Efficient Constructions of Bilinear Accumulators Ioanna - - PowerPoint PPT Presentation

β–Ά
efficient constructions of bilinear accumulators
SMART_READER_LITE
LIVE PREVIEW

Efficient Constructions of Bilinear Accumulators Ioanna - - PowerPoint PPT Presentation

Efficient Constructions of Bilinear Accumulators Ioanna Karantaidou, Foteini Baldimtsi Set Me Membership ip Bank, GMU, subscription- based service, etc Alice List of members I am Alice ce ... List of members as a Data structure


slide-1
SLIDE 1

Efficient Constructions

  • f Bilinear

Accumulators

Ioanna Karantaidou, Foteini Baldimtsi

slide-2
SLIDE 2

Set Me Membership ip

List of members ... Alice ... ... Bank, GMU, subscription- based service, etc

I am Alice ce

Alice

List of members as a Data structure

  • Size of List: O(n)
  • (at least one of) Additions/Deletions,

lookups depends on n

  • Privacy against list

holder/membership verification in a privacy preserving way: Expensive!

slide-3
SLIDE 3

Accu ccumulator Setting

VERIFIER MANAGER Set S

Acc.v

Initialize & Create Acc.v Accumulator Value: holds Set S

slide-4
SLIDE 4

Posit sitiv ive Acc ccumula lator: ad addin ing Use ser x

VERIFIER MANAGER Set S

Acc.v

User x Wx UpdateAlg update message x Add(x) Update Acc.v

slide-5
SLIDE 5

Posit sitiv ive Acc ccumula lator: provi ving membership ip

VERIFIER MANAGER Set S

Acc.v

User x Wx UpdateAlg Wx access/service update message x Add(x) Update Acc.v Accumulator value size: O(1) Witness size: O(1) UpdateAlg: O(1) Membership Verification time: O(1)

slide-6
SLIDE 6

Secu curity y Propertie ies s (mem ember ership ip)

... Alice ... Charlie Alice

Accumulator acc

Verification algorithm:VerMem(π‘₯𝑦)

𝑦 ∈ 𝑏𝑑𝑑 β†’ VerMem(π‘₯𝑦)=1 𝑦 βˆ‰ 𝑏𝑑𝑑 β†’ VerMem(π‘₯𝑦)=0 (or =1 with negligible prob.) Set/List

Verification=lookup

Alice is a member β†’ verification Bob is not a member β†’ verification

correctness soundness

slide-7
SLIDE 7

2 2 Types es of Accu cumulators

RSA based accumulators [CL02, LLX07, BdM93]

  • Accumulate odd prime numbers
  • Factorization of group hidden
  • Strong RSA assumption

Bilinear Pairing based accumulators [N05, CKS09, ATSM09, ZKP17]

  • Accumulate integers
  • Known order groups
  • Witness, accumulator value belong in pairing friendly groups
  • q-SDH assumption

Choice depends on the application!

slide-8
SLIDE 8

Common Iss ssues es with Known Accu cumulators

  • Unnecessary accumulator updates that cause high communication costs
  • Expensive non-membership operations
  • Computational overhead due to extra properties

Can we do better if we take advantage of the presence of a trusted entity (manager)?

Di Discu cuss ssio ion on the se secr cret key y model

  • Most known constructions have a trusted setup
  • Anonymous Credentials, subscription-based services, etc
slide-9
SLIDE 9

Our Resu sults

  • 1. Positive

Bilinear Accumulator with Optimal Communication Cost

  • 2. Universal Bilinear Accumulator with Constant Non-

Membership Witness Creation

  • 3. ZK Accumulator with Constant Non-Membership

Witness Creation and Update

slide-10
SLIDE 10

FIRST CONSTRUCTION Positive Bilinear Accumulator with Optimal Communication Cost

slide-11
SLIDE 11

(sk)

Posit sitiv ive e Bilin linea ear Accu ccumulator

𝐡𝑑𝑑. 𝑀 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙 𝐡𝑑𝑑. 𝑀 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙 (π’š+𝑑𝑙)

User x

Set S

Add(x) π‘₯𝑦 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙 upmsg

slide-12
SLIDE 12

Posit sitiv ive e Bilin linea ear Accu ccumulator Verif ific icatio ion

𝐡𝑑𝑑. 𝑀 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙 (π’š+𝑑𝑙) π‘₯𝑦 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙

π‘₯𝑦 = 𝐡𝑑𝑑. 𝑀(𝑦+𝑑𝑙)βˆ’1 π‘₯𝑦

(𝑦+𝑑𝑙)

= 𝐡𝑑𝑑. 𝑀 Public parameters: 𝑕, 𝑕𝑑𝑙 , (𝑕𝑑𝑙)2, (𝑕𝑑𝑙)3, … β†’ π‘₯𝑦

(𝑦+𝑑𝑙)

e(π’™π’š, π’‰π’š 𝒉𝒕𝒍 )=e(𝑩𝒅𝒅. π’˜ , 𝒉) (VerMem) Public parameters: 𝑕, 𝑕𝑑𝑙 ,(𝑕𝑑𝑙)2, (𝑕𝑑𝑙)3,…→ 𝑕𝑦 , 𝑕𝑑𝑙

slide-13
SLIDE 13

(sk)

Posit sitiv ive e Bilin linea ear Accu ccumulator

𝐡𝑑𝑑. 𝑀 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙 𝐡𝑑𝑑. 𝑀 = 𝑕 𝑦1+𝑑𝑙 … π‘¦π‘œ+𝑑𝑙 (π’š+𝑑𝑙)

User x

Set S

Del(x) upmsg

slide-14
SLIDE 14

Posit sitiv ive e Bilin linea ear Accu ccumulator

Minimum communication bound (on update messages) for positive accumulators= |d| (number of deletions)

Camacho, Philippe, and Alejandro

  • Hevia. "On the impossibility of batch

update for cryptographic accumulators." International Conference on Cryptology and Information Security in Latin America. Springer, Berlin, Heidelberg, 2010.

slide-15
SLIDE 15

Posit sitiv ive Bilin linear Acc ccumulator with Optim imal l Communicatio ion Cost-Fir irst try

(sk)

𝐡𝑑𝑑. 𝑀 = 𝑕𝑣

User x Add(x) π‘₯𝑦 = 𝑕𝑣 (π’š+𝑑𝑙)βˆ’1

𝐡𝑑𝑑. 𝑀 = 𝑕𝑣 (π’š+𝑑𝑙)βˆ’1

Del(x) upmsg

slide-16
SLIDE 16

Posit sitiv ive Bilin linear Acc ccumulator with Optim imal l Communicatio ion Cost-Fir irst try

(sk)

𝐡𝑑𝑑. 𝑀 = 𝑕𝑣

User x Add(x) π‘₯𝑦 = 𝑕𝑣 (π’š+𝑑𝑙)βˆ’1

𝐡𝑑𝑑. 𝑀 = 𝑕𝑣 (π’š+𝑑𝑙)βˆ’1

Del(x) upmsg

  • Communication

efficient

  • Dynamic (add,del)
  • Positive

(membership)

  • Correctness

holds and VerMem same

  • Soundness??
slide-17
SLIDE 17

Posit sitiv ive Bilin linear Acc ccumulator with Optim imal l Communicatio ion Cost-Fir irst try

Proof overview:

  • R (public parameters) runs an adversary A (public parameters)
  • A submits lists of to-be-added, to-be-deleted elements 𝑀𝐡, 𝑀𝐸
  • R simulates updates and witnesses
  • A breaks acc soundness
  • R breaks q-SDH assumption

q-SDH: Given (p, 𝐻, π»π‘ˆ, 𝑓, 𝑕), {𝑕𝑑𝑙}𝑗,𝑗 = 0, … , π‘Ÿ there is negligible probability of finding 𝑕

1 𝑑𝑙+𝑦 for 𝑦 ∈ β„€π‘ž

slide-18
SLIDE 18

Posit sitiv ive Bilin linear Acc ccumulator with Optim imal l Communicatio ion Cost-Fir irst try

Proof overview:

  • R (public parameters) runs an adversary A (public parameters)
  • A submits lists of to-be-added, to-be-deleted elements 𝑀𝐡, 𝑀𝐸
  • R simulates updates and witnesses
  • A breaks acc soundness
  • R breaks q-SDH assumption

Adaptive soundness not achieved

slide-19
SLIDE 19

Positive Bilinear ar Ac Accumu mulator wi with Optimal al Commu mmunication Cost- Modular ar Construction

(x,r) in A- sound positive additive acc r in NA- sound positive dynamic acc A-sound positive dynamic acc

Baldimtsi, Foteini, et al. "Accumulators with applications to anonymity-preserving revocation." 2017 IEEE European Symposium on Security and Privacy (EuroS&P). IEEE, 2017.

slide-20
SLIDE 20

Positive Bilinear ar Ac Accumu mulator wi with Optimal al Commu mmunication Cost- Modular ar Construction

  • r=F(x), where F() is a

pseudorandom function

  • Updates for deletions

Communication cost= |d| Optimal! No updates for positive accumulator that supports additions only

(x,r) in A- sound positive additive acc r in NA- sound positive dynamic acc A-sound positive dynamic acc

slide-21
SLIDE 21
  • Jan Camenisch, Markulf Kohlweiss, and Claudio Soriente. An accumulator basedon bilinear maps and efficient revocation for anonymous credentials. In PKC 2009
  • Lan Nguyen. Accumulators from bilinear pairings and applications. In CT-RSA 2005.

Positive Camenisch et al 09 Nguyen 05 this work (NA- sound) this work (A- sound) Add 1 1 1 1 Del 1 1 1 1 MemWitCreate 1 1 1 1 NonMemWitCreate

  • MemWitUpOnAdd

1 1 MemWitUpOnDel 1 1 1 1 NonMemWitUpOnAd d

  • NonMemWitUpOnDe

l

  • VerMem

1 1 1 1 VerNonMem

  • Manager storage

1 1 1 1 Parameters 2q q q q

  • Com. cost

|a|+|d| |a|+|d| |d| |d| Efficient ZKPs βœ“ βœ“ βœ“ βœ“ Adaptively-sound βœ“ βœ“ βœ“

slide-22
SLIDE 22

SECOND CONSTRUCTION Universal Bilinear Accumulator with Constant Non-Membership Witness Creation

slide-23
SLIDE 23

Addit itional l Properties (non-mem embership ip:NM)

Accumulator acc

NM verification algorithm: VerNonMem(π‘₯𝑦)

𝑦 βˆ‰ 𝑏𝑑𝑑 β†’ VerNonMem(π‘₯𝑦)=1 𝑦 ∈ 𝑏𝑑𝑑 β†’ VerNonMem(π‘₯𝑦)=0 (or =1 with negligible prob.) Set/List

NM verification=lookup

Bob is not a member β†’ NM verification Alice is a member β†’ NM verification

Charlie Alice ... Alice ...

correctness soundness

slide-24
SLIDE 24

Ge Gener eric c Univ iversal l Mo Modular Construct ction motiv ivatio ion: Non membership ip for y

Users (public parameters): S={𝑦𝑗}, polynomial division Manager (sk): ς𝑗=1

|S| (𝑦𝑗+𝑑𝑙) ∈ β„€, used as exponent

a (ς𝑗=1

|S| 𝑧𝑗) + 𝑐 𝑧 = 1

Users (public parameters)/Manager (sk): ς𝑗=1

|S| 𝑧𝑗 ∈ β„€, Euclidean algorithm

Bilinear ATSM09, S={𝑦𝑗}, 𝑦𝑗 ∈ β„€π‘ž RSA LLX07, S={𝑧𝑗}, 𝑧𝑗 primes

slide-25
SLIDE 25

Ge Gener eric c Univ iversal l Mo Modular Construct ction motiv ivatio ion: Non membership ip for y

Users (public parameters): S={𝑦𝑗}, polynomial division Manager (sk): ς𝑗=1

|S| (𝑦𝑗+𝑑𝑙) ∈ β„€, used as exponent

a (ς𝑗=1

|S| 𝑧𝑗) + 𝑐 𝑧 = 1

Users (public parameters)/Manager (sk): ς𝑗=1

|S| 𝑧𝑗 ∈ β„€, Euclidean algorithm

Bilinear ATSM09, S={𝑦𝑗}, 𝑦𝑗 ∈ β„€π‘ž RSA LLX07, S={𝑧𝑗}, 𝑧𝑗 primes

non-membership cost: |S|

slide-26
SLIDE 26

Ge Gener eric c Univ iversal l Mo Modular Construct ction Over vervi view ew

Can we replace non-membership with constant-runtime membership?? Yes, with a trusted manager

A-sound positive dynamic acc for S A-sound positive dynamic acc for D-S A-sound universal dynamic acc for S

Can we make sure that 𝐡𝐷𝐷1and 𝐡𝐷𝐷2 are disjoint? The accumulator manager always signs the most up to date value of the accumulator

π‘©π‘«π‘«πŸ π‘©π‘«π‘«πŸ‘

slide-27
SLIDE 27

Ge Gener eric c Univ iversal l Mo Modular Construct ction

(sk)

π‘©π‘«π‘«πŸ. π‡πŸπ¨(πŸπ›, βˆ…) π‘©π‘«π‘«πŸ‘. π‡πŸπ¨(πŸπ›, 𝐄)

slide-28
SLIDE 28

Ge Gener eric c Univ iversal l Mo Modular Construct ction

(sk) User x Add(x) π‘₯𝑦 = 𝐡𝐷𝐷1.π‘₯

π‘©π‘«π‘«πŸ (π‘»πŸ) π‘©π‘«π‘«πŸ‘ (π‘»πŸ‘) 𝑦 ∈ 𝑇2 π‘©π‘«π‘«πŸ (π‘»πŸΪ‚{π’š}) π‘©π‘«π‘«πŸ‘ (π‘»πŸ‘ βˆ’ {π’š})

π‘©π‘«π‘«πŸ. π›πžπž(𝐲) π‘©π‘«π‘«πŸ‘. 𝐞𝐟𝐦(𝐲)

slide-29
SLIDE 29

Ge Gener eric c Univ iversal l Mo Modular Construct ction

(sk) User x Del(x) π‘₯𝑦 = 𝐡𝐷𝐷2. π‘₯

π‘©π‘«π‘«πŸ (π‘»πŸ) π‘©π‘«π‘«πŸ‘ (π‘»πŸ‘) 𝑦 ∈ 𝑇1 π‘©π‘«π‘«πŸ‘ (π‘»πŸ‘Ϊ‚{π’š}) π‘©π‘«π‘«πŸ (π‘»πŸ βˆ’ {π’š})

π‘©π‘«π‘«πŸ. 𝐞𝐟𝐦(𝐲) π‘©π‘«π‘«πŸ‘. π›πžπž(𝐲)

π‘₯ π‘›π‘“π‘›π‘π‘“π‘ π‘‘β„Žπ‘—π‘ž π‘₯ (non-membership)

slide-30
SLIDE 30

Ge Gener eric c Univ iversal l Mo Modular Construct ction

User x π‘₯𝑦

VERIFIER π‘©π‘«π‘«πŸ.π–πŸπ¬ππŸπ§(𝐱𝐲) π‘©π‘«π‘«πŸ‘.π–πŸπ¬ππŸπ§(π‘₯𝑧)

π‘₯𝑧 User y

O(1)

slide-31
SLIDE 31

Ge Gener eric c Univ iversal l Mo Modular Construct ction

Note on Efficiency Concretes:

  • Generation (run once) linear

to Domain size

  • Add/Del of double cost

Asymptotics: All operations constant, independent of accumulated set S

slide-32
SLIDE 32

Ge Gener eric c Univ iversal l Mo Modular Construct ction- Soundnes ess

Theorem: A combination of accumulators 𝐡𝐷𝐷1, 𝐡𝐷𝐷2 is a universal dynamic adaptively-sound accumulator if 𝐡𝐷𝐷1, 𝐡𝐷𝐷2 are positive dynamic adaptively-sound accumulators of domain D and one is holding SβŠ‚D and the other one 𝑇 βŠ‚D and public updates are not permitted.

slide-33
SLIDE 33

Ge Gener eric c Univ iversal l Mo Modular Construct ction Soundnes ess

Theorem: A combination of accumulators 𝐡𝐷𝐷1, 𝐡𝐷𝐷2 is a universal dynamic adaptively-sound accumulator if 𝐡𝐷𝐷1, 𝐡𝐷𝐷2 are positive dynamic adaptively-sound accumulators of domain D and one is holding SβŠ‚D and the other one 𝑇 βŠ‚D and public updates are not permitted.

INTUITION: Information obtained by 2 accumulators with the same instantiation could be obtained by different states of 1 accumulator

slide-34
SLIDE 34

Ge Gener eric c Univ iversal l Mo Modular Construct ction Soundnes ess

Theorem: A combination of accumulators 𝐡𝐷𝐷1, 𝐡𝐷𝐷2 is a universal dynamic adaptively-sound accumulator if 𝐡𝐷𝐷1, 𝐡𝐷𝐷2 are positive dynamic adaptively-sound accumulators of domain D and one is holding SβŠ‚D and the other one 𝑇 βŠ‚D and public updates are not permitted.

INTUITION: Information obtained by 2 accumulators with the same instantiation could be obtained by different states of 1 accumulator PROOF: R has access to Add/Del oracle. A breaks ACC=(𝐡𝐷𝐷1, 𝐡𝐷𝐷2) soundness. R breaks 𝐡𝐷𝐷1 (positive) soundness

slide-35
SLIDE 35

Effic icien ency y Results

Positive Universal Camenischet al 09 Nguyen 05 this work (A- sound) Au et al 09 This work- Instantiation with Nguyen 05 Add 1 1 1 1 1 Del 1 1 1 1 1 MemWitCreate 1 1 1 1 1 NonMemWitCreate

  • |S|

1 MemWitUpOnAdd 1 1 1 1 MemWitUpOnDel 1 1 1 1 1 NonMemWitUpOnAdd

  • 1

1 NonMemWitUpOnDel

  • 1

1 VerMem 1 1 1 1 1 VerNonMem

  • 1

1 Manager storage 1 1 1 |S| 1 Parameters 2q q q q q?

  • Com. cost

|a|+|d| |a|+|d| |d| |a|+|d| |a|+|d| Efficient ZKPs βœ“ βœ“ βœ“ βœ“ βœ“ Adaptively-sound βœ“ βœ“ βœ“ βœ“ βœ“

slide-36
SLIDE 36

THIRD CONSTRUCTION ZK Accumulator with Constant Non-Membership Witness Creation and Update

slide-37
SLIDE 37

Appli licatio ion: ZK Accu cumula lator

Esha Ghosh , Olga Ohrimenko , Dimitrios Papadopoulos , Roberto Tamassia and Nikos Triandopoulos "Zero-Knowledge Accumulators and Set Operations" IACR Cryptology ePrint Archive 2015 (2015): 404.

MANAGER

Acc.v Member/non- member, witness x?

S ??

Adv

slide-38
SLIDE 38

Appli licatio ion: ZK Accu cumula lator

MANAGER

Member/non- member, witness x?

S ??

Goal:

  • Adv can learn only the latest query

answers How:

  • Randomness in the exponent

Adv

Acc.v

slide-39
SLIDE 39

Appli licatio ion: ZK Accu cumula lator- Wh Why

  • 1. Verifier is ZK

adversary

  • 2. ZK

accumulator is managed by a trusted entity

  • 3. Randomness

harms efficiency

slide-40
SLIDE 40

Ap Application: : ZK Ac Accumu mulator- the need fo for ran andomness

MANAGER

Acc.v=𝑕𝑠(𝑦1+𝑑𝑙)(𝑦2+𝑑𝑙)

S = {𝑦1, 𝑦2}

Acc.v=𝑕𝑠 𝒔′(𝑦1+𝑑𝑙)(𝑦2+𝑑𝑙)(π’š+𝑑𝑙)

S = {𝑦1, 𝑦2, π’š}

Add(x) Public parameters: 𝑕, 𝑕𝑑𝑙 ,(𝑕𝑑𝑙)2,(𝑕𝑑𝑙)3,… 𝑇 = {𝑦1, 𝑦2} 𝑏𝑑𝑑. 𝑀 = 𝑕(𝑦1+𝑑𝑙)(𝑦2+𝑑𝑙) = 𝑕𝑦1𝑦2+(𝑦1+𝑦2)𝑑𝑙+𝑑𝑙2 = 𝑕𝑦1𝑦2(𝑕𝑑𝑙)𝑦1+𝑦2(𝑕𝑑𝑙)2

  • 1. First element’s witness is g (generator is public information)
  • 2. A guess about S can be verified with public information
  • 3. A witness can be updated with public information (still valid?)
slide-41
SLIDE 41

Ap Application: : ZK Ac Accumu mulator- the need fo for ran andomness

MANAGER

Acc.v=𝑕𝑠(𝑦1+𝑑𝑙)(𝑦2+𝑑𝑙)

S = {𝑦1, 𝑦2}

Acc.v=𝑕𝑠 𝒔′(𝑦1+𝑑𝑙)(𝑦2+𝑑𝑙)(π’š+𝑑𝑙)

Sβ€² = {𝑦1, 𝑦2, π’š}

aux’={rr’}

Set S (aux=r)

Add(x)

S ?? r ?? Verification works!

slide-42
SLIDE 42

Ap Application: : ZK Ac Accumu mulator- Non-membership usual ally

𝑕𝑣 𝑕𝑠𝑣

Non-membership verification requires r

Users (public parameters): S={𝑦𝑗}, polynomial division Manager (sk): ς𝑗=1

|S| (𝑦𝑗+𝑑𝑙) ∈ β„€, used as exponent

Bilinear ATSM09, S={𝑦𝑗}, 𝑦𝑗 ∈ β„€π‘ž

slide-43
SLIDE 43

Ap Application: : ZK Ac Accumu mulator- Non-membership by Ghosh et al al

𝑇 ∩ {𝑦} β‰  βˆ… π‘₯𝑧 = 𝑋

1, 𝑋 2 = (𝑕(π‘Ÿ1 𝑑𝑙 +𝛿(𝑧+𝑑𝑙)) π’”βˆ’1, π‘•π‘Ÿ2 𝑑𝑙 βˆ’π›Ώ ς𝑗=1

S (𝑦𝑗+𝑑𝑙) )

𝑓 𝑋

1, 𝑩𝒅𝒅. π’˜ 𝑓 𝑋 2,𝑕𝑦𝑕𝑑𝑙 = 𝑓(𝑕, 𝑕)

Remove accumulator randomness

slide-44
SLIDE 44

Ap Application: : ZK Ac Accumu mulator- Non-membership by Ghosh et al al

𝑇 ∩ {𝑦} β‰  βˆ… π‘₯𝑧 = 𝑋

1, 𝑋 2 = (𝑕(π‘Ÿ1 𝑑𝑙 +𝜹(𝑧+𝑑𝑙)) π‘ βˆ’1, π‘•π‘Ÿ2 𝑑𝑙 βˆ’πœΉ ς𝑗=1

S (𝑦𝑗+𝑑𝑙) )

𝑓 𝑋

1, 𝐡𝑑𝑑. 𝑀 𝑓 𝑋 2, 𝑕𝑦𝑕𝑑𝑙 = 𝑓(𝑕, 𝑕)

Add query/witness specific randomness

slide-45
SLIDE 45

Ap Application: : ZK Ac Accumu mulator- Non-membership by Ghosh et al al

𝑇 ∩ {𝑦} β‰  βˆ… π‘₯𝑧 = 𝑋

1, 𝑋 2 = (𝑕(π‘Ÿ1 𝑑𝑙 +𝜹(𝑧+𝑑𝑙)) π‘ βˆ’1, π‘•π‘Ÿ2 𝑑𝑙 βˆ’πœΉ ς𝑗=1

S (𝑦𝑗+𝑑𝑙) )

𝑓 𝑋

1, 𝐡𝑑𝑑. 𝑀 𝑓 𝑋 2, 𝑕𝑦𝑕𝑑𝑙 = 𝑓(𝑕, 𝑕)

Add query/witness specific randomness

(+) r not needed for verification (-) no witness update algorithm Update β†’ NonMemWitCreate: O(|S|)

slide-46
SLIDE 46

Ap Application: : ZK Ac Accumu mulator- Modular ar Construction

𝑇 ∩ {𝑦} β‰  βˆ… π‘₯𝑧 = 𝑋

1, 𝑋 2 = (𝑕(π‘Ÿ1 𝑑𝑙 +𝜹(𝑧+𝑑𝑙)) π‘ βˆ’1, π‘•π‘Ÿ2 𝑑𝑙 βˆ’πœΉ ς𝑗=1

S (𝑦𝑗+𝑑𝑙) )

𝑓 𝑋

1, 𝐡𝑑𝑑. 𝑀 𝑓 𝑋 2, 𝑕𝑦𝑕𝑑𝑙 = 𝑓(𝑕, 𝑕)

Add query/witness specific randomness

(+) r not needed for verification (-) no witness update algorithm Update β†’ NonMemWitCreate: O(|S|)

Solution: instantiate our generic modular universal construction with ZK accumulators with membership operations Result: Non-membership witness creation, Non- membership witness update: O(1)

slide-47
SLIDE 47

Summary

In the secret key model:

  • 1. We can hit optimal communication cost (Positive Bilinear

Accumulator with Optimal Communication Cost)

  • 2. We can have constant non-membership (Universal Bilinear

Accumulator with Constant Non-Membership Witness Creation)

  • 3. We can have constant ZK (ZK Accumulator with Constant

Non-Membership Witness Creation and Update)

slide-48
SLIDE 48

Available on ePrint soon