Throughput and Delay Scaling of General Cognitive Networks 11/47 LSystem Models LBasic Models Assumptions Network extension:a unit square. n primary users {X)1 and m secondary users {)1 Independently and identically distributed (i.i.d.)in the network extension. Random permutation traffic,no cross network traffic. 4口“0·4E，4色t1=月QG
Throughput and Delay Scaling of General Cognitive Networks 11 / 47 System Models Basic Models Assumptions I Network extension: a unit square. I n primary users {Xi} n i=1 and m secondary users {Yj} m j=1 . I Independently and identically distributed (i.i.d.) in the network extension. I Random permutation traffic, no cross network traffic
Throughput and Delay Scaling of General Cognitive Networks 12147 LSystem Models LBasic Models Interference Model The Physical Model:a transmission between two primary users is successful if the SINR at receiver satisfies: SINR= Received Power No+IPP+Is→P 2a (1) where, Ipp:Interference from other primary TXs Isp:Interference from all secondary TXs If successful,transmits with bandwidth W. Define the physical model for secondary users similarly. Similar results also hold for the Gaussian Channel Model. 4口40·4E4色下空引=
Throughput and Delay Scaling of General Cognitive Networks 12 / 47 System Models Basic Models Interference Model The Physical Model: a transmission between two primary users is successful if the SINR at receiver satisfies: SINR = Received Power N0 + IP→P + IS→P ≥ α (1) where, IP→P : Interference from other primary TXs IS→P : Interference from all secondary TXs I If successful, transmits with bandwidth W. I Define the physical model for secondary users similarly. I Similar results also hold for the Gaussian Channel Model
Throughput and Delay Scaling of General Cognitive Networks 13/47 LSystem Models LBasic Models Asymptotic Capacity Definition Asymptotic per-node capacity X(n)of the network is said to be e(g(n))if there exist two positive constants c and c'such that: limn→ooPr{λ(n)=cg(n)is feasible}=1 limno Prf(n)=c'g(n)is feasible<1 4口“③·4E4色卡21=
Throughput and Delay Scaling of General Cognitive Networks 13 / 47 System Models Basic Models Asymptotic Capacity Definition Asymptotic per-node capacity λ(n) of the network is said to be Θ(g(n)) if there exist two positive constants c and c 0 such that: limn→∞ Pr {λ(n) = cg(n) is feasible} = 1 limn→∞ Pr {λ(n) = c 0g(n) is feasible} < 1
Throughput and Delay Scaling of General Cognitive Networks 14147 LSystem Models LOperation Rules Operation Rules Essential difference between cognitive networks and normal(standalone)networks Should be as general as possible. To formulate the concept of priority: Primary users disregard secondary users. Secondary users must take care of both themselves and primary users. 4口“0·4E，4色t1=月QG
Throughput and Delay Scaling of General Cognitive Networks 14 / 47 System Models Operation Rules Operation Rules I Essential difference between cognitive networks and normal (standalone) networks. I Should be as general as possible. I To formulate the concept of priority: I Primary users disregard secondary users. I Secondary users must take care of both themselves and primary users
Throughput and Delay Scaling of General Cognitive Networks 15/47 LSystem Models LOperation Rules Operation Rules Rule 1 (the scheduling decision model of primary users) A transmission is eligible to be scheduled if: Received Power、 a+e No+Ip-P (2) Note that /sp is not included,i.e.,the primary scheduler may not be aware of secondary users. Though it could be arbitrarily small,the allowance e is necessary to provide opportunities for secondary users. 4口40·4E，4色下空1=
Throughput and Delay Scaling of General Cognitive Networks 15 / 47 System Models Operation Rules Operation Rules Rule 1 (the scheduling decision model of primary users) A transmission is eligible to be scheduled if: Received Power N0 + IP→P ≥ α + (2) I Note that IS→P is not included, i.e., the primary scheduler may not be aware of secondary users. I Though it could be arbitrarily small, the allowance is necessary to provide opportunities for secondary users