Technology

Threat of Drones and Drone Detection

The Threat of Drones

Drones and other Unmanned Aerial Systems (UAS) have the potential to conduct a coordinated attack on Government offices and Critical Infrastructure of any nation. The threat of drones is presently the most dangerous and crucial in the field of Government Security and Critical Infrastructure Protection. The rapidly evolving threat of Drones is unconventional, un-precedented and it requires a new set of high technology equipment to detect and defeat.

Apart from the threat to security and critical infrastructure, another serious concern is the invasion of privacy by intruding drones. Drones are used to take photos/ record video from residential areas without the permission of the residents. The presence of a drone within the vicinity of a residential area needs to be detected so that appropriate action can be taken to protect the privacy of the residents.

Drone Detection

The commonly used methods for drone detection are video/IR detection, Radar detection and RF detection. Among these methods, RF detection is the cost-effective and long-range detection method.

The most effective way to detect drones is with Radio Frequency (RF) detection technology. It offers long-range detection and can detect all types of drones used today. These RF detectors use ultra wideband antennas to detect and track the RF emission from the drone. Drones emit RF signals all the time and these signals are known to be within certain frequency bands. RF drone detectors continuously scan these frequency bands to identify any signal which resembles the signature of a drone's signal.

Challenges in drone signal detection within ISM bands:

Drones operate in license-free ISM bands. The 2400 MHz & 5800 MHz ISM bands are used by many systems such as Wi-Fi, Bluetooth, ZigBee, NB-IoT etc.
Detection of a drone signal among all other signals within the ISM band is done based on RF-signature analysis of the drone signal.
The detection problem is complicated by the presence of random narrow-band interference within the ISM bands due to harmonics and other interference sources
Conventional methods such as Spectral Shape Correlation or Dynamic Threshold perform badly in the presence of random narrow-band interference.
This limits the detection range as at low SNR, the interfering signal will alter the spectral shape greatly and will reduce the Probability of Detection or may introduce a high rate of False Alarms.

Artificial Neural Network (ANN) based drone signal detection:

SecuDome Drone detector employs an Artificial Neural Network ( ANN) based algorithm for the detection of the drone signal within the ISM band
The AI engine uses multiple parameters such as Bandwidth, Slope ( of the rising edge and falling edge) and two other important parameters of the signal for detection.
The AI engine is trained using a large number of Drone and Non-Drone signals. The training signals are captured over the air from various indoor and outdoor environments.
After training, the AI engine offers near-zero false alarm and a 99.99% probability of detection

Illegal Mobile Phones and Mobile Phone Detection

Modern offices are digital hives of wireless devices such as 3G/4G modems, IoT connected vending machines etc. Many of these devices are inadvertently connected directly to the internet without any security firewalls and offer an opening for cyber attacks. Such devices need to be identified by AI-based RF signature analysis.

Compounding this problem is the presence of illegal phones in security-critical offices, prisons and "no phone" areas. These illegal phones need to be detected and confiscated to eliminate the security risk.

Detection of illegal wireless devices in Offices

3G/4G modems are brought into office inadvertently to connect to the Internet. It provides a direct link to hackers to access computer/network
Modern offices are digital hives of IoT connected devices and machines. IoT enabled printers, vending machines etc. offer an easy gateway for cyberattacks
What is needed is a detector which can monitor the RF environment 24/7 and detect RF threats real-time

Detection of illegal phones within prisons and “No phone” areas

Presence of illegal phones within the prison is a major concern for the internal security of a nation. There are many systems which aim to detect and de-activate illegal mobile phones. Unfortunately, the problem is still widespread and new methods are needed to solve it.

Existing mobile detectors are based on conventional Threshold Detection in spectral-domain offer low detection range.
This is especially true for 3G signal which uses WCDMA modulation. WCDMA has inherent Low probability of Detection ( LPD) feature which makes it vert hard to detect under low SNR conditions
New detection algorithms such as cyclic spectral correlation are needed for an enhanced range of detection
The detection system needs to be cost effective for wider adoption. The detection system needs work in both stand alone mode and networked mode to suit operational requirements

Example : WCDMA detection

As an example of the signal detection methods employed in Ryderoo MDK-21, a brief description of the cyclic correlation method is provided here. This method is similar to the Strip Spectral Correlation Analysis used widely for the detection of Low Probability of Detection (LPD) signals such as WCDMA.

The picture shows the presence of a WCDMA signal which is difficult to detect using conventional spectral-domain methods such as threshold detection or energy measurements. Likewise, advanced signal processing methods are implemented in MDK-21 to enhance the signal detection capability and to offer a longer range of detection.

For more details about the system, kindly contact our team at info@ryderoo.com