A Defense Technology Blog
See All Posts
  • Civil UAVs Need GPS Anti-Spoofing, But Who Pays?
    Posted by Graham Warwick 9:19 AM on Jul 19, 2012

    The leader of a university research team that showed how easy it is to hijack a UAV by spoofing its GPS won the rapt attention of a congressional hearing this morning (July 19).

    Todd Humphreys, an assistant professor with the University of Texas at Austin, is recommending that, for operation in national airspace, all civil UAVs above 18 lb. be required to have spoof-resistant navigation. And he goes beyond that, recommending that GPS-based timing and positioning systems used in national critical infrastructure be required to be spoof-resistant.

    Humphreys was addressing a hearing of the subcommittee of the House Committee of Homeland Security that is responsible for oversight of the Department of Homeland Security (DHS) — an agency that was conspicuous by its absence from the hearing on UAS and security.

    His recommendations are based on well-publicized tests in which UT Austin researchers were able to take control of a small unmanned helicopter— a relatively sophisticated Adaptive Flight Hornet Mini, representative of small UAVs likely to be operated by law enforcement — by spoofing its civil-GPS navigation system.

    Hornet Mini (Photo: Adaptive Flight)

    Tests included a dry run in the UT Austin stadium, then an exercise at White Sands Missile Range, N.M., where the civil-GPS spoofer was placed on a hilltop about half a mile from where the UAV would be flying. The UAV was commanded by its ground operator to hover at 50 ft. The spoofer began transmitting weak counterfeit GPS signals, achieving meter-level alignment with the authentic signals at the UAV’s GPS antenna. The spoofer then rapidly increased its counterfeit signal power, bringing the UAV under its control. By inducing a false upward drift in the UAV’s perceived GPS location, the spoofer fooled its flight-control system into commanding a dive. At about 10 ft. above the ground, a human pilot assumed control of the UAV to prevent it crashing.

    Graphic: University of Texas Radionavigation Laboratory

    Despite the dramatic demonstration, Humphreys isn't too worried — yet. The UAVs that will be used initially by law enforcement are small, and sophisticated GPS spoofers "are likely outside the capability of organized crime or terrorist organizations...[but] well within the capability of near-peer nation states," he says. But the problem will escalate as more UAVs move into national airspace and, Humphreys emphasizes, "unauthenticated civil GPS . . . [is a] bigger problem than UAS, even manned aircraft have some vulnerability" along with the financial, energy and other sectors.

    In his written testimony, Humphreys discusses a number of potential defenses against spoofing of the civil GPS signals. He dismisses the possibility of using military GPS anti-spoofing technology — the so-called SAASM receiver — because it would more than double the price of small UAVs and kill the market, incur the logistic headache of distributing encryption keys, and run the risk of the technology falling into the wrong hands.

    Nonetheless the Association of Unmanned Vehicle Systems International, which submitted testimony but did not testify, continues to list SAASM as one of the potential solutions to civil-GPS vulnerability.

    Instead, Humphreys reccomends that: "The Department of Homeland Security commit to funding development and implementation of a cryptographic authentication signature in one of the existing or forthcoming civil GPS signals. The signature should at minimum take the form of a digital signature interleaved into the navigation message stream of the WAAS signals. Better would be to interleave the signature into the CNAV or CNAV2 GPS navigation message stream. Best would be to implement the signature as a spread spectrum security code interleaved into the spreading code of the L1C data channel."

    Humphreys says the Pentagon has "indicated its willingness" to incorporate a crypo-authentication signal, or digital watermark, into civil GPS, but does not have the funds to do it. "I believe it will fall to the DHS to fund it, but I am not optimistic," he said. "In the best scenario it will be five years before we see a solution, and I am becoming more pessimistic."

    His statement that "the DHS may have a role to play" in new regulations on GPS and UAV security was welcomed by a congressional panel clearly furious that the department had refused to appear before them — a refusal they chose to interpret as a clear statement by the DHS that it has no responsibility for the security of UAVs in national airspace. It is a position the DHS is unlikely to be able to hold.

    Tags: ar99, unmanned

  • Recommend
  • Report Abuse

Comments on Blog Post