Donate for the Cryptome archive of files from June 1996 to the present

18 July 2013

RockCam Spy Project

Gregory Perry (Gregory.Perry[at] sends:

Lockheed Martin / Defense Threat Reduction Agency / RockCam Project

As we discussed prior, I was the Chief Technology Officer for Advanced Wireless Automation

an imagery intelligence hardware provider in partnership with Lockheed Martin, the Defense Threat Reduction Agency, and the United States Department of Defense.  The particular group at Lockheed was their fourth corporate division with roots in the original Skunkworks R&D group, and this project reported directly to the LMCO CEO and Board of Directors.

During this project, we designed a high resolution, secure wireless-enabled covert imaging hardware platform that was installed into fake rocks placed at various locations, to be used for clandestine imagery intelligence and surveillance applications (project name ROCKCAM).

My particular role as CTO was to liaison with Lockheed Martin engineers and the Defense Threat Reduction Agency officials to get the project through Department of Defense Factory Acceptance Testing, so that the project could be manufactured by Lockheed and then rolled out to other DoD components for homeland surveillance applications and wartime theatre IMGINT operations.  The device itself was an embedded Linux-based SoC encapsulated in a fake rock, and which included a spread spectrum frequency hopping 900MHz radio section configured in a self-healing wireless mesh topology so that the rocks could communicate between each other and use each deployed RockCam as a wireless relay/repeater.  A covert CMOS-based C-mount pinhole imager was installed into each rock, coupled with a motion sensing PIR sensor that would trigger the imager, which in turn would then use a high speed FPGA to compress via a proprietary wavelet compression engine the intercepted image, and then encrypt and transport the imagery intelligence data (initially static high resolution images, later video with corresponding audio) to a centralized Network Security Operations Center (NSOC).  The base of each rock was packaged with deep cycle lead acid batteries, with an expected lifecycle of at least three years after being installed in the area to be surveilled.

The device was intended as a covert imaging platform that could be remotely accessed via handheld computer or PDA to retrieve the imagery data in remote locations, and there was a well-defined API that I worked on with a third company called M2M which ran the Network Operations Center where all of the captured data would be long term stored and mined.

The key to Lockheed / DTRA passing DoD Factory Acceptance Testing was a highly secured encrypted transport layer for the imagery data being relayed between the rocks, so that conventional signals intelligence methods would be unable to view the encrypted image/video streams.  I designed the public key encryption implementation which used a combination of x.509 certificates for the PKI in conjunction with Diffie Helman negotiated AES transport tunnels. After getting Lockheed past Factory Acceptance Testing with the DoD, I was told that the project had been cancelled, that my ownership in the company was effectively worthless, and I was released from my employment with AWA at that time.

A short time later a very similar "RockCam" started appearing in the mainstream news as the result of a spy row between Britain and Russia, with Tony Blair having recently acknowledged the existence of the program:

I've attached to this email the M2M protocol layer interfacing specifications documentation used for interfacing the RockCam with M2M for remote covert imagery transport.  Since the project was ostensibly cancelled by Lockheed Martin and the Defense Threat Reduction Agency, I don't think there should be any national security concerns with the release of any of this information.  It was suggested to me that this particular hardware platform may have become the basis for a missile nose cone fly-by-wire guidance system as well as a drone imaging module, which would make sense given the FPGA/wavelet compression engine that was able to compress high resolution imagery data (120:1 ratio with a better SNR than other competing wavelet standards such as JPEG2K).  However, I have no knowledge of any such derivative projects manufactured by Lockheed, Boeing, or otherwise, so I don't think there should be any national security concerns with auctioning off this hardware and related engineering blueprints to the highest bidder on Ebay.

I am also in possession of a year or so worth of electronic mail correspondence and communications by and between AWA, Lockheed, M2M, and various components such as DTRA that may be of use to other burgeoning startups interested in building covert imager hardware, so all of those documents, the prototype hardware that I have here, and some pretty pictures will be listed on Ebay today or tomorrow for NATO-friendly nation states to peruse and bid on.

I will send you that Ebay auction link once it's live and online.

The prototype hardware I have left is somewhat smokey as the auction will state, as my house accidentally caught on fire last year in the area where the prototype hardware was being stored.