12E/ IoT Swarms + SSI in constrained networks

IoT Swarms + SSI in Constrained Networks

Wednesday 12E

Convener: Geovane Fedrecheski

Notes-taker(s): Geovane Fedrecheski

Tags for the session - technology discussed/ideas considered: IoT Swarms, SSI in IoT,

Low-overhead SSI constructs

Discussion notes, key understandings, outstanding questions, observations, and, if appropriate to this discussion: action items, next steps: 

Presentation link -- here.

Summary: This session was a discussion about three topics: IoT Swarms, the challenges of SSI in constrained networks, and preliminary results on how to overcome them. The results showed that, while a DIDComm message with a DID Document as payload used almost 1 kilobyte, a binary approach can be used to cut it to just about 200 bytes.

IoT Swarms enable resource sharing among autonomous IoT devices. The presenter mentioned some papers published in this regard [1][2], including one that analyses using SSI in IoT and Swarm systems [3].

One of the challenges identified by this last paper is the overhead of using SSI, which poses a challenge for adoption on constrained IoT networks. For example, while the Long Range (LoRa) communication, often used in IoT systems, only allows payloads of up to 240 bytes, a single DID Document typically occupies 500 bytes or more. Similarly, messages using DIDComm tend to use at least 1 kilobyte, which prevents its use on constrained networks.

The presenter then showed the results of his exploration to build a binary version of DID Documents and DIDComm, and presented comparisons regarding message footprint, as shown in the Figures 1 and 2 below.

[[File:./media/image1.png|403x277px]]

Figure 1. Binary versions of DIDComm and DID Documents are needed to allow transmission in LoRa networks. The payload, in blue, is a DID Document. The overhead, in orange, is the protocol overhead due to the message signature.

[[File:./media/image2.png|389x193px]]

Figure 2. A binary version of DIDComm is needed to allow transmission of regular payloads in LoRa networks. In this example, the payload is a random CBOR-encoded message with 21 bytes. The overhead, in orange, is the protocol overhead due to the message signature and encryption.

In this session, the presenter did not share the methods, i.e., how to reduce message sizes, since they are part of a new and ongoing paper. Once the paper is published, however, the methods will be disclosed and shared with the IIW community.

[1] De Biase, Laisa Caroline Costa, et al. "Swarm economy: a model for transactions in a distributed and organic iot platform." IEEE Internet of Things Journal 6.3 (2018): 4561-4572.

[2] Fedrecheski, Geovane, et al. "SmartABAC: enabling constrained IoT devices to make complex policy-based access control decisions." to appear in the IEEE Internet of Things Journal (2021).

[3] Fedrecheski, Geovane, et al. "Self-sovereign identity for IoT environments: a perspective." 2020 Global Internet of Things Summit (GIoTS). IEEE, 2020.