspace |
| Transmission Channel | Nationally unified; recommend a dedicated channel in 960–1164 MHz aeronautical band (coordinate with MIIT) |
| Broadcast Rate | Once per second nominal; adaptively increase when velocity/position changes rapidly |
| Transmit Power Classes |
|
| Interoperability Format | Standardized data packet structure for multiple application systems and network sharing |
The relationship between required transmit power Ptx and desired range R can be approximated by the free‑space path loss formula (assuming omnidirectional antennas):
$$
P_{\text{tx}} (dBm) = P_{\text{rx,min}} + 20\log_{10}\left(\frac{4\pi R}{\lambda}\right) + \text{fade margin}
$$
where Prx,min is the receiver sensitivity, λ is the wavelength, and fade margin accounts for obstacles and interference. For example, at 1 GHz (λ≈0.3 m), a range of 5 km with a receiver sensitivity of −100 dBm and 10 dB margin yields Ptx ≈ −100 + (20*log(4π×5000/0.3)) +10 ≈ −100 + 66.4 +10 = −23.6 dBm, which corresponds to about 4 mW, easily achievable with low‑cost transmitters.
3.2 Wireless Electronic Fence Technology
An electronic fence is a controlled airspace defined by geographic coordinates (latitude, longitude, altitude). Drones approaching such areas receive warnings; if they attempt to enter, they are alerted or forcibly prevented from doing so. This is an essential component of drone regulation.
Currently, many drones store electronic fence data preloaded by manufacturers, which can be updated online. However, such data may not be complete or timely, and cannot reflect temporary control measures—for instance, during major events or emergencies where a zone is restricted for a few days. Therefore, the electronic fence system must support real‑time acquisition and forced update of control data. Two methods are proposed:
- Method 1 – Public Mobile Network: This is straightforward: the fence data can be uploaded to the drone via a smartphone or other mobile network‑connected controller. The drawback is that mobile network coverage may be poor in some areas, and during major events, temporary radio silence may be enforced, making the mobile network unavailable.
- Method 2 – Dedicated Wireless Beacon Stations: Set up dedicated beacon transmitters at important or temporarily restricted airspace. Drones flying nearby receive the beacon signal and are warned, or the data is directly fed into the flight controller to prevent entry. This method is more flexible for ad‑hoc control: a temporary beacon can be installed for a few days and removed afterward, without affecting future flights in that area.
Regardless of the method, a uniform data format for electronic fence transmission must be defined so that manufacturers can integrate it into their systems. The beacon stations should use a nationally unified frequency channel and modulation scheme. Again, I recommend coordinating with MIIT to allocate a channel in the 960–1164 MHz aeronautical band. The two methods can also be combined depending on the scenario.
4. Recommendations for Implementation
The solution I have described involves both flight management and manufacturing regulation. It requires supporting technical standards, especially for the unique identification code, position broadcast, and electronic fence data. With these in place, the drone regulation process becomes visible, controllable, and actionable, effectively solving the “blindness” and “expense” problems. This will enable gradual liberalization of airspace for drones, promoting the healthy development of the industry and its application across various sectors. To achieve this, I recommend the following actions to the relevant authorities:
| # | Recommendation |
|---|---|
| 1 | Develop a “Civil Drone Type Certification Specification” as soon as possible. Mandate unique ID, APS‑B, and electronic fence capabilities as compulsory requirements. Translate flight safety requirements into detailed technical indicators. Consult with drone manufacturers to gain their understanding and cooperation. |
| 2 | Implement a production type certification system: all drones manufactured in or used within China must comply with the certification specification. |
| 3 | Regulate imports: require all imported drones intended for use in China to pass type certification. |
| 4 | Regulate sales: all drones sold within China must be type‑certified. |
| 5 | Since the drone industry is global, standards should be developed with both domestic and international perspectives. Actively collaborate with ICAO early on to push Chinese standards to become international standards, leading the global drone regulation framework and contributing to the international community. |
In summary, drone regulation is not an insurmountable problem if we adopt a pragmatic, technology‑based approach. By leveraging the proven concept of cooperative surveillance (like ADS‑B) and creating a simplified APS‑B system combined with wireless electronic fences, we can achieve cost‑effective, scalable, and real‑time monitoring of consumer drones. This will not only enhance safety and security but also unlock the full potential of drones in society.
