In this project you will write a program to produce a discrete time simulation of a queue as shown in Fig. 1. Time is slotted on the input and the output. Each input packet follows a Bernoulli process. In a given time slot the independent probability that a packet arrives in a time slot is p, while the probability that the packet will be serviced is q. One packet fills one time slot.        

The queue can store up to four packets (not the five shown in the diagram above). All packets are processed on a first come - first served basis. Assume that when a packet is serviced all other packets in a queue (if any) are shifted instantaneously towards the output. Each slot departures from the queue are processed before arrivals. 

In your discrete event simulation the program will mimic the operation of the queue and collect statistics. More specifically, you will need to collect (a) throughput and (b) delay statistics for different values of p (p = 0.02, 0.04 up to 1.0 in steps of 0.02), and for a fixed value of q = 0.75. To obtain an accurate statistics you should simulate at least ten thousand time slots for each value of p. Note that you ARE NOT allowed to implement the model equation in the program - but you can use them as a check.

The average throughput is just the number of serviced packets divided by the number of time slots. The average delay of the queue is an average number of time slots a packet is waiting in a queue before it gets serviced (i.e., it is the total number of time slots which all serviced packets spend in the queue divided by the total number of serviced packets). For the delay statistics, it is convenient to store your packets in a linked list and associate the time slot tag with each packet.