On the Effect of a Random Access Protocol on the Performance of the Session-Based Data Subsystem in GSM/GPRS

Abstract

Research into mixed services in mobile cellular networks is highly topical. Session-based applications (e.g., Web browsing) form significant and increasing data traffic in cellular networks. In such mixed services, random access (RA) protocols impose significant effects in performance metrics of the new data users. To date, analytical facility has been limited by the absence of closed-form models for the RA protocols that exist in time-division multiple access (TDMA) cellular networks with mixed voice (as circuit switched) and data with finite number of sessions (as packet switched). In this novel paper, we research an RA protocol based on S-Aloha with data as a session for mobile TDMA cellular networks and present an analytical framework. This work is germane to Global System for Mobile Communications/General Packet Radio Service systems. This paper begins with a general model description of a voice and data subsystem, and an explanation of its admission schemes. As well as the model, an analytical framework based on equilibrium point analysis is derived, and an algorithm is then formulated to complement the analytical framework. Using this novel analytical framework, together with the associated algorithm, performance metrics of the system such as throughput, mean access delay, and mean delay in the queue are obtained. Stability conditions are also highlighted. Numerical results are corroborated using simulation.