Semidefinite Programming Approach to Gaussian Sequential Rate-Distortion Trade-offs
Author(s)
Tanaka, Takashi; Baek, Kwang Ki; Parrilo, Pablo A.; Mitter, Sanjoy K
DownloadAccepted version (1.350Mb)
Terms of use
Metadata
Show full item recordAbstract
Sequential rate-distortion (SRD) theory provides a framework for studying the fundamental trade-off between data-rate and data-quality in real-time communication systems. In this paper, we consider the SRD problem for multi-dimensional time-varying Gauss-Markov processes under mean-square distortion criteria. We first revisit the sensor-estimator separation principle, which asserts that considered SRD problem is equivalent to a joint sensor and estimator design problem in which data-rate of the sensor output is minimized while the estimator's performance satisfies the distortion criteria. We then show that the optimal joint design can be performed by semidefinite programming. A semidefinite representation of the corresponding SRD function is obtained. Implications of the obtained result in the context of zero-delay source coding theory and applications to networked control theory are also discussed.
Date issued
2017-04Department
Massachusetts Institute of Technology. Laboratory for Information and Decision Systems; Massachusetts Institute of Technology. Department of Materials Science and EngineeringJournal
IEEE Transactions on Automatic Control
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Citation
Tanaka, Takashi, Kwang-Ki K. Kim, Pablo A. Parrilo and Sanjoy K. Mitter. "Semidefinite Programming Approach to Gaussian Sequential Rate-Distortion Trade-offs." IEEE Transactions on Automatic Control 62, Issue: 4 (April 2017).
Version: Author's final manuscript
ISSN
0018-9286
1558-2523