( ) Nebulas Rank (In-and-Out Degree) Wilbur

Similar documents
위해 사용된 기법에 대해 소개하고자 한다. 시각화와 자료구조를 동시에 활용하는 프로그램이 가지는 한계와 이를 극복하기 위한 시도들을 살펴봄으로서 소셜네트워크의 분석을 위한 접근 방안을 고찰해 보고자 한다. 2장에서는 실험에 사용된 인터넷 커뮤니티인 MLBPark 게시판

왜 2.0 인가? 비트코인이 아직 발전의 초기단계라면 왜 벌써 이를 뛰어넘는 2.0 플랫폼이 필요한 것일까? 우선 비트코인 기술자체가 완성된 것이 아니고 지속적인 개선과 발전이 필요. 하지만 비트코인은 이미 50억달러가 넘는 경제적 이해관계가 걸려 있는 네트웤. 1차적

새로운 생태계

DBPIA-NURIMEDIA

Yggdrash White Paper Kr_ver 0.18

무선데이터_요금제의_가격차별화에_관한_연구v4.hwp

월간 CONTENTS 3 EXPERT COLUMN 영화 점퍼 와 트로이목마 4 SPECIAL REPORT 패치 관리의 한계와 AhnLab Patch Management 핵심은 패치 관리, 왜? 8 HOT ISSUE 2016년에 챙겨봐야 할 개인정보보호

°í¼®ÁÖ Ãâ·Â

<333820B1E8C8AFBFEB2D5A B8A620C0CCBFEBC7D120BDC7BFDC20C0A7C4A1C3DFC1A42E687770>

Data Industry White Paper

I

13-08.hwp

학습영역의 Taxonomy에 기초한 CD-ROM Title의 효과분석

03-최신데이터

THE JOURNAL OF KOREAN INSTITUTE OF ELECTROMAGNETIC ENGINEERING AND SCIENCE. vol. 29, no. 10, Oct ,,. 0.5 %.., cm mm FR4 (ε r =4.4)


R을 이용한 텍스트 감정분석

164

High Resolution Disparity Map Generation Using TOF Depth Camera In this paper, we propose a high-resolution disparity map generation method using a lo

09권오설_ok.hwp

DBPIA-NURIMEDIA

DBPIA-NURIMEDIA

리텀 백서 새로저장-작은용량

03¼ºÅ°æ_2

À±½Â¿í Ãâ·Â

강의지침서 작성 양식

#Ȳ¿ë¼®

<4D F736F F D20B1E2C8B9BDC3B8AEC1EE2DC0E5C7F5>

À¯Çõ Ãâ·Â

6

삼국통일시나리오.indd

정보기술응용학회 발표

< FBEC8B3BBB9AE2E6169>

아바타 캐릭터 패션의 컬 러마케팅 전략 형성에 관한 연구 (pp ) - 김영식 임미라 Contents 논문요약 Abstract 1. 서론 n 본론 1. 웹의 발달과아바타의 개념 및활용현황 2. 실제와사이버상의 아바타 패션 트랜드 경향 3. 색채의 연상, 상징

THE JOURNAL OF KOREAN INSTITUTE OF ELECTROMAGNETIC ENGINEERING AND SCIENCE Nov.; 26(11),

2002 Game White paper 2002 Game White paper

산업백서2010표지

김기남_ATDC2016_160620_[키노트].key

±èÇö¿í Ãâ·Â

08SW

PowerPoint 프레젠테이션

001지식백서_4도

example code are examined in this stage The low pressure pressurizer reactor trip module of the Plant Protection System was programmed as subject for

¼º¿øÁø Ãâ·Â-1

<4D F736F F F696E74202D E20C0CEC5CDB3DD20C0C0BFEB20B9D720BCADBAF1BDBA20B1E2BCFA E >


데이터베이스-4부0816

2002 Game White paper 2002 Game White paper

Microsoft PowerPoint - AC3.pptx

제19권 제3호 Ⅰ. 문제제기 온라인을 활용한 뉴스 서비스 이용은 이제 더 이 상 새로운 일이 아니다. 뉴스 서비스는 이미 기존의 언론사들이 개설한 웹사이트를 통해 이루어지고 있으 며 기존의 종이신문과 방송을 제작하는 언론사들 외 에 온라인을 기반으로 하는 신생 언론사

歯목차45호.PDF

歯목차.PDF

Microsoft PowerPoint - G3-2-박재우.pptx

sna-node-ties

½Éº´È¿ Ãâ·Â

THE JOURNAL OF KOREAN INSTITUTE OF ELECTROMAGNETIC ENGINEERING AND SCIENCE. vol. 29, no. 6, Jun Rate). STAP(Space-Time Adaptive Processing)., -

<313120B9DABFB5B1B82E687770>

PowerPoint 프레젠테이션

ÀÌÀç¿ë Ãâ·Â

<30312DC1A4BAB8C5EBBDC5C7E0C1A4B9D7C1A4C3A52DC1A4BFB5C3B62E687770>

? Search Search Search Search Long-Tail Long-Tail Long-Tail Long-Tail Media Media Media Media Web2.0 Web2.0 Web2.0 Web2.0 Communication Advertisement

, N-. N- DLNA(Digital Living Network Alliance).,. DLNA DLNA. DLNA,, UPnP, IPv4, HTTP DLNA. DLNA, DLNA [1]. DLNA DLNA DLNA., [2]. DLNA UPnP. DLNA DLNA.

<B9AEC8ADB0E6C1A6BFACB1B820C1A63137B1C720C1A633C8A C2F720BCF6C1A4BABB292E687770>

hapter_ i i 8 // // 8 8 J i 9K i? 9 i > A i A i 8 8 KW i i i W hapter_ a x y x y x y a /()/()=[W] b a b // // // x x L A r L A A L L A G // // // // /

09오충원(613~623)

ePapyrus PDF Document

WeToken白皮书

Microsoft PowerPoint - XP Style

Ä¡¿ì³»ÁöÃÖÁ¾

BRAVO WHITEPAPER V7 (1)_kor

<B1DDC0B6BCD2BAF1C0DA20B0A1C0CCB5E5BACF28C1A4B8AEBFEB292DC6EDC1FD28342E E687770>

<C7A5C1F620BEE7BDC4>

(JBE Vol. 21, No. 1, January 2016) (Regular Paper) 21 1, (JBE Vol. 21, No. 1, January 2016) ISSN 228

< FB1B9C1A6B1DDC0B6BFACB1B820352D3128BCF6C1A4B9DDBFB5292E687770>

[ReadyToCameral]RUF¹öÆÛ(CSTA02-29).hwp

<32382DC3BBB0A2C0E5BED6C0DA2E687770>

특집....,.,., (good jobs) (rent-sharing) (fairness)..... Ⅱ. 임금과생산성구조의분석모형 ) 1),,,, 2_ 노동리뷰

38이성식,안상락.hwp

07.045~051(D04_신상욱).fm

<4D F736F F F696E74202D20454C53BDC3BCBCC1B6C1BEC0EFBCDBC0C720C0EFC1A1B0FA20B9FDB8AE >

korean_1.6

Copyright 0, Oracle and/or its affiliates. All rights reserved.,.,,,,,,,,,,,,.,...,. U.S. GOVERNMENT RIGHTS Programs, software, databases, and related

슬라이드 1

[Nebulas] Non-tech Whitepaper_KR_Final(1).pages

슬라이드 제목 없음

179

February

歯이혜경.PDF

歯MW-1000AP_Manual_Kor_HJS.PDF

6.24-9년 6월

레이아웃 1

0125_ 워크샵 발표자료_완성.key

<4D F736F F D20C3D6BDC C0CCBDB4202D20BAB9BBE7BABB>

< FC1A4BAB8B9FDC7D D325FC3D6C1BEBABB2E687770>

45-51 ¹Ú¼ø¸¸

02김헌수(51-72.hwp

THE JOURNAL OF KOREAN INSTITUTE OF ELECTROMAGNETIC ENGINEERING AND SCIENCE Dec.; 26(12),

<BCBCB9CCB3AA E32392E687770>

07_À±¿ø±æ3ÀüºÎ¼öÁ¤


Transcription:

: Nebulas Rank Nebulas 2018 6 v1.0.1

1... 1 2... 3 2.1... 3 2.2... 4 2.3 ( )... 4 3... 6 3.1... 6 3.2... 7 4 Nebulas Rank... 10 4.1... 10 4.2 (In-and-Out Degree)... 12 4.3 Wilbur... 14 5 Core Nebulas Rank ( )... 17 5.1... 17 5.2 ( )... 18 5.3... 18 6 Core Nebulas Rank... 18 6.1?... 18 6.2 Core Nebulas Rank... 19 7 Extended Nebulas Rank... 20 7.1 Extended Nebulas Rank... 20 7.2 Extended Nebulas Rank... 20 8... 21... 22 A.... 24 B.... 27

1., (DApp). (Openness).,,.,. TTP Cookie.,.,.,.,,..,.,.,,.,.. :...,... 1

Nebulas Rank., Nebulas Rank.. (,, ).,.,. Nebulas Rank Nebulas Rank, Coinmarketcap.com. Nebulas Rank ( 0.84), Nebulas Rank. Nebulas, Nebulas Rank ( ). Nebulas Rank Nebulas Rank Core Nebulas Rank Extended Nebulas Rank. Core Nebulas Rank.. Extented Nebulas Rank Core Nebulas Rank,., Core Nebulas Rank Extented Nebulas Rank. Nebulas Rank, Nebulas Rank,, Nebulas Rank Nebulas Rank. :, Nebulas Rank Nebulas ( ) (2018 4 v1.0.2)[3]. ( ). 2

2.. 2.1 2008 10 Satoshi Nakamoto,.., 3.,., (Namecoin)[5] (DNS), Open Assets (Colored coins),. (Turing-Complete),.. (Ethereum) [7]. (Smart Contracts),.. (Ethereum Virtual Machine, EVM) EVM., (DApp). DApp,,,., (Killer Dapps)., (Externally owned account, EOA).,... 2015 Chris Skinner (Value Web), (Value Ecosystem) (Value Exchanges) (Value Stores) 3

(Value Management Systems),. Chris. 2.2,.,.,,. PageRank [9]. Google PageRank., PageRank,, PageRank. PageRank Fleder, Kester, Pilai [10], PageRank. PageRank PageRank. Web 2.0 PageRank., PageRank, PageRank LeaderRank. PageRank, LeaderRank. LeaderRank ( ). PageRank (Sybil Attack). LeaderRank (, ),,.,. Nebulas Rank NEM, [17] (Proof of Work) (Proof of Stake) NEM (Proof of Importance), NCDawareRank[14]. NCDawareRank ( ) SCAN[15] [16] [17].,. 2.3 ( ) Nebulas Rank,. Hopcroft [18], PageRank 4

. Zhang, ( ) [19]. PageRank (Symmetric Network) [20][12]. : 1... 2. (out-degree),. 3.,. 4.,. Nebulas Core Nebulas Rank,. 5

3. Core Nebulas Rank. ( ) Core Nebulas Rank. 3.1..!. L U (L, U)!!. : L = (A, D, T)! (1) A D T!,!,! ( ). (tetrad) : D = {a d, a A, d R*}! (2) T = {(s, t, w, τ)}! (3) a! d a d d s, t, w τ! (!., 0 )!.!!,,. : u dom a. u U, a A! (4) 6

: A(u ) = { a A : u dom a}! (5) :! (6)...... 3.2,. [21],. u 1, u 2 U : A(u 1 ) A(u 2 ) = ϕ Nebulas Rank.,. (Velocity Factor)... : Nebulas ( )[3].., :. R.S Kroszner 6 2,795, [22].,. [23], V!. : M V = P Y! (7) 7

! M V P,!,!!,, (,!.) (GDP). (7). M Y 1 p! Nebulas (Nebulas NAS 4 % ), 2100.!., (7) : V (M + Δm) (s,t,w,τ) T w! (8) M = P Y Δm!.!, (New Keynesian Models)..,.. (, ),!!.. Y!, GDP,.,.,. (8)!. Y! : P Y P Y Y = C(a)! (9) a A C(a) a!!, Core Nebulas Rank. Nebulas.,. ( : Nebulas Proof of Devotion ) 8

( ).. 9

4 Nebulas Rank Core Nebulas Rank... : Core Nebulas Rank. [t 0 T, t 0 ]! : Θ(t 0 ) = {(s, t, w, τ) t 0 T τ t 0 w > 0 s t}! (10) Θ(t o ) d c τ!,,!!,!,!. a A C(a) Θ(t o )!!! Core Nebulas Rank : C(a) = Ω(β(a)) Ψ(γ(a)) β(a) a γ(a)! (11)!!!.. s : Nebulas ( )[3] Core Nebulas Rank : 1.!. 2. LeaderRank.,.., PageRank ( : LeaderRank) [20].. 4.3 21. K 4.1 β(a) [t 0 T, t o ] n, : B i B i+ 1 a A B o, B 1,, B n. : 10

d a o, d a 1,, d a n : d a (0), d a (1),, d a (n) d(i) a < d (i+ a 1),0 i n 1 β(a)!,! : (a) = ( d a (k) for n =2 k, k =1, 2, 3,... (d a (k) + da (k+1))/2 for n =2 k +1,k =1, 2, 3,... (12).,. w = 10, =2 v 1 v 2 w = 100, =1 w = 100, =5 w = 10, =3 H(v) ={(v, v1, 100, 1), (v1, v2, 10, 2), (v2, v3, 10, 3), (v3,v,10, 4)} v v 3 w = 10, =4.1: v 1 v 2 w = 100, =5 w = 90, =1 11 v v 3

!!! 4.2 (In-and-Out Degree).2:.1. (in-and-out)... : v H(v) H(v) {(v, v 1, w 1, τ 1 ), (v 1, v 2, w 2, τ 2 ),, (v i, v i+ 1, w i, τ i ),, (v n, v, w n+ 1, τ n+ 1 )},!! 1 i n: τ 1 τ i+ 1 (v 1, v 2, 100, 5) n.1,,..! : H 1 (v 1 ), H 2 (v 2 ),, H n (v n ) H i (v i ) (s i m, t i m, w i m, τ i m)!!, H i (v i ) w i m H i (v i )!!,! 0 : (s i, t i, w i, τ i ) T : w i w i m E((s, t, w, ),w m )= (s, t, w wm, ) if w 6= w m if w = w m Θ (t 0 ) = Θ(t 0 ) H i (v) {E(t), t H i (v i )} i = 1,2,, n! (13).2.1 ( ). 12

v!! p(v) :! p(v) = w (14) i (s i,v,w i,τ i ) Θ (t 0 ) v! : v,!! γ(v)! q(v) = w (15) i (v,t i,w i,τ i ) Θ (t 0 ) : q(v) arctan! p(v) ) (16) G(v) = (p(v) + q(v)) e 2 sin2 ( π 4.3 14. G(v) =(p(v)+q(v)) e 2sin2 ( 4 arctan q(v) p(v) ) 10 5 0 4 q(v) 2 0 0 2 p(v) 4.3: 13

γ(v) = ( θ G(v) G(v) + μ )λ! (17)! θ, μ, λ 4.3 Wilbur. Core Nebulas Rank. Nebulas Rank. Core Nebulas Rank f (x), Wilbur 1. Core Nebulas Rank,,. x 1 x 2 f (x) : 1. 0. x f (x 1 + x 2 ) > f (x 1 ) + f (x 2 ) x 1 > 0,x 2 > 0 (18) f(x) =x f(x) =x/(1 + e a+b x ) Nebulas Rank Nebulas Rank Factor.4: Nebulas x 1 x 2 2.. lim f (x 1 + x 2 ) = f (x 1 ) + f (x 2 ) x 1 > 0,x 2 > 0 x 1,x 2 (19) 1 (Sybil Attack) 1970 TV (Sybi)l. Dr. Cornelia Wilbur. Wilbur. 14

.,...4. f (x) = x /(1 + e a+ b x ) a > 1,b < 0 (20) A., (11) : C(v) = β(v) 1 + e γ(v) a+ b β(v) 1 + e c+ d γ(v) (21).5: Nebulas Rank Market Capitalization of Ethereum (USD) 2 1.5 10 9 Nebulas Rank ETH Market Cap. 0 10 20 30 40 50 60 Time (day) Core a, b, c, d. Core Nebulas Rank. Nebulas 2017 5 1 2017 6 30 ( : 3629091 3955158 ) ETH ( ) [24]..5 ETH Core Nebulas Rank ETH ( ) (21) Core Nebulas Rank. 15

Core Nebulas Rank. p 4.48 10 17 < 0.001 0.84427, (p ). (11) Core Nebulas Rank. 16

5 Core Nebulas Rank ( ) Core Nebulas Rank Nebulas Rank..........,,. 5.1 (21).., 0 β o. γ.,. β,.,,.,, 5.3...... 4.3 (21)..,. (Cycle-removal Algorithm). γ = 2β o C =.. (1 + e a+ b β 0)(1 + e c+ 2d β 0).. 2β 2 0. C = γ 2T β o T 2T β0 2. (1 + e a+ b β o)(1 + e ec + c c d β o ), 17

5.2 ( ) Sybil Attack P2P [25]. P2P. P2P. ID., ID. P2P (Redundancy),, ID. P2P ID ID ID. ID. ID. P2P. [26].... Core Nebulas Rank.. β o,. N N β = N γ K β K = 2 N 5.1,,. : C = N K β2 0 N (1 + e a+ b β 0 N )(1 + e c+ K d β 0) = Kβ 2 0 (1 + e a+ b β 0 N )(1 + e c+ K d β 0) (22) 5.3.. 6 Core Nebulas Rank Core Nebulas Rank Core Nebulas Rank. 6.1? Core Nebulas Rank.. Nebulas Rank. 18

Nebulas Rank :, IPFS, Genaro[27][28], Core Nebulas Rank. Core Nebulas Rank, Core Nebulas Rank., TPS. Core Nebulas Rank. Core Nebulas Rank Core Nebulas Rank., Core Nebulas Rank Core Nebulas Rank., Core Nebulas Rank.,, Core Nebulas Rank, Core Nebulas Rank. Core Nebulas Rank, Core Nebulas Rank. Core Nebulas Rank... 6.2 Core Nebulas Rank Core Nebulas Rank, Core Nebulas Rank,. Core Nebulas Rank., Nebulas Force Core Nebulas Rank., Core Nebulas Rank (LLVM IR ). Nebulas (NVM), Core Nebulas Rank Core Nebulas Rank.,. 19

7 Extended Nebulas Rank Core Nebulas Rank. Core Nebulas PoD( ) DIP( ), Core Nebulas Rank.,, Extended Nebulas Rank. Extended Nebulas Rank Core Nebulas Rank Nebulas. 7.1 Extended Nebulas Rank Nebulas. DApp. DApp... (Acyclic Graph), Page Rank Nebulas Rank. 7.2 Extended Nebulas Rank.,. Extendede Nebulas Rank,, Core Nebulas Rank. Extended Nebulas Rank, Core Nebulas Rank.., Extented Nebulas Rank Core Nebulas Rank. Extended Nebulas Rank Extended Nebulas Rank,. Extended Nebulas Rank. 20

8 Nebulas Rank.. Nebulas : Nebulas Rank..,., DApp. Dapps.. Nebulas... Nebulas. 21

[1] S. Meiklejohn, M. Pomarole, G. Jordan, K. Levchenko, D. McCoy, G. M. Voelker, and S. Savage, A fistful of bitcoins: characterizing payments among men with no names, in Proceedings of the 2013 conference on Internet measurement con- ference, pp. 127 140, ACM, 2013. [2] Http cookie. https://en.wikipedia.org/wiki/http_cookie. [3] Nabulas Technical White Paper. https://nebulas.io/docs/ NebulasTechnicalWhitepaper.pdf. Accessed: 2018-04-01. [4] S. Nakamoto, Bitcoin: A peer-to-peer electronic cash system, 2008. [5] Namecoin. https://namecoin.org. [6] Openassets protocol. http://github.com/openassets/ open-assets-protocol. [7] V. Buterin et al., Ethereum white paper, 2013. [8] Forget fintech welcome to the valueweb. http://thefinanser.com/2015/ 02/forget-fintech-welcome-to-the-valueweb.html/. [9] L. Page, S. Brin, R. Motwani, and T. Winograd, The pagerank citation ranking: Bringing order to the web., tech. rep., Stanford InfoLab, 1999. [10] M. Fleder, M. S. Kester, and S. Pillai, Bitcoin transaction graph analysis, arxiv preprint arxiv:1502.01657, 2015. [11] Q. Li, T. Zhou, L. Lu, and D. Chen, Identifying influential spreaders by weighted LeaderRank, Physica A: Statistical Mechanics and its Applications, vol. 404, pp. 47 55, 2014. [12] A. Cheng and E. Friedman, Manipulability of pagerank under sybil strategies, 2006. [13] NEM Technical Reference. http://nem.io/nem_techref.pdf. Accessed: 2017-08-01. [14] A. N. Nikolakopoulos and J. D. Garofalakis, NCDawareRank, Proceedings of the sixth ACM international conference on Web search and data mining - WSDM 13, no. February 2013, p. 143, 2013. [15] X. Xu, N. Yuruk, Z. Feng, and T. A. Schweiger, Scan: a structural clustering algorithm for networks, in Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 824 833, ACM, 2007. [16] H. Shiokawa, Y. Fujiwara, and M. Onizuka, Scan++: efficient algorithm for finding clusters, hubs and outliers on large-scale graphs, Proceedings of the VLDB Endowment, 22

vol. 8, no. 11, pp. 1178 1189, 2015. [17] L. Chang, W. Li, L. Qin, W. Zhang, and S. Yang, pscan: Fast and exact structural graph clustering, IEEE Transactions on Knowledge and Data Engineering, vol. 29, no. 2, pp. 387 401, 2017. [18] J. Hopcroft and D. Sheldon, Manipulation-resistant reputations using hitting time, in International Workshop on Algorithms and Models for the Web-Graph, pp. 68 81, Springer, 2007. [19] J. Zhang, R. Zhang, J. Sun, Y. Zhang, and C. Zhang, Truetop: A sybil-resilient system for user influence measurement on twitter, IEEE/ACM Transactions on Networking, vol. 24, no. 5, pp. 2834 2846, 2016. [20] A. Cheng and E. Friedman, Sybilproof reputation mechanisms, in Proceed- ings of the 2005 ACM SIGCOMM workshop on Economics of peer-to-peer systems, pp. 128 132, ACM, 2005. [21] M. Swan, Blockchain: Blueprint for a new economy. O Reilly Media, Inc., 2015. [22] R. S. Kroszner, Liquidity and monetary policy, 2007. [23] R. Selden, Monetary velocity in the united states, 1956. [24] CoinMarketCap. https://coinmarketcap.com/. [25] D. Quercia and S. Hailes, Sybil attacks against mobile users: friends and foes to the rescue, in INFOCOM, 2010 Proceedings IEEE, pp. 1 5, IEEE, 2010. [26] Wikipedia contributors, Sybil attack Wikipedia, the free encyclopedia, 2018. [Online; accessed 25-June-2018]. [27] Ipfs. https://ipfs.io/. [28] Genaro. https://genaro.network/en/. 23

!!!! A. A.1 Proof of Property 1 proof x 1 > 0,x 2 > 0!. For any!, we have f (x 1 + x 2 ) = = = x 1 + x 2 1 + e a+ b (x 1+ x 2 ) x 1 1 + e + x 2 a+ b (x 1+ x 2 ) 1 + e a+ b (x 1+ x 2 ) x 1 1 + e b x 2 e a+ b x 1 + x 2 1 + e b x 1 e a+ b x 2 b < 0 0 < e (a+ b) x 1 < 1,0 < e b x 2< 1, In formula 21, we have!, so! moreover, x 1 1 + e b x 2 e a+ b x 1 > x 1 1 + e a+ b x 1 = f (x 1 ) x 2 1 + e b x 1 e a+ b x 2 > x 2 1 + e a+ b x 2 = f (x 2 ) is actually: f (x 1 + x 2 ) > f (x 1 ) + f (x 2 ) A.2 Proof of Property 2 proof x 1 > 0,x 2 > 0!. For any!, we have f (x 1 + x 2 ) f (x 1 ) f (x 2 ) = x 1 + x 2 1 + e x 1 x 2 a+ b (x 1+ x 2 ) 1 + e a+ b x 1 1 + e a+ b x 2 x 1 x 1! (23) = ( ) 1 + e b x 2 e a+ b x 1 1 + e a+ b x 1 x 2 x 2 + ( ) 1 + e b x 1 e a+ b x 2 1 + e a+ b x 2 g (x 1, x 2 ) h (x 1, x 2 ) Here we use function! represents the left part,! represents the right part: x 1 x 1! (24) g (x 1, x 2 ) = 1 + e b x 2 e a+ b x 1 24 1 + e a+ b x 1

!!!! x 2 x 2! (25) h (x 1, x 2 ) = 1 + e b x 1 e a+ b x 2 1 + e a+ b x 2 x 1 x 2 So (23) for! and!, their limits can be represented as: lim x 1,x 2 we have [ f (x 1 + x 2 ) f (x 1 ) f (x 2 )] = lim g (x 1, x 2 ) + x 1,x 2 lim h (x 1, x 2 ) x 1,x 2 g (x 1, x 2 ) = = < < x 1 x 1 1 + e b x 2 e a+ b x 1 1 + e a+ b x 1 x 1 e a+ b x 1 (1 e b x 2) (1 + e b x 2 e a+ b x 1) (1 + e a+ b x 1) x 1 e a+ b x 1 (1 + e a+ b x 1) (1 + e b x 2 e a+ b x 1) (1 + e a+ b x 1) = x 1 e a+ b x1 1 + e b x 2 e a+ b x 1 x 1 e a+ b x 1 1 + e a+ b x 1 = 1 + x 1 1 e a+ b x 1 x Calculate limit for!, according to L Hospital s rule, 1 1 + ea+ b x lim x 1 + x 1 = lim 1 x (e a b x ) e a+ b x = lim x 1 b e a b x b < 0 lim b e a b x = In formula 21, we have!, so!, moreover, x lim x 1 + x = 0 1 e a+ b x g (x 1, x 2 ) > 0, According to A.1, we have! so according to sandwich theorem: 25

!!! lim x 1,x 2 g (x 1, x 2 ) = 0 Similarly, we can get: lim x 1,x 2 h (x 1, x 2 ) = 0 So, lim x 1,x 2 [ f (x 1 + x 2 ) f (x 1 ) f (x 2 )] = 0 26

B. v1.0 v1.0.1 4.3 1 2 v1.0.2 27