Distributed net-centric architecture of m2m acquisition units with optical GVA

Abstract

This paper describes the architecture of a low-latency symmetric multiprocessing optical soft memory system to cluster computing inside the core of an adaptive optical signal processor with the aid of soft decision algebraic polynomial algorithms. The optical system hardware is shown to evolve along with the iterator instantiations of the soft algorithm that forms the core of the memory map. The system enables efficient cache coherence protocols used in unit multiprocessors to be run across a homogeneous cluster in optical soft memory systems. We define a structure called the Optical Generalized Viterbi Algorithm Data Structure (Optical GVA DS) that makes up the system map for adaptive optical signal processing. The system executes transforms where algorithms for handling the entire data vector is processed, shortening the computational complexity effectively. Thus the optical soft memory system as described by the evolving Optical GVA DS iterator instantiates enables the design of parallel processors to handle modulated data in the optical domain. This is of importance in the realization of distributed netcentric architectures and forms the basis of large-scale real-time data processing and acquisition in m2m units.