Calculation of average time of packet delay in the storage buffer of a single-channel system with self-similar traffic

2017;
: pp. 87-91
1
О.S. Popov Odessa national academy of telecommunications
2
О.S. Popov Odessa national academy of telecommunications

First step for the calculation of service quality characteristics in a single-channel packet communication system is to estimate the Hurst exponent for self-similar traffic, after which, according to the well-known Norros formula, the average number of packets in the system N is calculated. However, such an algorithm does not allow calculating two very important service quality characteristics, such as the average waiting time of packets in the cumulative buffer (not in the system as a whole) and the waiting probability of the service start of the packet. In the paper the new method for approximating the probability distribution function of the system states is proposed, where a simple exponential function with the  distribution parameter was used for the approximating function. From this approximating function the new formula for calculating the waiting probability for the service start of the packet in a one-channel system with self-similar traffic is obtained. This method of calculation is based on the phenomenon that packets in self-similar traffic are not smoothly distributed over different time intervals. They are grouped into "blocks" within certain time intervals, but there are hardly any of them within the others. Therefore, for such traffic, in the distribution function of the number of packets in a single-channel system, the probability  of a complete absence of packets in it increases. The results obtained in the paper will be useful for the further development of monitoring subsystems of power comlexes.

  1. V. Krylov and S. Samokhvalova, The theory of teletraffic and its applications, St. Petersburg, Russia: BHV-Petersburg, 2005. (Russian)
  2. N. Ilkka, “A storage model with self-similar input”, Queuing Systems, vol. 16, 1994.
  3. L. Kleinrock, The theory of mass service. Trans.from English, Moscow, Russia: Mashinostroyenie, 1979. (Russian)
  4. A. Lozhkovskyi, “Traffic model in multi-service packet-switched networks”, Scientific Papers of O.S. Popov Odessa National Academy of Telecommunications, vol. 1, pp. 63-67, 2010. (Russian)
  5. B. Mandelbrot, The Fractal Geometry of Nature, Computing in mathematics, physics, biolog; Trans. from English, Moscow, Russia: Institute of Computer Science, 2002. (Russian)
  6. A. Lozhkovskyi, “Simplified calculation of the Hurst coefficient by the R/S-analysis method”, in Proc. 2nd IEEE  International Conference on Advanced Information and Communication Technologies-2017 (AICT-2017), pp. 254-257, Lviv, July 4-7, 2017.
  7. I. Lesovoy and G.Pylypenko, “Imitation modeling of the routing process based on fuzzy logic”, IAPGOS, vol. 1, pp. 16-18, 2018.
  8. A .Lozhkovskyi and Ye. Levenberg,  “Investigation of simulating methods for self-similar traffic flows“, in Proc. 4th IEEE International Scientific-Practical Conference «Problems of Infocommunications. Science and Technology», PIC S&T’2017, pp. 410-413, Kharkiv, Ukraine, October 10–13, 2017.