Title :
Optimum channel and source integrated (CSI) coding
Author_Institution :
Coll. of Staten Island, City Univ. of New York, NY, USA
Abstract :
A novel analog channel and source integrated (CSI) coding formulation is offered that inherently arises from a systematic generalization of a recently offered minimum mean squared error (MMSE) linear state space (LSS) predictive transform (PT) source coding formulation. The CSI coding formulation leads to both suboptimum and optimum coders. The suboptimum coder has a decoder that is relatively simple to design and implement but has a stability that is a function of the assumes signal source statistics. On the other hand, the optimum coder has a decoder that is more complex to design and implement but has the advantage of being stable regardless of the assumed signal source statistics and is, in addition, strictly optimum for the ease where the true describing equations for the channel and signal source are of the additive white noise and white LSS types, respectively
Keywords :
decoding; encoding; estimation theory; linear predictive coding; optimal systems; optimisation; random processes; state-space methods; analog channel; equations; linear state space predictive transform; minimum mean squared error; optimum coders; source coding formulation; source integrated coding; suboptimum coders; Bandwidth; Control systems; Decoding; Equations; Multidimensional signal processing; Signal design; Source coding; Stability; State-space methods; Statistics;
Conference_Titel :
Aerospace and Electronics Conference, 1994. NAECON 1994., Proceedings of the IEEE 1994 National
Conference_Location :
Dayton, OH
Print_ISBN :
0-7803-1893-5
DOI :
10.1109/NAECON.1994.332864
