Project for Performance Evaluation of Computer Systems and Networks - University of Pisa.
The project requirements are:
In a slotted random-access network, N couples transmitter-receiver share the same communication medium, which consists of C separate channels. Multiple attempts to use the same channel in the same slot by different transmissions will lead to collision, hence no receiver listening on that channel will be able to decode the message. Assume that each of the N transmitters generate packets according to an exponential inter-arrival distribution, and picks its channel at random on every new transmission. Before sending a packet, it keeps extracting a value from a Bernoullian RV with success probability p on every slot, until it achieves success. Then it transmits the packet and starts over. If a collision occurs, then the transmitter backs off for a random number of slots (see later), and then starts over the whole Bernoullian experiment. The number of back-off slots is extracted as U(1; 2^(x+1)), where x is the number of collisions experienced by the packet being transmitted.
The aim of the project is the Assessment of the Effectiveness of the Slotted Random- Access Network Protocol described in the latter paragraph.