Development and evaluation of an autonomous wireless sensor actuator network in logistic systems

To introduce the concept of the autonomous system, this chapter starts by reviewing the central system concept and structure. Then the central system constrains are explained and it is shown how the distributed system is developed to cope with the limitations of the central system. The distributed control system emergence following distributed system concepts is studied. Another term which is used for the distributed control system is decentralized. In one section the usage of this term is explained. The autonomous system concept based on these bases with consideration of its meaning in other realms is discussed. The autonomous structure has already attracted the attention of philosophers in political science in the field of the federal system. In order to obtain a clearer understanding of this term, it is tried to show the difference between autonomous and other common terms. In this survey the self-decision making is offered as a core element of the autonomous concept. Following this concept and considering the point that the autonomous system inherits the distributed system characters, the autonomous wireless sensor actuator network is introduced. In the last section it is explained how the autonomous structure in logistic systems can conduct the competition of the entities from the resources to the goals. In this way the necessity of developing the autonomous objects such as intelligent containers manifested itself. Development and evaluation of an autonomous wireless sensor actuator network in logistic systems 6 1.1) Central system According to Encyclopedia Britannica, a system a regularly interacting or interdependent group of items forming a unified whole erally three characteristics are counted for a system: structure, behavior and interconnection. Structure is about the composition of items which are called entities hereafter. Behavior refers to the character which the system portrays by itself when it is looked at from outside of the system. Interconnectivity describes the connectivity between the entities inside the system and it is known as a system network. A central system contains an entity which decides about the behavior of the system or plays a switching role in connection between the entities. In terms of networking there is an entity in the central system whose role is to make connection between entities, e.g. the center receives a message and sends it to the destination. In the central decision-making system, the supreme entity (center) makes decisions for the other entities and for the whole system and has authority over other entities. In political science such a system is also called an authority system [01King]. Decision-making is taken as a factor because it covers many aspects such as control, organization of the entities, task distribution and resource allocations. Referring to the system characteristic, the system network is independent from decision-making structures because in some systems the entities can make decisions for themselves but the data exchange between them is done via a center. Therefore the central system is referred to it is more precise to say from which aspect. A primary model of the central system is shown in Fig. 1.1. Elazar in [15Elaz] names this model Center-Periphery model and Tanenbaum in [12Tanb] names this model Star. The Center-Periphery model is preferred to Star because it represents the center peripherals roles which connect to the subordinates. In this figure the center entity is called supreme entity or easily center and the connected entities are subordinated to this center. The arrows show the connection path of the center and entities. The figure shows that the entities are not dependent on or connected to each other directly. In other words, the unification of the system is totally achieved by the center. If the center fails, the system will break apart. Now suppose that more entities are supposed to attach to the center, since the peripheral number of the center is limited, at one point the center will go out of its peripherals. After this point in order to add more entities to the center one solution is to add more resources and peripherals to the center. This solution is costly because not only does the old center becomes useless but also based on what Tanenbaum in