Unmanned Systems Technology 038 l Skyeton Raybird-3 l Data storage l Sea-Kit X-Class USV l USVs insight l Spectronik PEM fuel cells l Blue White Robotics UVIO l Antennas l AUVSI Xponential Virtual 2021 report

98 PS | Secure UAV network comms H ow can unmanned vehicles trust each other? That might seem like an odd question, but many advanced operational concepts involve UAVs cooperating in groups, so secure comms between them are essential (writes Peter Donaldson). Among the potential threats to such operations is the use of ‘malicious’ UAVs as vectors for cyber-attacks, to which wireless networks formed between UAVs can be vulnerable because of their dynamic topology, exchanging information via different routes as their relative positions change. But now researchers at universities in Tehran and Isfahan, in Iran, have come up with a way to use ‘smart agent’ software to provide security for UAVs on networks using an ad hoc on-demand distance vector (AODV) protocol to find the best routes between network nodes. In AODV networks, routes between nodes are created only when they are needed, and deleted when no longer required. When a node aboard a UAV wants to communicate with another and can find no route to the destination in its existing routing table, it sends out a request to other nodes that can provide a route reply. AODV is potentially vulnerable to sinkhole, black hole and grey hole attacks. In a sinkhole attack, the intruder tries to attract network traffic with fake routes. In a black hole attack, the malicious UAV receives data packets but deletes them instead of delivering them, while in a grey hole attack it delivers a portion of the packets but deletes the rest, making its activities harder to discover. In the first of the solution’s two phases, the software analyses the routing behaviour of UAVs on the network to identify intruders, for example by looking for any that, suspiciously, have received data packets without sending them on, and with any of these vehicles that have sent out at least one reply packet in response to a route request from another UAV deemed hostile. In the second phase, the smart agents use a three-step negotiation process to persuade reliable UAVs to reject data from any malicious nodes. The agents themselves take the form of packets of code that migrate among nodes on the network, and their programming includes cryptographic hash functions used to validate the comms between nodes. Cryptographic hash functions are one- way algorithms that when applied to a message produce a unique hash value that cannot be reversed to reveal the original message. They also make any change from the original in a forwarded message obvious, because the cascade effect means that even the tiniest change in the message will produce an entirely different hash value. The researchers tested this secure agent UAV (SAUAV) method using NS- 3, a discrete event comms network simulation tool developed primarily for research and educational use. SAUAV, they say, is more efficient than other security approaches in terms of key parameters such as detection rate, false positive rate, false negative rate and packet delivery rate. So next time you see a group of UAVs looking at each other suspiciously, you’ll have an idea of the kind of intrigues in which they could be enmeshed. Now, here’s a thing “ ” June/July 2021 | Unmanned Systems Technology In AODV networks, routes between UAV nodes are created only when they are needed and deleted when no longer required, but they can be vulnerable