Make your own free website on Tripod.com

INFORMATION & KNOWLEDGE IN DYNAMIC SYSTEMS

An original look to the system's TRINITY

Copyright by Josip Pajk, December 1997.

Dynamic systems structure

Dynamic systems are the most widespreaded kind of systems we know, and the sole kind of systems cybernetics should deal with. Unlike common material and energetic(al) structures which are entirely dependent about their environment, or static systems which behaviour is limited and invariant (predictable) in time, dynamic systems can be recognized for their behaviour which is not static (always the same output for the same input), but capable to evolve in accordance with their specific inherent informational structure. Various real dynamic systems (organisms, organizations, artificial systems) can be all pictured by a model with an identical (informational) structure. This model is built from three distinctive blocks:

[dynsys.gif]

  • The learning block where system specific structures (information) are extracted from the (material and energetical) data set acquired in the environment.
  • The memory block where these new structures are stored and compared with the old (structured) ones producing the knowledge (active category) that defines the system state (behaviour).
  • The output block which produces the response of the system, i.e. material and energetic(al) structures to "deliver" in the environment (to other systems).

    This model can be applied to all three structural types the system is built of. Namely, all dynamic systems are capable to acquire from the environment, extract (select) and incorporate (memorize) parts of various material and energetic(al) structures for the formation of their unique compound system structure, in which the informational structure is dependent and, at the same time, also defines the transitions in the other two structural layers of the dynamic system. These transitions from one structural state to another on all three system layers are made possible by four multidimensional vectors that connects the three system's blocks. The vectors on the informational layer could be identified as:

  • U(t) Input vector of Data from the environment
  • V(t) Inner vector of Information extracted by the system
  • X(t) Inner (Knowledge) state vector of the system
  • Y(t) Output data vector of the (Response) to the environment

    The internal state of knowledge influences the way in which information are extracted from the input set of data as well as the ultimate system response. The autopoiesis (self-consciousness) process inside the system (the V-X circuit) is cyclic and permanent, regardless if there is or not an input vector U applied to the system.

    Consequently, there is no other possibility of changing (control) the systemís inner states from the outside (environment, other systems) besides the application of an input data vector U(t), nor there is another mean for the identification of the system's internal change of state, besides tracking and comparing it's response (output) Y to the applied (input) data U. Evidently, the so called "control system", can never be completely assured if it's control signals will produce the desired response (behaviour) of the "controlled" dynamic system.

    Therefore, a conclusion can be made, that: there is no information outside of the system. There are only material or/and energy structured data flows (signals). Concisely speaking, it's not correct to address these processes as "transfer of information" between systems, because just in the system, information can be extracted from the data set (signals) acquired from the environment. Information must somehow change the inner state (knowledge) structure of the system. The same set of data will obviously result in a different set of information in two different systems, or even in the same system in a different time interval (if it's inner state, or knowledge, has been changed meanwhile).

    Particular material and energetical structures produced by the environment or other systems (even the environment can be analysed as a dynamic system), means something (bring information) only for those systems that have the appropriate material, energetical and informational structure on which these particular structures have some kind of influence. Structures made of matter and energy may interact, make transitions from one state to another, but there is no information without (or, out of) a system with the capability to exploit it. Exploitation means that the system has a goal. The basic goal of any system is it's survival. Just after the fulfilment of this goal the system can "think" about the fulfilment of it's other goals, whatever they are.

    The observer

    As it can be seen from the previous contentions, there is a strong correlation between the observer and the extracted information (Observer dependent knowledge).

    There are only two distinctive methods of observer operation:

  • Passive (monitoring) which does not affect the observed system (structure) in any way, gathers information from the observed energetic structural changes, mainly from those radiated that are influencing the systemís environment (i.e. instruments, antennas, electromagnetic detectors).
  • Active (search) methods are used for the identification of rigid (material) structures. It is necessary for the observer system to produce and project some kind of energetic or material structure toward the observed system (structure) in order to gain a response (reflection) from which information can be extracted (i.e. radar, various scout or measuring missions).

    A book or a CD (or even a plant seed) is nothing else but a material structure. Energetical structures like broadcast signals means nothing until acquired by a system (observer), able to extract from them the needed information by the application of an appropriate method.

    Evolution

    To be continued...