Han-Chieh Chao. President, National Ilan University, ILan, Taiwan

Transcription

1 Han-Chieh Chao President, National Ilan University, ILan, Taiwan

2 Dr. Han-Chieh Chao President National Ilan University, I-Lan, Taiwan July 18, 2011 at Seville, Spain 2

3 Han-Chieh Chao TEL: ext. 200 PhD from Purdue University Current Position:President of National Ilan University (NIU), Taiwan Jointly Appointed Professor of Institute of CSIE and Department of Electronic Engineering, NIU. Jointly Appointed Professor of National Dong Hwa University, Hualen, Taiwan Honorary Professorship of Beijing Jiaotong University Yantai University Xiamen University and Lanzhou University FIET (Fellow of IET) FBCS CITP (Fellow and Chartered IT Professional of British Computer Society) Editor-in-Chief: IET Communications, IJAHUC, JIT, IJIPT

4 (

5

6 National Center for Traditional Arts (Ilan)

7 Taiping Mountain (Ilan) Chilain Mountain (Ilan)

8 Turtle Island (Ilan)

9 National Ilan University Main Entrance Gate

10 Campus View (I)

11 Campus View (II)

12 Campus View (III)

13 Year School Name Taiwan Provincial Ilan School of Agriculture and Forestry (Founded by Japanese on 1926 年 5 月 12 日創立 ) Taiwan Provincial Ilan Agriculture Vocational School Taiwan Provincial Ilan Agriculture and Engineering Vocational School Taiwan Provincial Ilan Agriculture and Engineering Vocational High School National Ilan Junior College of Agriculture and Technology 1998 National Ilan Institute of Technology 2003 National Ilan University 13

14 14

15 accessed from 15

16 Main Campus Wujei Campus Sansin Campus Practicing Farm Practicing Forest Farm etc 15.4 Hectare 15.0 Hectare Hectare 1.0 Hectare Hectare 0.4 Hectare 16

17 NIU Administrative Academic Academic Affairs Student Affairs General Affairs R&D Affairs Extension Education Library Computer Center Accounting Human Resource Military Training Physical Training Engineering EECS BioResources Humanity & Management General Education 17

18 Organization of Colleges College of Engineering Civil Engineering Chemical and Material s Engineering Mechanical and Electro- Mechanical Engineering Environmental Engineering Institute of Architecture and Sustainable Planning College Bioresource Food Science Animal Science Horticulture Natural Resource Biomechatronics Engineering Biotechnology Agriculture Promotion Committee College Humanities & Management Institute of Management Economics Foreign Languages & Literature Humanities and Science Center Language Center EECS Electronics Electrical Engineering Institute of CSIE Degree Program of EECS 18

19 ID Category 人數 Note Student General 4486 Extensive 848 College Graduate School PhD Program M W M W M W M College W Teacher Full-time 241 Including 4 temporarily transfer Adjunct 92 Staff 84 Including 2 security guard Jinator 40 19

20 Current Status about Cloud Computing Why Cloud Computing What is Cloud Computing Cloud Computing Security Cloud Security Alliance (CSA) National Institute of Standards and Technology (NIST) Security Solutions Conclusion 2011/7/16 20

21 Introduction of IEEE Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 21

22 Introduction of IEEE Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 22

23 Defined as Worldwide Interoperability for Microwave Access by WiMAX Forum Refers to interoperable implementations of the IEEE wireless networks standards Provides fixed and mobile Internet access A solution of the last mile problem An important role on the network technology of Broadband Wireless Access (BWA) in the future 23

24 Released on Dec Published on June 2004 Point-to-point (PTP) Line-of-sight (LOS) d Published on Feb Known as Integrate with , c and a Point-to-multi-point (PMP) Non-line-of-sight (NLOS) e Known as e Mobile WirelessMAN Mobile WiMAX Approved on Dec m Approved on May 2009 Advanced Air Interface with data rates of 100Mbit/s mobile and 1Gbit/s fixed Mobile WiMAX Release 2 or WirelessMAN- Advanced j Multihop Relay Specification 24

25 25

26 26

27 It is declared that WiMAX system can supply higher transmit data rate and wider signal coverage than 3G system Range: 30 mile radius from base station Speed: 70 Mega bits per second But the real network planning and testing results are imperfect 27

28 28

29 BS s No. Distance from BS to user TCP Bandwidth (Mbps) UDP Bandwidth (Mbps) UL DL UL DL m m km m m km Annotation File Size (TCP/UDP) 64K/8K (byte) 64K/8K (byte) 64K/8K (byte) 64K/8K (byte) 64K/8K (byte) 64K/8K (byte) 29

30 IEEE standard organization start to establish Mobile Multi-hop Relay (MMR) Study Group (SG) Already passing IEEE 802 Executive Committee (EC) approbation and set up IEEE j Relay Task Group (TG) The IEEE j is based on the IEEE e standards 30

31 The most different in j is the frame structure in PHY layer and the definition in MAC layer Features Type e j Topology PMP only Tree structure (PMP compatible, not ad hoc nor mesh) Number of hops Single hop Multi-hop Traffic aggregation No Yes over multi-hop path System capacity Lower Higher within BS coverage area Coverage Lower Higher Construction cost Higher Lower PHY support OFDMA OFDMA extension 31

32 Utilize the technology and relevant mechanism of Relay Station in order to: enhance WiMAX user s transmission capacity extend a wider communication coverage area To compare with BS, low cost and easy to use is the main advantage in IEEE j 32

33 Does not forward the frame header information Apply to centralized scheduling Lower complexity but two hop relaying only Enhance the system capacity in BS s coverage 33

34 According to the scheduling mode to generate or forward the frame header information But forwarding framing information might cause interference between neighbor RSs which limit the system capacity Different scheduling mode accompanies different system complexity Extend the BS s coverage through multi-hop relaying 34

35 The most different between T-RS and NT-RS is the transmission method of framing information Features Type T-RS NT-RS Coverage extension No Yes Number of hops 2 2 or more Interference in RS cell None High Signal strength (in BS coverage) High Same as e Signal strength (out of BS coverage) None Medium RS cost Low High 35

36 Two types of network planning Uniform Data Rate Coverage Set up RS in the BS coverage Always set up near the BS boundary coverage or critical shadow fading area Coverage Extension Set up the RS out of the BS coverage To expand the BS coverage 36

37 IEEE j can Solve the shadowing effect Extend the communication range 37

38 To satisfy the limitation: User s utility rate Maximum hop count Communication interference Ultimate goal: Fit the deployment condition Satisfy user s requirement Minimize the construction cost 38

39 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 39

40 [4] Lin P., Ngo H., Qiao C. M., Wang X., Wang T. and Qian D, Minimum Cost Wireless Broadband Overlay Network Planning, Proceedings of International Symposium on a World of Wireless, Mobile and Multimedia Networks The coverage problem is formulated as a mixed integer programming 2. Authors propose some heuristic algorithms for overlay networks 40

41 [8] Lin B., Ho P.-H., Xie L.-L. and Shen X, Relay Station Placement in IEEE j Dual-Relay MMR Networks, Proceedings of IEEE ICC Author investigate how to decide the optimal placement of RSs 2. The proposed algorithm is based on dual-relay architecture 41

42 [10] Lu H.-C. and Liao W.-J., Joint Base Station and Relay Station Placement for IEEE j Networks, Proceedings of IEEE GLOBECOM Authors only consider at most two-hop relaying in their proposed algorithm 2. The planning time is not considered in this research 42

43 [11] Yu Y., Murphy S. and Murphy L, Planning Base Station and Relay Station Locations in IEEE j Multi-hop Relay Networks, Proceedings of 2nd IEEE Broadband Wireless Access Workshop, colocated with IEEE CCNC The network planning time is raised exponential in big scale network 2. Only suit small size network planning problem 43

44 [12] Yu Y., Murphy S. and Murphy L, A Clustering Approach to Planning Base Station and Relay Station Locations in IEEE j Multi-hop Relay Networks, Proceedings of IEEE ICC Authors consider the transmitting power of subscriber station 2. The proposed method is based on clustering technique 44

45 [13] Chen C. Y., Liang Y. S., Yu C. M., Ho C. H. and Kuo S. Y, Increasing Reliability for IEEE j Mobile Multi-hop Relay Networks Planning, Proceedings of IEEE 15th Pacific Rim International Symposium on Dependable Computing (PRDC), Shanghai, China, November 16-18, The most related work with our research 2. The tree algorithm is proposed 3. BSs are placed on the vertices of selected candidate nodes which will increase the construction cost 45

46 Authors Features BS Numbers in Network Problem Definition Solution RS Relaying Hop Count Planning Time Construction Cost Lin P. et al. [4] Multiple Mixed Integer Programming Heuristic Two-hop - High Lin B. et al. [8] Single ILP Dual-relay Architecture Two-hop - - Lu H.-C. et al. [10] Multiple ILP Graph Theory Two-hop - Medium Yu Y. et al. [11] Multiple ILP Standard Approach (Brach & Bound) Two-hop High (Minutes) High Yu Y. et al. [12] Multiple ILP Clustering Approach Two-hop Medium (Minutes) High Chen C. Y. et al. [13] Multiple ILP Graph Theory Multi-hop Proposed Algorithm Multiple ILP Graph Theory Multi-hop Low (Seconds) Low (Seconds) Medium Low - : means the authors don t consider this factor 46

47 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 47

48 Assumptions: The locations of BSs or RSs and target nodes are known The link between BSs and RSs belongs to valid or invalid are known The tree structure is used to connect BSs and RSs in IEEE j network. Available link between candidate node j and candidate node i Requirement in our model: The maximum depth of routing: ηi = max j y i, j If η i >1, the maximum tree is limited to l 48

49 Definition of Variables ILP for 16j coverage Variable 1 { z1 k V =,..., z } V 1 = k 2 { q1 s V = V 2 = s E1 E 2 x i, j a i, j wi ui l E 1 ηi,..., q } Definition Set of candidate nodes in the network Number of candidate nodes in the network Set of target nodes in the network Number of target nodes in the network Set of links within candidate nodes Set of links between a candidate node and a target node The available links within a candidate node and a candidate node The available links within a candidate node and a target node The signal quality of a specified target point The maximal signal quality for the available target node The depth of the routing tree within The maximum depth of the routing tree 49

50 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 50

51 Three assumptions: 1. All the locations of candidate nodes and target nodes are known, and the link between whole nodes belong to valid or invalid are known too. 2. All the communication range of BSs are consistent in our planning case, and this assumption is equivalent to all RSs. 3. The construction cost of RS is 1 unit, and the construction cost of BS is 3 times of RS 51

52 Tree algorithm (proposed in [13]) 52

53 Algorithm :Supergraph Tree Algorithm ( G, z, δ, ) Input : G : underlying graph Ω = V, Ω z r VGr G z : vector δ : minimum required average utility : maximum depth of the routing tree 1 repeat S = f r' = S [1] until Ω z G z z S[1] r = γ ( G, Ω = C S = f S[1] S[1] = C = C = C until U(z) δ G l = if V l r' d S = f d S = z until V S[1] l d Ω = Ω 1 1 while G S = f r d while G = φ f = C = indicating the placement of = V, = φ ( G, Ω, Ω ) \ S[1] and Ω G = γ ( G, Ω ), where G repeat repeat 2 is unconnected ( G, Ω, V RS 1 2 RS ( G, Ω, V ( G, V Gr BFS, l, BS r Gr \ S [1]) is the supergraph of is unconnected BS BS Gr ', Ω ) r = Ω r = ( V ( G ), where G l fd ( G, \ V, V ) BS ) ) 2 r \ N( G, S [1]) Gr l, E Gr G, where G 2 2 = ( V r' l ) = ( V l, E ) BS and RS Gr ', E Gr ' ) 53

54 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 54

55 Candidate node set = {z 1 z 15 } = Ω 1 Target node set = {q 1 q 30 } = Ω 2 Network topology Selected subgraphs 55

56 Planning results Tree algorithm S-Tree algorithm Algorithm BS RS Total cost Tree S-Tree

57 Simulation parameters Variables/parameters Network Size Definition 10x10 (km) Candidate Nodes 15 Target Nodes 30 BS Radius RS Radius BS Cost RS Cost 4200 (m) 1000 (m) 3 (unit) 1 (unit) Maximum hop count 4 Minimum utility rate 30 (%) 57

58 Calculation time with varying number of candidate nodes 1,2 1 Calculation Time (sec) 0,8 0,6 0,4 Tree S-Tree 0, Nuumber of Candidate Nodes 58

59 Construction cost with varying BS radius and RS fixed to 1 km Construction Cost (units) Tree S-Tree BS Communication Radius (km) 59

60 Construct cost with ten different network topologies Construction Cost (units) Tree S-Tree Mean of Tree Mean of S-Tree Network Topology Scenario 60

61 Construct cost with varying the utility rate limitation Construction Cost (units) Tree S-Tree Minimum Utility Rate (%) 61

62 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 62

63 We have four major contributions: 1. We formulate the j coverage problem which called 16jcoverage, and propose S-Tree algorithm for solving 16jcoverage based on graph theory. 2. The multiple BSs and RSs network can be constructed by our proposed algorithm, instead of single BS and multiple RSs in others paper. 3. The network planning time presents linearly raised with increased candidate and target nodes, rather than exponential calculation time in other research. 4. We provide the realistic planning case and simulation results. 63

64 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 64

65 We propose the S-Tree algorithm which can achieve the lowest planning time and construction cost. Due to the proposed algorithm may construct more than one BS which introduces interference, we will take the interference between BSs into consideration as our future work. 65

66 Introduction Related works Problem definition Proposed mechanism Results Contributions Conclusions and future works References 66

67 1. Lin B., Ho P.-H., Xie L.-L., Shen X., and Tapolcai J, Optimal Relay Station Placement in Broadband Wireless Access Networks, IEEE Transactions on Mobile Computing, 2010, vol. 9, pp Peters S.W. and Heath R.W, The future of WiMAX: Multihop relaying with IEEE j, IEEE Communications Magazine, 2009, vol. 47, pp Genc V., Murphy S., Yu Y. and Murphy J, IEEE J Relay-based Wireless Access Networks: An Overview, IEEE Wireless Communications, 2008, vol. 15, pp Lin P., Ngo H., Qiao C. M., Wang X., Wang T. and Qian D, Minimum Cost Wireless Broadband Overlay Network Planning, Proceedings of International Symposium on a World of Wireless, Mobile and Multimedia Networks, Velez F.J., Carvalho V., Santos D., Marcos R.P., Costa R., Sebastiao P. and Rodrigues A, Aspects of cellular planning for emergency and safety services in mobile WiMax networks, Proceedings of 1st International Symposium on Wireless Pervasive Computing, Jan Theodoros T. and Kostantinos V, WiMax Network Planning and System's Performance Evaluation, Proceedings of IEEE WCNC 2007, pp Lin B., Ho P.-H., Xie L.-L. and Shen X, Optimal relay station placement in IEEE j networks, Proceedings of IEEE IWCMC 2007, pp

68 8. Lin B., Ho P.-H., Xie L.-L. and Shen X, Relay Station Placement in IEEE j Dual-Relay MMR Networks, Proceedings of IEEE ICC 2008, pp Niyato D., Hossain E., Kim D. I. and Han Z, Joint Optimization of Placement and Bandwidth Reservation for Relays in IEEE j Mobile Multihop Networks, Proceedings of IEEE ICC 2009, pp Lu H.-C. and Liao W, Joint Base Station and Relay Station Placement for IEEE j Networks, Proceedings of IEEE GLOBECOM 2009, pp Yu Y., Murphy S. and Murphy L, Planning Base Station and Relay Station Locations in IEEE j Multi-hop Relay Networks, Proceedings of 2nd IEEE Broadband Wireless Access Workshop, colocated with IEEE CCNC Yu Y., Murphy S. and Murphy L, A Clustering Approach to Planning Base Station and Relay Station Locations in IEEE j Multi-hop Relay Networks, Proceedings of IEEE ICC 2008, pp Chen C. Y., Liang Y. S., Yu C. M., Ho C. H. and Kuo S. Y, Increasing Reliability for IEEE j Mobile Multi-hop Relay Networks Planning, Proceedings of IEEE 15th Pacific Rim International Symposium on Dependable Computing (PRDC), Shanghai, China, November 16-18, Becvar, Z., 2007, in IFIP International Federation for Information Processing, Volume 245, Personal Wireless Communications, eds. Simak, B., Bestak, R., Kozowska, E., (Boston: Springer), pp

69 69