This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1 ECE & CSE ECE-CSE 861: Introduction to Computer Communication Networks Ness B. Shroff ECE & CSE Lectures 12-13 ECE & CSE L= w Littles Law L = expected number of packets in the system = arrival rate w = expected waiting time Littles Law ECE & CSE Littles Law (Proof) 4 3 1 2 5 6 t1 t2 t3 w 1 t4 w 2 w 3 w 4 T t5 t6 w 5 w 6 Cj Packets that arrive into system Time w j is defined as the waiting time in system for customer (packet) j c j is defined as the jth customer arriving into system t j is defined as the arrival time of customer c j Note: Queueing system is not FIFO (why?) ECE & CSE Littles Law More Definitions: I j (t) = 1, c j is in system at time t (not departed) = 0, otherwise N t = number of customers in the system at time t 1 Hence, ( ) (1) Now ( ) (waiting time for customer ) and ( ) = # of arrivals (of customers) by time = max ( : ) t j j j j j N I t w I t dt j t t j t t = = = 2 ECE & CSE Littles Law In our system for j 4, ( ) ( ) (2) T j j j w I t dt I t dt = = > T T dt t I dt t I w 4 4 4 4 ) ( w fact, In ) ( Because all customers c j have departed by time T. Note that the second part of equation is not true if c j had not departed before time T because w 4 T ECE & CSE Littles Law Now integrate N t (the number of customers in the system) ( ) 1 1 ( ) 1 0 ( ) 1 ( ) ( ) from(1) ( ) (interchanging thesumand theintegral) , for j 4 T T T t t j j j j T T j j T j j N dt I t dt I t dt I t dt w = = = = = = = = ( ) 1 Error term(negative) t t u j j N du w = = + Now for an arbitrary t 0, N t 0 ECE & CSE...
View Full Document
- Winter '11