Method and device for establishing data service model

Abstract

The embodiment of the invention provides a method and a device for establishing a data service model. The method comprises: when the number of users is not greater than a first threshold value, adopting a first data service model to describe the increase of cell throughput along with the increase of the number of the users, and to describe a process in which an individual user throughput decreases; when the number of the users increases and exceeds the first threshold value, adopting a second data service model to describe decrease of the cell throughput, and to describe the process in which the individual user throughput decreases; and acquiring the best relation between the number of the users and the throughput according to the first data service model and the second data service model. The data service model is established in a statistical manner, the Markov process is not used as the basis, and the model is proposed based on actual test results; meanwhile, parameters alpha and beta of the model can be corrected according to a scene, so that the model has a broader application range; and based on the data service model, the relation between the number of the users and the throughput is described and the maximum cell throughput is predicted.

Description

technical field [0001] The invention relates to mobile communication technology, in particular to a method and device for establishing a data service model. Background technique [0002] Existing business models are only applicable to circuit domain models, such as GSM communication systems and TD-SCDMA voice systems. Because each user monopolizes a resource, such as frequency point, time slot or code channel, it is reasonable to adopt the Irish model. The existing business model uses a discrete-state Markov process, and the business model is as follows: figure 1 As shown, the discrete-state Markov process assumes that: the arrival of the telephone call flow obeys the Poisson process, and the duration of each call obeys the negative exponential distribution of the parameter μ. The system has s trunks. If there is no idle trunk, the incoming call is rejected and the call no longer enters the system. This is a typical birth-death process, and its arrival rate and departure ...

AI Technical Summary

Problems solved by technology

[0006] The existing technology has the following problems: the existing business model is not suitable for the resource sharing communication system with a scheduling mechanism, and the scheduling mechanism of modern mobile communication usually adopts proportional fairness, which needs to consider the size of the user's data to be transmitted and historical throughput information And the current signal-to-noise ratio (SNR, SignaltoNoiseRatio), this model itself is no longer a Markov process

Images