ZIP: 231
Title: Memo Bundles
Owners: Jack Grigg <jack@electriccoin.co>
        Kris Nuttycombe <kris@electriccoin.co>
        Daira-Emma Hopwood <daira@electriccoin.co>
        Arya Solhi <arya@zfnd.org>
Credits: Sean Bowe
         Nate Wilcox
Status: Draft
Category: Consensus / Wallet
Created: 2024-04-26
License: MIT
Discussions-To: <https://github.com/zcash/zips/issues/627>

Terminology

The key words “MUST”, “MUST NOT”, “SHOULD”, and “MAY” in this document are to be interpreted as described in BCP 14 ¹ when, and only when, they appear in all capitals.

The character § is used when referring to sections of the Zcash Protocol Specification. ²

Abstract

Currently, the memo sent in a shielded output is limited to at most 512 bytes. This ZIP proposes to allow larger memos, and to enable memo data to be shared between multiple recipients of a transaction.

Motivation

In Zcash transaction versions v2 to v5 inclusive, each Sapling or Orchard shielded output contains a ciphertext comprised of a 52-byte note plaintext, and a corresponding 512-byte memo field. ³ Recipients can only decrypt the outputs sent to them, and thus can also only observe the memo fields included with the outputs they can decrypt.

The shielded transaction protocol hides the sender(s) (that is, the addresses corresponding to the keys used to spend the input notes) from all of the recipients. For certain kinds of transactions, it is desirable to make one or more sender addresses available to one or more recipients (for example, a reply address). In such circumstances it is important to authenticate the sender addresses, to give the recipient a guarantee that the address is controlled by a sender of the transaction; failure to authenticate this address can enable phishing attacks. These Authenticated Reply Addresses require zero-knowledge proofs, and for the Orchard protocol these proofs are too large to fit into a 512-byte memo field.

It is also desirable, for clients with more stringent bandwidth constraints, to be able to transmit encrypted notes to the client without including the encrypted memo data. In the current light client protocol ⁴, this is done by truncating the note ciphertext to just the part that encrypts the memo. However, that has the effect of truncating the authentication tag, and so the resulting decryption algorithm does not meet standard security notions for an authenticated encryption scheme. It is a goal of this proposal to rectify this, simplifying the security argument.

Instead of the memo data, this ZIP proposes that it is possible to indicate whether a memo is present for the recipient. When using the light client protocol, a recipient need not download full transaction information if this indication tells them that they have not received any memo in the transaction.

At present, it is not possible to transmit the same memo data to multiple transaction recipients without redundantly encoding that data, and sending memo data greater than 512 bytes requires sending multiple outputs; the problem is compounded when attempting to send more than 512 bytes to each recipient. By separating memo data from the decryption capability for those memos, it admits a greater variety of applications that utilize memo data, while decreasing the amount of data that needs to be stored on-chain overall.

Requirements

• Recipients can receive memo data that is greater than 512 bytes in length.
• Multiple recipients, across any of the shielded pools, can be given the capability to view the same memo data.
• The exact number and exact lengths of distinct decryptable memos should not be revealed, even to the transaction recipients, although an upper bound on the total length of memo data that the observer does not have the capability to view will be leaked to transaction recipients, and the overall maximum possible length of memo data will be revealed on-chain.
• A recipient can determine whether or not they have been given the capability to view any memo solely by decrypting the note ciphertext.
• Memo chunks within a transaction can be individually pruned from block storage without preventing the transaction from being verified when transmitting block data to peers.
• The ciphertext of the note alone can be decrypted using a scheme that meets a standard security notion for authenticated encryption, without more than a small constant overhead in ciphertext size.

Non-requirements

• Recipients do not need to be able to receive multiple memos per note. This capability can however be enabled under the existing proposal by “chaining” memos, including the decryption key for another memo within the memo that is decryptabble by a recipient.

Specification

Since this proposal is defined only for v6 and later transactions, it is not necessary to consider Sprout JoinSplit outputs. The following sections apply to both Sapling and Orchard outputs.

Changes to the Zcash Protocol Specification

The following changes affecting the definitions of note plaintexts and note ciphertexts, and the algorithms for encryption and decryption.

In § 3.2.1 ‘Note Plaintexts and Memo Fields’:

• Change

  Each Sapling or Orchard note plaintext (denoted np) consists of

     (leadByte ⦂ 𝔹^𝕐, d ⦂ 𝔹^[ℓ_d], rseed ⦂ 𝔹^{𝕐[𝟛𝟚]}, memo ⦂ 𝔹^{𝕐[𝟝𝟙𝟚]})

  to

  The form of a Sapling or Orchard note plaintext depends on the version of the transaction in which it will be included; specifically whether that version is pre-v6, or v6-onward.

  Each pre-v6 Sapling or Orchard note plaintext (denoted np) consists of

     (leadByte ⦂ 𝔹^𝕐, d ⦂ 𝔹^[ℓ_d], rseed ⦂ 𝔹^{𝕐[𝟛𝟚]}, memo ⦂ 𝔹^{𝕐[𝟝𝟙𝟚]})

  Each v6-onward Sapling or Orchard note plaintext (denoted np) consists of

     (leadByte ⦂ 𝔹^𝕐, d ⦂ 𝔹^[ℓ_d], rseed ⦂ 𝔹^{𝕐[𝟛𝟚]}, K^memo ⦂ 𝔹^{𝕐[𝟛𝟚]})

In § 5.5 ‘Encodings of Note Plaintexts and Memo Fields’ ⁵:

• Change the paragraph that describes “The encoding of a Sapling or Orchard note plaintext” to refer to “The encoding of a pre-v6 Sapling or Orchard note plaintext”.

• Add a new paragraph at the end of the section:

  The encoding of a v6-onward Sapling or Orchard note plaintext consists of:

  8-bit leadByte

  88-bit d

  64-bit v

  256-bit rseed

  32-byte Kmemo

  • A byte 0x03, indicating this version of the encoding of a v6-onward Sapling or Orchard note plaintext.
  • 11 bytes specifying d.
  • 8 bytes specifying v.
  • 32 bytes specifying rseed.
  • 32 bytes specifying K^memo.

  A value consisting of 32 0xFF bytes for K^memo is used to indicate that there is no memo for this note plaintext.

In § 4.7.2 ‘Sending Notes (Sapling)’ ⁶ and § 4.7.3 ‘Sending Notes (Orchard)’ ⁷:

• Add a reference to this ZIP specifying the construction of the memo bundle and derivation of K^memo in the case of a v6-onward note plaintext.

• Change

  Let np = (leadByte, d, v, rseed, memo) .

  to

  Let np be the encoding of a Sapling note plaintext using leadByte, d, v, rseed, and either memo for a pre-v6 note plaintext or K^memo for a v6-onward note plaintext.

  replacing “Sapling” with Orchard in the case of § 4.7.3.

In § 4.20.1 ‘Encryption (Sapling and Orchard)’ ⁸:

• Change

  Let np = (leadByte, d, v, rseed, memo) be the Sapling or Orchard note plaintext. np is encoded as defined in § 5.5 ‘Encodings of Note Plaintexts and Memo Fields’.

  to

  Let np be the encoding of the Sapling or Orchard note plaintext (which may be pre-v6 or v6-onward), as defined in § 5.5 ‘Encodings of Note Plaintexts and Memo Fields’.

• Add another normative note to that section:

  • C^enc will be of length either 580 or 100 bytes, depending on whether np is a pre-v6 or v6-onward note plaintext.

In § 4.20.2 ‘Decryption using an Incoming Viewing Key (Sapling and Orchard)’ ⁹ and § 4.20.3 ‘Decryption using a Full Viewing Key (Sapling and Orchard)’ ¹⁰:

• Replace memo ⦂ 𝔹^{𝕐[𝟝𝟙𝟚]} with memoOrKey.
• Specify that the type of memoOrKey is 𝔹^{𝕐[𝟝𝟙𝟚]} when decrypting a pre-v6 note ciphertext, or 𝔹^{𝕐[𝟛𝟚]} when decrypting a v6-onward note ciphertext. In the latter case, it is used as K^memo to decrypt the memo bundle as described in subsequent sections of this ZIP.

Memo bundle

A memo bundle consists of a sequence of 256-byte memo chunks, each individually encrypted. These memo chunks represent zero or more encrypted memos.

Each transaction may contain a single memo bundle, and a memo bundle may contain at most memo_chunk_limit memo chunks. This limits the total amount of memo data that can be conveyed within a single transaction to memo_chunk_limit * 256 bytes.

memo_chunk_limit is a parameter to this specification, to be decided upon by the community. The authors of this ZIP propose a maximum of 64 chunks, resulting in a maximum total memo data length of 16 KiB.

Memo bundles are encoded in transactions in a prunable manner: each memo chunk can be replaced by its representative digest.

Memo encryption

During transaction construction, each output with memo data is assigned a 32-byte memo key K^memo. These keys SHOULD be generated randomly, and MUST NOT be used to encrypt more than one memo within a single transaction. If an output has no memo data, it is assigned the memo key consisting of 32 0xFF bytes.

In note plaintexts of v6-onward transactions, the 512-byte memo field is replaced by K^memo .

The transaction builder generates a 32-byte salt value salt from a CSPRNG. A new salt MUST be generated for each memo bundle.

The symmetric encryption key for a memo is derived from its K^memo as follows:

   encryption_key = PRFKmemoexpand([0xE0] || salt)

The first byte 0xE0 should be added to the documentation of inputs to PRF^expand in § 4.1.2 ‘Pseudo Random Functions’ ¹¹.

If the generated key is 32 0xFF bytes, the transaction constructor MAY repeat this procedure with a different salt, in order to avoid the recipient misinterpreting the output as having no memo data. Since that has negligible probability, it alternatively MAY omit this check.

Each memo is padded to a multiple of 256 bytes with zeroes, and split into 256-byte chunks. Each memo chunk is encrypted with ChaCha20Poly1305 ¹² as follows:

   IETF_AEAD_CHACHA20_POLY1305(encryption_key, nonce, memo_chunk)

where nonce = I2BEOSP₈₈(counter)||[final_chunk] .

This is a variant of the STREAM construction ¹³.

• counter is a big-endian chunk counter starting at zero and incrementing by one for each subsequent chunk within a particular memo.
• final_chunk is the byte 0x01 for the final memo chunk, and 0x00 for all preceding chunks.

Finally, the encrypted memo chunks for all memos are combined into a single sequence using an order-preserving shuffle. Memo chunks from different memos MAY be interleaved in any order, but memo chunks from the same memo MUST have the same relative order. The following diagram shows an example shuffle of three memos:

[
    (memo_a, 0),
    (memo_b, 0),
    (memo_a, 1),
    (memo_c, 0),
    (memo_c, 1),
    (memo_a, 2),
]

Memo decryption

When a recipient decrypts a shielded output, they obtain a memo key K^memo. From this they derive encryption_key as above, and then proceed as follows:

• Set counter = 0 and final_chunk = 0x00.
• Attempt to decrypt each memo chunk in order. Pruned memo chunks are skipped.
• Each time decryption succeeds for a memo chunk, increment counter by 1, and then continue attempting to decrypt subsequent chunks.
• Once all memo chunks have been trial-decrypted once, set final_chunk = 0x01 and then attempt to decrypt the memo chunks again, starting immediately after the last successfully-decrypted chunk (or at the start if none were), and without incrementing counter.
  • This step can be made secret-independent by attempting to decrypt every memo chunk again, and ignoring the results of all chunks up to and including the last successfully-decrypted chunk.
• If no memo chunk decrypts successfully with final_chunk = 0x01, discard any memo chunks that were decrypted, and return nothing. Otherwise, concatenate the decrypted memo chunks in order and return the concatenation as the memo.

If any chunk of the memo encrypted to memo_key has been pruned, the decryption process above returns nothing (as final_chunk will be set to 0x01 with the wrong counter value), ensuring that a malformed memo is not returned.

Encoding in transactions

† These fields are present if and only if fAllPruned == 0.

If fAllPruned == 0, then:

• nonceOrHash represents the nonce for deriving encryption keys.
• Each bit of pruned, in little-endian order, indicates the type of the corresponding entry in vMemoChunks. A bit value of 0 indicates that the entry will be of type byte[272] representing an encrypted memo chunk. A bit value of 1 indicates the entry will be a byte[32] and contains the memo_chunk_digest for a pruned chunk.

If fAllPruned == 1, then:

• nonceOrHash represents the overall hash for the memo bundle as defined in Transaction sighash.
• The nMemoChunks, pruned, and vMemoChunks fields will be absent.

Transaction sighash

memo_chunk_digest = H(AEAD(MemoChunk, memo_key))
memo_bundle_digest = H(concat(memo_chunk_digests))

The memo bundle digest structure is a performance optimization for the case where all memo chunks in a transaction have been pruned.

TODO: finish this to be a modification to the equivalent of ZIP 244 for transaction v6.

Transaction fees

(This section will become a modification to ZIP 317.)

A memo bundle may contain two free chunks if there are any shielded outputs in the transaction. Otherwise, each memo chunk requires marginal_fee as defined in ZIP 317 ¹⁴.

Network protocol

Nodes must reject GetData responses having an fAllPruned value that is nonzero, or any byte of pruned that is nonzero.

Open issues

Limit on the number of memo chunks

memo_chunk_limit == 64 is recommended. This results in a maximum of 16 KiB of memo data per transaction.

Rationale

Memo bundle size restriction

Restricting the total amount of memo data in a bundle, for example to 16 KiB, limits the rate at which the chain size can grow cheaply (from a computational perspective; memo bundles are much easier to produce than proofs or signatures).

The current behaviour for previous transaction versions (no limit on the number of memos) is not altered by this ZIP, because memos in those transactions are tied to individual shielded outputs (incurring their computational cost), and are not natively aggregatable.

Memo chunk size

TODO: this table needs to be recalculated with a 16 KiB limit.

With 10 KiB limit on amount of memo data as the constant in this table, the maximum number of unique memos you can create, and the cost in bytes of that memo data plus auth when using a 32-byte memo key, is:

In the “256 20-out” case you have a distinguisher compared to old transactions, in that you can tell the transaction is sending at most 256 bytes per recipient rather than 512 if it is sending the max number of memos. But that’s inherently baked into the decision to use a smaller memo chunk size (and it is still possible for the chunks to all be a single memo sent to all outputs, or anything in between).

Memo key size

If we used a 16-byte memo key instead of 32 bytes, the transaction size overhead becomes:

The decrease in overhead is relatively modest in most cases, but more noticeable for small memos with a 256-byte memo chunk.

However, 128-bit keys don’t meet Zcash’s target security level of 125 bits, as argued in ¹⁵.

The benefits of 256-bit keys are:

• They incur only a small transaction size overhead above the minimum key size that would meet the target security level.
• This key length matches what we already use elsewhere for symmetric keys.

Encryption format

Including a per-transaction salt in the derivation of encryption_key gives protection against accidental (or intentional) reuse of K^memo reuse across multiple transactions. We do not protect against K^memo reuse within a transaction; it is up to the transaction builder to ensure that the same K^memo is not used to encrypt two different memos (and if they did so, normal clients would either never observe the second memo, or would decrypt parts of each memo and get a nonsensical and potentially insecure “spliced” memo).

We do not include commitments to the shielded outputs in the derivation of encryption_key for two reasons:

• It would force the transaction builder to fully define all shielded outputs before encrypting the memos, which might prevent potential use cases of PCZTs ¹⁶.
• We don’t want to unnecessarily prevent the ability to create a transaction with a memo bundle and no shielded outputs, as there may be use cases for, e.g. a fully-transparent transaction with encrypted memo, or a ZSA issuance transaction with exposed memo data using a well-known K^memo.

Pruned encoding

The separation of memo data from note data, and the new ability to easily store variable-length memo data, opens up an attack vector against node operators for storing arbitrary data. The transaction digest commitments to the memo bundle are structured such that if a node operator is presented with a memo key (i.e. they are given the capability to decrypt a particular memo), they can identify and prune the corresponding memo chunks, while still enabling the transaction to be validated as part of its corresponding block and broadcast over the network.

The transaction encoding permits pruning at the individual chunk level in order to facilitate pruning an individual memo from a transaction without affecting the other memos. This enables node operators to be responsive to, for example, GDPR deletion requests.

Note that broadcasting a partially-pruned transaction means that the pruned chunks no longer contribute to the upper bound on memo data.

The prunable structure does not introduce a censorship axis; memo bundles do not reveal which memo chunks correspond to which memos, and therefore a network adversary cannot selectively censor individual memos. They can censor any/all chunks within specific transactions, however shielded transactions do not reveal their senders, recipients, or amounts, and thus also cannot be individually targeted for censorship.

Transaction fees

Making the fee linear in the number of chunks has the following properties:

• The required fee to add more memo chunks scales at the same rate as adding logical actions, so it isn’t a cheaper mechanism for an adversary to bloat chain size.
• A “baseline transaction” (one spent note, one output to an external recipient with a memo, one change output without a memo) has the same fee as before.
• A “broadcast transaction” (many outputs to different recipients all given the same memo) is the same fee as before (but a smaller transaction).
• A “many memos transaction” (many outputs to different recipients all with unique memos) is at most around twice the fee as before.

Combined with the memo bundle size restriction, the maximum additional fee for a memo bundle over prior transactions is 0.0019 ZEC.

Reference implementation

TBD

References