The Optimal Configuration Method And The MC Simulation For The Multi-channel Sharing System

The multi-channel sharing technique has already been widely applied to communication field such as satellite communication, mobile communication and other type of communication systems. The multi-channel sharing means that any idle channels can be occupied to communicate by any users in the system, therefore the user numbers in the system could be more than the channel numbers. Adopting the multi-channel sharing technique makes the channel utilization ratio increase and the call congestion phenomenon emerge at the same time. In order to ensure both the call congestion ratio satisfying and the channel utilization ratio high enough, researches on the rational configuration method of the system is extremely essential.In this thesis, the multi-channel sharing system is considered as a stochastic service system. Based on the steady state solution of the birth and death process which state space is a limited set, utilized M/M/n/n/m queue model, the call congestion ratio formula and the channel utilization ratio formula fitting to the limited user system are derived. Drawing support from the visual data analyzing function of the software MATLAB, the value divergence between the formulas and the corresponding one fitting to the infinite user system are compared. The source and the regularity of deviation when Erlang B formula or the corresponding channel utilization ratio formula is used in a limited user system is dissected, and the condition available of the two groups of formulas are pointed out. According to the formulas that are applicable to the different type of systems, the optimal configuration methods and their MATLAB source programs are proposed.Computer simulation is a new subject combined with system engineering, modem mathematical method and computer technology. Computer simulation method has already become an effective tool and indispensable way in scientific research except theory analysis and scientific experiment. Monte-Carlo method is called MCM for short. It is base on the probability theory and the computational mathematics, so MCM is also named statistical simulation approximate computation method. Monte-Carlo simulation is just a computer mathematical simulation method utilized MCM. And its application field is getting wider and wider.In this thesis, the MCS of the multi-channel sharing system for limited users is achieved. The simulation sequence has not only confirmed that the optimal configuration methods obtained from theory model is reliable, but also supplied another convenient and effective method to make the system analysis. The more important significance of MCS is to reflect the running state of the stochastic service system completely and visually. Therefore, in order to make the system perfect we'll analyze and improve further.As example, the optimal configuration methods proposed in this thesis are applied to the large-scale telephone line expander system. The application has verified that the multi-channel sharing system for limited users has reached the design requirements at aspect of the call congestion ratio and the channel utilization ratio.In addition, the MCS method and its MATLAB source programs proposed in this thesis are universal to any sense of limited source Markov loss queue system. Improving slightly, the new simulation program available of the limited source Markov loss-delay queue system could be obtained.