Secure Signal Processing in The Cloud: (Enabling Technologies For Privacy-Preserving Multimedia Cloud Processing)

Secure Signal Processing in the
Cloud
[Enabling technologies for privacy-preserving multimedia cloud processing]
Under the Guidence of

Mr.J.Sudhakar
Technical seminar
By
N.Mamatha
(12RO1D5826)
INTRODUCTION
From a technological point of view, there are
currently some challenges that multimedia clouds
still need to tackle to be fully operational.
The most important issues that can hold back the
widespread adoption of the cloud, and of any
outsourcing scenario in general, are actually security
and privacy.
Both concepts are very close to each other in the
cloud as there can be no privacy without security.
Privacy is a more specific requirement, and it is
related only to sensitive data and/or processes.
CLOUD AND PRACTICAL SCENARIOS

Cloud computing comprises the provision of
computing and
storage services to users through the Internet
A public cloud provides determined services to
individuals and
organizations
cloud services are presented in three layers
1. Infrastructure as a service (IaaS)
2. Platform as a service (PaaS)
3. Software as a service (SaaS)
Service Layers in cloud

Infrastructure as a service (IaaS): The lower
architectural layer, representing the raw cloud

hardware resources.
Platform as a service (PaaS): The second layer,
usually providing an application server and a
database in which the customer can develop and run
his/her own applications.
Software as a service (SaaS): This is the most
extended use of the cloud as such, where a series of
applications are provided to the final user.
At any of these three layers of service provision, there are

three essential functionalities that summarize the purpose of a
cloud: storage, transmission, and computation
Functional roles played by the

stakeholders of a cloud system from a
privacy aware perspective:
Cloud infrastructure: All the storage and
processing outsourced to the cloud takes place at
the cloud infrastructure.
Software/application provider: Software
developers that
produce applications run on top of the cloud
infrastructure/
platform and are offered as a SaaS to end users.
Payment provider: Party in charge of billing for
the consumed
cloud resources
Customers/end users: Parties that contract
and make use
of a service on the cloud infrastructure
Exemplifying scenarios of signal

processing applications showcasing the
privacy problem in the cloud:
1.
Outsourced biometric recognition: In a

biometric recognition system, some biometric
information is contrasted against the templates
stored in a database.
2.
e-Health: e-Health is the paradigmatic scenario

where sensitive signals are involved, and the need
for privacy is always present there.
3.
Outsourced adaptive or collaborative

filtering: Filtering itself is an essential block of
signal processing, present in any system we can
think of, from voice processing in a smartphone to
complex volume rendering in the production of
asynthetic three-dimensional movie.
FORMALIZING THE PRIVACY

PROBLEM:
Any party that participates in a secure protocol
interchanges and obtains a series of intermediate
messages until the protocol finalizes and the result is
provided to the appropriate parties.
The set of all these messages conforms the transcript of
the protocol, and the untrusted parties may use it for
dishonest purposes.
Semihonest parties follow the predetermined protocol
without any deviation
Malicious parties will deviate from the protocol and may
introduce fake data
Typically, privacy-preserving protocols tackle the
semihonest case and present further extensions for
dealing with the more realistic malicious case.
MEASURING PRIVACY
Once the adversary model is defined- a proper privacy
framework should clearly establish means to quantitatively
measure the privacy, or,
conversely, the private information leakage.
The evaluation of the leakage that a given protocol produces
determines its suitability for a set of privacy requirements.
The typical cryptographic measures for security and secrecy
usually rely on complexity theory and hardness assumptions.
signal processing measures for the conveyed information in a
signal are based on fundamental information-theoretic
magnitudes.
PRIVACY TOOLS FROM SPED

Homomorphic processing
Searchable encryption and private information
retrieval(PIR)
Secure multiparty computation (SMC) and garbled
circuits(GCs)
Secure (approximate) interactive protocols
Practical limitations of privacy

tools
The restrictions and peculiarities of a cloud scenario
essentially
limit two parameters:
The bandwidth of the customercloud link, which cannot
be continuously active, and
The computational overhead for the customer.
MAPPING COMPLEX TO REAL SOLUTIONS

There are many challenges that SPED must face to
provide efficient
privacy-preserving solutions in a cloud scenario.
Specifically, several tradeoffs have to be optimized :
For a given privacy-preserving solution, a balance must be
found among the following four magnitudes:
Computational load,
Communication(bandwidth and interaction rounds),
Accuracy (error propagation),and
Privacy level (differential privacy).
Ideally, a privacy-preserving e-Health cloudified scenario

should conform
to the generic architecture depicted in Figure as follows
Fig: Generic architecture for a SPED-privacy protected health cloud.
CONCLUSION
Throughout this we have motivated the need for privacy
when outsourcing processes to cloud environments
The first and most fundamental issue deals with the
definition and quantification of privacy in the cloud.
The range of cloud applications is rich and varied, from very
simple spreadsheet applications to the rendering of
synthetic video scenes or finding the solution to complex
optimization problems.
References
[1] W. Zhu, C. Luo, J. Wang, and S. Li, Multimedia cloud
computing, IEEE Signal
Process. Mag., vol. 28, no. 3, pp. 5969, May 2011.
[2] M. Jensen, J. O. Schwenk, N. Gruschka, and L. L. Iacono,
On technical
security issues in cloud computing, in Proc. IEEE Int. Conf.
Cloud Computing,
Bangalore, India, Sept. 2009, pp. 109116.
[3] J. R. Troncoso-Pastoriza and F. Prez-Gonzlez, CryptoDSPs
for cloud
privacy, in Proc. CISE 2010, Hong Kong, China, LNCS 6724.
[4] M. Armbrust, A. Fox, R. Griffith, A. D. Joseph, R. H. Katz, A.
Konwinski,
G. Lee, D. A. Patterson, A. Rabkin, I. Stoica, and M. Zaharia,
Above the clouds: A
Berkeley view of cloud computing, EECS Dept., Univ.
California, Berkeley, Tech.
Rep. UCB/EECS-2009-28, Feb. 2009.
[5] M. Osadchy, B. Pinkas, A. Jarrous, and B. Moskovich, Scifi:
A system for

Secure Signal Processing in The Cloud: (Enabling Technologies For Privacy-Preserving Multimedia Cloud Processing)

Uploaded by

Copyright:

Available Formats

Secure Signal Processing in The Cloud: (Enabling Technologies For Privacy-Preserving Multimedia Cloud Processing)

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Secure Signal Processing in The Cloud: (Enabling Technologies For Privacy-Preserving Multimedia Cloud Processing)

Uploaded by

Copyright:

Available Formats

Secure Signal Processing in the

Under the Guidence of

CLOUD AND PRACTICAL SCENARIOS

Service Layers in cloud

architectural layer, representing the raw cloud

At any of these three layers of service provision, there are

Functional roles played by the

Exemplifying scenarios of signal

Outsourced biometric recognition: In a

e-Health: e-Health is the paradigmatic scenario

Outsourced adaptive or collaborative

FORMALIZING THE PRIVACY

PRIVACY TOOLS FROM SPED

Practical limitations of privacy

MAPPING COMPLEX TO REAL SOLUTIONS

Ideally, a privacy-preserving e-Health cloudified scenario

Fig: Generic architecture for a SPED-privacy protected health cloud.

You might also like