Although anticipatory systems are frequently found in many natural phenomena, formal logic and reasoning (by implication) have not been successful in analyzing these systems. Robert Rosen, in his seminal work on anticipatory and complex systems, categorized these systems as those that cannot be modeled by formal reasoning based on implication and further traced their connections to Newtonian formalism where causality is replaced by implication. Rosen demonstrated his modeling relation in the context of mechanics (R. Rosen, 1985; 1991). The author examines information flows in these systems, and explores the system in the communication and computation context. It is concluded that existing communication and computation theories in common with physics/mechanics are missing the "final cause" category defined by Aristotle!.

This paper attempts to explore a conceptual model of any generic software for capturing its macroscopic behavior via identification of attributes analogous to volume, pressure, entropy, etc., as known in physical systems. Logical volume, pressure and temperature are defined while retaining the concepts of entropy. Based on this model the totality of error control strategies via the constrained number of variables, their ranges, data typing, program structuring, etc., is termed software hardening which is parallel to liquefaction and solidification phenomena in physics. This model provides a framework for conceptual unification of various prevailing techniques of software reliability improvement, such as modularity, object orientation, software reuse, etc., which have similar effects of reducing the number of states traversed and thus are all ways towards software hardening.

In this paper an FM detector capable of suppressing the degradation in receiver performance due to the presence of an interfering signal is presented. Optimum receiver structures based on maximum-aposteriori estimation procedures are first derived and then a practical demodulator based on the optimum receivers is examined. The receiver consists of two phase-locked loops (PLL) interconnected in a manner to permit one PLL to lock onto and track the Stronger received FM signal while the other loop tracks the weaker of the two received FM signals. The detector has the capability of demodulating both the desired received FM as well as the interferer even for the case when both signals are co-channel. Experimental results demonstrating such capability even in the presence of strong input Gaussian noise are presented.