2

DARPA (US Defense Advanced Research Projects Agency) has released a Request for Proposal for a Secure Messaging Platform using a blockchain framework.

The following is the Objective paragraph from DARPA's call for proposal:

OBJECTIVE: Create a secure messaging and transaction platform that separates the message creation, from the transfer (transport) and reception of the message using a decentralized messaging backbone to allow anyone anywhere the ability to send a secure message or conduct other transactions across multiple channels traceable in a decentralized ledger.

The request for proposal is available in full, here: DARPA Call for Proposal for Secure Messaging Platform.

  • Why are contemporary secure messaging platforms like TextSecure, Cryptocat, XMPP/OTR, Bitmessaging and so many more NOT enough?
    • What do these contemporary solutions lack, as the Secure Messaging Scorecard suggests many applications have all these features?
  • What does blockchain provide when it comes to Secure Messaging and why it is necessary?

How can Military Interdepartmental Purchase Requests (MIPR) be implmented using the secure ledger or blockchain?

In particular, this paragraph of the proposal is not making sense to me

The messaging platform would act as the transport for a cyptographically sound record of all transactions whether they be MIPRs, contracts, troop movements or intelligence. Troops on the ground in denied communications environments would have a way to securely communicate back to HQ and DoD back office executives could rest assured that their logistics system is efficient, timely and safe from hackers. The benefits are broad and could even be applied to domains such as space. With crowded skies it’s important to maintain situational awareness of all satellites and those concerned with space situational awareness/telemetry or air traffic control could instantly share data between nations using a separate but equivalent ledger implementation thus removing questions as to the authenticity and integrity of the data.

What does transport mean here? How does blockchain help troops communicate in denied communication environment help? I am also not clear about the argument on situational awareness of satellites and how does blockchain enable sharing data between nations.

Simple answers would really help me and others like me. Thanks

2
  • I can't answer why DARPA wants this, but a system somewhat like ripple with a consensus algorithm could probably fit the bill. Transactions would be encrypted, ledger updates could be broadcast maybe via satellite, and entities creating transactions/messages could probably work with very limited upstream bandwidth. Jun 23, 2016 at 8:56
  • @Hans-MartinMosner Can you put this and add some more text as an answer. Please also explain and cite the source for "broadcast via satellite". I would accept it as the answer Jun 23, 2016 at 16:15

2 Answers 2

2

Here's what I think how such a system could work:

  • A relatively small but geographically distributed network of well-connected nodes maintain a ripple-like ledger (https://ripple.com/files/ripple_consensus_whitepaper.pdf) using a consensus algorithm. Each version of the ledger builds upon the previous version and contains transactions of different types.
  • Transactions can contain messaging information, material orders with associated payments, etc. It may be necessary to encrypt each transaction individually with public key(s) so that it is only readable by the intended recipient(s). In addition, each transaction has to be digitally signed by the sending party.
  • Ledger updates (i.e. blocks with new transactions) are signed by the consensus-forming servers and broadcast using whatever means are available to reach the intended recipients.
  • One possible ledger transmission path could be a P2P network based on the internet (simple and widely available.)
  • For recipients that don't have reliable internet connection, a satellite link might be possible. Military forces operate a wide array of satellites (http://en.wikipedia.org/wiki/Military_satellite), and it is conceivable that high-bandwidth downlinks could be used to transfer ledger data (mostly incremental blocks, plus occasional full ledgers to allow recipients to synchronize when they missed some incremental blocks.)
  • Participants who generate transactions must be able to send them to the consensus building network. For participants with internet connections this is easy, those who rely on the satellite downlink would either have access to a low-bandwidth uplink or to other radio communications.
  • Encryption of transactions is an issue: On one hand, one would not want to encrypt messages using keys known by many participants due to the high risk of accidental or malicious key disclosure, on the other hand, transactions that rely on previous ledger data (such as orders utilizing an order budget account which must not be overdrawn by the ordering entity) need to be read by the consensus building system to be accepted or rejected.
  • For example, transactions sent to the consensus network could be encrypted with the public keys of both the intended recipient and the consensus building system.
  • One other possible approach would be to use a homomorphic encryption approach which allows the consensus building nodes to operate on the data without knowing what it actually means (see http://www2.technologyreview.com/news/423683/homomorphic-encryption/). However, I have no idea whether these approaches are usable for this kind of information processing yet.
2
  • Why did you mention Ripple specifically? Jun 23, 2016 at 22:51
  • Because its consensus based approach with short ledger close times (few seconds) seems to fit the bill better than the PoW or PoS algorithms used in many crypto currencies. And it's the only one I'm relatively familiar with, but of course there might be other algorithms which are just as good or better for this task. Jun 24, 2016 at 3:57
3

As I see it, the key part is at the very end: "traceable in a decentralized ledger". That feature isn't described in the EFF page you cite, and I'm not aware that any of the services listed provide it.

To see why this is important, imagine a scenario like the following:

Major Smith receives a message at 0400 hours from Battalion HQ about impending enemy action. But he's too busy playing Angry Birds and doesn't bother to pass it along to his units in the field. The enemy attacks at 1100 hours, the field units are unprepared, and they are defeated with heavy casualties. Major Smith is afraid he will be court-martialed for failing to share the intelligence with his troops. So he writes a message, backdated to 0500 hours, addressed to Captain Jones and ordering him to contact the field units and give them the information. He also secretly orders (or bribes) Corporal Johnson, an IT administrator, to alter the message system's logs to show his forged message as having been sent at the appropriate time. This way, he can blame Captain Jones for failing to act on the message, or at a minimum, plausibly claim there must have been some sort of communication failure that kept Jones from receiving it.

If there is a secure decentralized ledger of all messages sent, then Smith can't insert his fake backdated message in it, and there's no single person he can bribe in order to get it done. So he can't dodge responsibility for shirking his duty.

A secure decentralized ledger is precisely what a blockchain provides.

0

Not the answer you're looking for? Browse other questions tagged or ask your own question.