1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Optimal Gateway Selection in Multi-domain Wireless Networks: A Potential Game Perspective Yang Song, Starsky H.Y. Wong, and Kang-Won Lee Wireless Networking Research Group IBM T. J. Watson Research Center Mobicom 2011 Research was sponsored by US Army Research and UK Ministry of Defense under W911NF-06-3-0001. IBM Research 1 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Overview Motivation 1 Gateway Selection Game 2 Equilibrium Selective Learning 3 Performance Evaluation 4 Conclusions 5 IBM Research 2 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Coalition Networks with Multiple Domains Scenario: - Coalition networks with heterogenous groups. - Inter-connected via wireless links, e.g., IEEE 802.11, WiMAX, UAV, satellite, 3G/4G etc. Example: Joint military missions, US-UK Disaster rescue teams, fire-fighters and police officers Wireless sensor networks of different organizations, e.g., Internet of Things (IoT), Smart Planet Solutions IBM Research 3 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Interoperability Issue IBM Research 4 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Interoperability Issue Problems: Inter-domain communication is non-trivial for heterogenous domains Different network protocol, security schemes, policies Security and policy enforcement, traffic analysis IBM Research 4 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Interoperability Issue Problems: Solution: Inter-domain communication is non-trivial for heterogenous domains Different network protocol, security schemes, policies Security and policy enforcement, traffic analysis IBM Research 4 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Interoperability Issue Problems: Solution: Designate gateway nodes Inter-domain communication is non-trivial for heterogenous Gateways domains S1 D1 Different network protocol, security schemes, policies D2 Security and policy S2 enforcement, traffic analysis Domain A Domain B IBM Research 4 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Cost Efficient Gateway Selection Each pair of nodes has a cost , Destination e.g., routing metric cost, such as Gateways Domain B hop count, RIP, AODV etc. Euclidean distance Source ETX, ETT, RTT Domain A Domain C Energy consumption etc. IBM Research 5 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Cost Efficient Gateway Selection Each pair of nodes has a cost , Destination e.g., routing metric cost, such as Gateways Domain B hop count, RIP, AODV etc. Euclidean distance Source ETX, ETT, RTT Domain A Domain C Energy consumption etc. For a single domain Intra-domain cost IBM Research 5 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Cost Efficient Gateway Selection Each pair of nodes has a cost , Destination e.g., routing metric cost, such as Gateways Domain B hop count, RIP, AODV etc. Euclidean distance Source ETX, ETT, RTT Domain A Domain C Energy consumption etc. For a single domain For the network Intra-domain cost Inter-domain backbone cost IBM Research 5 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Cost Efficient Gateway Selection Each pair of nodes has a cost , Destination e.g., routing metric cost, such as Gateways Domain B hop count, RIP, AODV etc. Euclidean distance Source ETX, ETT, RTT Domain A Domain C Energy consumption etc. For a single domain For the network + Intra-domain cost Inter-domain backbone cost Question: How to select the set of gateways s.t. the overall cost is minimized? IBM Research 5 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Gateways Domain B Source Domain A Domain C IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Source Domain A Domain C IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Source Domain A Domain C Distributed solution IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Each domain may designate gateway for its own benefit Source (self-interested / lack of coordination) Domain A Domain C Distributed solution IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Each domain may designate gateway for its own benefit Source (self-interested / lack of coordination) Domain A Domain C Distributed solution Equilibrium efficiency IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Each domain may designate gateway for its own benefit Source (self-interested / lack of coordination) Domain A Domain C Reluctance in revealing its own intra-domain topology Distributed solution Equilibrium efficiency IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Each domain may designate gateway for its own benefit Source (self-interested / lack of coordination) Domain A Domain C Reluctance in revealing its own intra-domain topology Distributed solution Equilibrium efficiency Local information only IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Challenges Destination Combinatorial nature of Gateways Domain B solution space Each domain may designate gateway for its own benefit Source (self-interested / lack of coordination) Domain A Domain C Reluctance in revealing its own intra-domain topology Distributed solution Equilibrium efficiency Local information only potential game theory & equilibrium selective learning IBM Research 6 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Network Model M : the set of domains in the s = { g 1 , g 2 , · · · , g |M| } : the coalition network joint gateway selection profile of the network N m : the set of nodes in the domain Satellite/UAV/3G/4G link: cost η (expensive), to enforce g i m = 1: node i is selected as always-on connectivity the gateway node and g i m = 0 o.w. and � i m = argmax i ∈N m g i A pair of node i and j : m be the selected gateway node c ( i , j ) ≥ 0 is the associated symmetric link cost, c ( i , j ) = η m , · · · , g |N m | g m = { g 1 m , g 2 } : the m if out of range gateway selection strategy of c ′ ( i , j ) � min ( c ( i , j ) , η ) domain m IBM Research 7 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Gateway Selection Game For each single domain Minimize ( Local information and observation only) � � c ′ � � � � i , � i m , � � U m ( g m , g − m ) = c i m + i n (1) i � = � n � = m , n ∈M i m , i ∈N m IBM Research 8 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Gateway Selection Game For each single domain Minimize ( Local information and observation only) � � c ′ � � � � i , � i m , � � U m ( g m , g − m ) = c i m + i n (1) i � = � n � = m , n ∈M i m , i ∈N m Destination Player: each domain m ∈ M Gateways Domain B Strategy space: N m Source Domain A Domain C IBM Research 8 / 19
1. Motivation 2. Gateway Selection Game 3. Equilibrium Selection Learning 4. Evaluation 5. Conclusions Gateway Selection Game For each single domain Minimize ( Local information and observation only) � � c ′ � � � � i , � i m , � � U m ( g m , g − m ) = c i m + i n (1) i � = � n � = m , n ∈M i m , i ∈N m Destination Player: each domain m ∈ M Gateways Domain B Strategy space: N m Questions Source Domain A Domain C IBM Research 8 / 19
Recommend
More recommend