|
An Adaptive Technique for Data Compression and Robustness at the Tactical Edge
June 2010
Yan Grushevsky, DSCI-XPRT Solutions, Inc.
Junghoon Lee, DSCI-XPRT Solutions, Inc.
George F. Elmasry, DSCI-XPRT Solutions, Inc.
Manoj Jain, DSCI-XPRT Solutions, Inc.
Michael Vitt, DSCI-XPRT Solutions, Inc.
Gregory Hadynski, AFRL/RIGC
Bruce Metcalf, The MITRE Corporation
ABSTRACT
An adaptive technique for reliable and fast transmission of streaming data at the tactical edge is presented. This technique combines data compression and data robustness to optimize for BW, reliability and speed of transmission in an adaptive matter. Speed of transmission and reducing BW needed for transmission is achieved by exploring lossless data compression of the payloads of the packet stream. Once all redundancy in the packet stream is removed, a packet-erasure-based FEC is applied that further optimize for BW and speed of transmission in addition to adding robustness to the packet stream that assures successful transmission within a speed of service given by tactical requirements. Lossless compression uses an adaptive Arithmetic Coding technique that provides a level of data compression close to theoretical limits (approaching the Shannon Bound) and does not require any prior knowledge of the statistical nature of the packet payload stream (performs compression on-the-fly). The FEC uses an RS code that can work over a general purpose process (does not require specific HW) and optimizes for the amount of redundancy needed. Robustness is achieved in the presence of high percentage of packet erasures caused by blockage, fading or malicious attack. The presented approach guarantees the delivery of a given message within a speed of service defined by tactical requirements and minimizes the BW needed for transmission. The paper presents the technique, shows supporting analysis and presents simulation results.
Additional Search Keywords
n/a
|
