PoS(ISCC2015)043. Research of Wireless Network Coverage Self- Optimization Based on Node Self-Adaption Model. Wenzhi Chen. Bojian Xu 1.

Size: px

Start display at page:

Download "PoS(ISCC2015)043. Research of Wireless Network Coverage Self- Optimization Based on Node Self-Adaption Model. Wenzhi Chen. Bojian Xu 1."

Ernest Parrish
5 years ago
Views:

1 Research of Wireless Network Coverage Self- Optimizatio Based o Node Self-Adaptio Model 1 Iteratioal School Beijig Uiversity of Posts ad Telecommuicatios Beijig, ,Chia m @163.com Wezhi Che 2 Iteratioal School Beijig Uiversity of Posts ad Telecommuicatios Beijig,100000, Chia m @163.com This paper poits that owadays, the widely dissemiated wireless etwork with base statios ad fouds the Node Self-Optimizatio Model which treats statios as research objects i order to eable the BSs (base statios) to regulate the sigal trasmittig power by themselves; at meawhile, uder the aalysis of actual wireless sigals trasmittig coditios, a uique Path-Loss Model with probability factor which has ever bee cosidered before has bee established, i this sese, the proper power value for BS which is able to meet certai coverage ca be educed more accurately through calculatio so that the wireless etwork s coverageshortage problem ca be solved thoroughly. I additio, for the optimizatio of BSs power by usig efficiecy of the whole etwork, we itroduce the coverage overlappig ratio as a buildig block of desigig macro power regulatig arithmetic ad the fid the Redudacy- Optimizatio Algorithm. I compariso with the covetioal power cofiguratio of BSs, the utilizatio of this algorithm is able of decliig the coverage redudacy ad thus icreasig the overall power by usig the efficiecy effectively. ISCC , December, 2015 Guagzhou, Chia 1 2 Wezhi Che Copyright owed by the author(s) uder the terms of the Creative Commos Attributio-NoCommercial-NoDerivatives 4.0 Iteratioal Licese (CC BY-NC-ND 4.0).

2 1. Itroductio Nowadays, the wireless etwork has bee itegrated with the huma s life, from TD-LTE wide area commuicatio with high data rate to the WLAN with the core of routers[1]. The wireless etwork is almost everywhere; however, the extesive existece of wireless etwork fails to always provide users with high-speed ad stable commuicatio ad furthermore there is early o relevat research that ca solve the problem thoroughly[2]. I our daily life[3], we may ofte face the situatio that our phoes ca t detect the sigal of a certai WiFi or the sigal is too weak to support etwork services; therefore, the best sigal coverig rage which is able of supportig every mobile termial (MT) to commuicate smoothly ad provides by BS ought to be explored deeply. Cosequetly, we iitially fouded the Node Self-Adaptio Model which ca make BSs regulate the coverage rage itelligetly; furthermore uder the premise of eough coverig rage, the overall power-savig arithmetic is put forward, which would exert effect of savig atural resources[4]. 2.Node Self-Adaptio Model 2.1 Model Backgroud I the wireless etwork base statio, the source of sigals ad the foudatio of commuicatio ca be foud. I fact, excludig some special occasios, there is early o differece betwee BSs withi a certai regio; therefore, i this model, the base statios ca be treated as idetical odes. Covetioally, the mai duty of BS is trasmittig sigals ad the emittig power has to be set by people; as a result, it s difficult to realize coverage selfoptimizatio. However, i the Node Self-Adaptio Model, the self-regulatio of BS s trasmittig power ca be achieved via iteractio betwee BS ad MT [5][6]. 2.2 Maitaiig the Itegrity of the Specificatios Uder the Node Self-Adaptio Model, all BSs trasmittig power will ot be regulated by people, o the cotrary, it will be cotrolled by a server. Suppose p 0 be the smallest receivig power for MT which is able to support the fudametal commuicatio ad it likes a kid of threshold. Whe MT coects with a certai BS, MT s sigal receivig power ca be set to the server through curret etwork i real time the the server would aalyze ad deal with the accepted data, which is followed by the trasmittig power regulatig o that BS util it ca meet the eeded commuicatio of cliet. (The process is illustrated i Fig. 1) Figure 1: Process of the Model 3. Establishmet of Self-Optimizatio 2

3 3.1 Path-Loss Model Uder Coverage I the Node Self-Adaptio Model metioed above, the server has to regulate BS s sigal trasmittig power; but the wireless commuicatio problems faced by MT are always caused by the formatio of coverage dead zoe, resultig from the isufficiet BS trasmittig power. Hece the determiatio of regulated power that provides by the server that is able to meet ecessary coverage eeds to be discussed more deeply. Based o the actual situatio, the coverage rage of a sigle BS ca be approximately regarded as a circle with radius: r ad the value of r is almost determied by the trasmittig power of the BS. Durig the process of sigal propagatig, the power loss would icrease with loger distace betwee MT ad the sigal ceter. I this case, the path-loss ca be expressed as: P L ( d i d 0) (3.1) Amog the expressio, d is the sigal propagatig distace. d 0 is a referece distace ad i wireless chael the value of d ca be set as 1m. Expoet 0 i is relevat to the sigal propagatig eviromet ad the further discussio of i is as follows. Whe the height of BS s atea is fixed, the value of i is estimated to 2 i free space; however, uquestioable sigal may be affected by obstacle probably durig propagatig ad the impedimet would make i bigger. Suppose p (d ) be the probability of beig impeded durig propagatig ad it is related to d. I additio, α represets the atteuatio of sigal amplitude caused by obstacle. I this sese: i=2+2 p (d )* l α (3.2) Furthermore, whe the trasmittig power of BS is costat, the average sigal receivig power of MT will logarithmically decay with distace. As a result, the sigal receivig power of MT ca be expressed as (i decibel) : P r =P C P L (d 0 ) 10 (2+2 p (d ) * l α) * l ( d d 0) (3.3) I the equatio, P is the trasmittig power of BS [7]. C P L (d 0 ) is the power loss for the referece poit which is d 0 meters away from the coverage ceter ad it is always measured i reality. 3.2 Optimized Power Value for Coverage-Shortage By utilizig the Path-Loss Model, the optimized power value for solvig wireless etwork coverage-shortage problem ca be determied by calculatio. Uder the premise of MT, it is able of detectig the curret etwork sigal while cosiderig the coditio that MT s receivig power fails to support the commuicatio as demaded. 3

4 Figure 2: Picture based o Path-loss model As show i Fig. 2, the mobile termial M is located at the shaded area B. I order to provide M with ormal commuicatio, it is ecessary to elarge the efficacious coverage area of BS (the efficacious coverage area of the BS is A). Assumig that the receivig power of M be P, the o the basis of computig ad the Path-Loss Model, ew radius of coverage after M power regulatio ca be reached: e P C P M P L(d 0) r=d 0 * (3.4) 10 (2+2 p( r )*l α ) I this represetatio, the value of p (r ) is determied by actual coditio. With the calculated value: r, we could draw the regulated BS power which is able to meet the coverage: (3.5) Amog the expressio, P R is the BS power after regulatio. P C is BS s origial trasmittig power. p (r 0 ) is the probability of beig impeded uder the origial coverage rage. What s more, i order to avoid the sigal coverage margial effect ad simplifyig calculatio, we leave out the referece power P L (d 0 ). The BS ca just employ the calculated power: P to R meet the coverage eeds [8][9]. 4.Coverage Redudacy Self-Optimizatio 4.1 Coverage Overlap Ratio I the wireless etwork, as the coverage rage of BS is estimated to be a circle, the overlap regios betwee several base statios are uavoidable; however, too much coverage redudacy will ot oly result i the power waste but also exacerbate the sigal iterferece betwee adjacet BSs. Cosequetly, it is pricipal to optimize overall power deploymet of base statios. I the Node Self-Adaptio Model, suppose all base statios iside of a certai regio costitute a ode set: ξ. Defie coverage overlap ratio as the proportio of the whole overlappig area ad efficacious coverage area of all odes, recordig as D the [10]: D= R + R +R + +R R R R R = R 1 R 2 R 3 R R i R i R i (4.1) 4

5 I the expressio, R is the coverage area of oe of the odes. is the efficacious i Ri coverage of all odes i ξ. The value of D ca reflect coverage redudacy extet ad the lower D is the higher power by usig efficiecy of the etwork. 4.2 Coverage Redudacy Optimizatio Algorithm I order to improve the power utilizig efficiecy of ξ, uder the premise of providig eough coverage for all mobile termials, descedig coverage overlappig area ca effectively reduce the value of D. As Fig. 3 shows the shaded parts: C 1 ad C 2 which represet the overlapped area of adjacet odes. The dimiish the weights of C, 1 C i 2 R i to reduce coverage redudacy. Figure 3: Picture-based o Coverage Redudacy Based o o the Node Self-Optimizatio Model, we desiged Algorithm [11]: Firstly, choose oe ode from ξ radomly ad ame it as O, the fid all adjacet odes of O ad their trasmittig powers through the server, form the power set{p, p, p,..., p,}: ad coordiate set: { ( x 1, y 1 ),(x 2, y 2 ),(x 3, y 3 ), ( x, y )} (I the actual cellular etwork, the value of is 6). After that, with the assistace of distace formula betwee two poits, it is able to get the distaces betwee O ad other odes, form the set: {d, d, d,, d } Accordig to the Path-Loss Model, the set of coverage radiuses of O ad its adjacet odes ca be obtaied by calculatio:{r 0, r 1, r 2, r 3,, r }. The choose oe of O s eighbor: O 1 radomly ad compute its coverage radius: r i also the physical distace betwee O 1 ad O : d i. Afterwards, compare d with i (r i ) : If d i >(r i ), the the trasmittig power of O 1 eeds to be icreased so as to elarge its coverage radius ad realize d i =( r i ). If iitially d i =(r i ), the the power of O should be kept uchaged. If 1 d i <(r i ), we defie this circumstace as shadow ad further optimizig about this is eeded. At the momet of optimizig the coverage overlap, the optimizatio of coverage shortage also eeds to be cosidered ad guarateed. I this case, the circumstace of shadow has to be subdivided. Upo receivig the power formula provided i Path-Loss Model, it s easy to compute the sigal trasmittig distace whe the losig power is P (threshold value) ad 0 record it as: R 0. The if (r i ) R 0 <d i, the power of O 1 has to be icreased util 5

6 physical distace equals to the subtractio of coverage radiuses ad threshold distace. O the cotrary, if (r i ) R 0 >d i, the the power has to be decreased util both sides are equal. Fially, if (r i ) R 0 =d i, the power of O 1 remais uchaged. After the process of O 1 metioed above, aother adjacet ode of O 1 would be chose radomly, amed as O 2. If O 1 satisfies the coditio of shadow, regulate the power of O 2 ad realize d j =( r j ). If O 1 does ot satisfy shadow, regulate the power of O 2 ad realize (r j ) R 0 =d. j Do the same process metioed above for all O s adjacet odes ad fially guaratee that every other oe ought to have the power value to be calculated after shadow process [12]. 4.3 Algorithm Prelimiary Verificatio For the purpose of puttig forward the Coverage Redudacy Optimizatio Algorithm, which just provides a feasible solutio for improvig the wireless etwork coverage by usig efficiecy, it has ot bee put ito practice; i this case we just give simple verificatio of this arithmetic here with the assistace of geometric illustratio. Firstly, we cosider the coditio that there are three base statios iside a local wireless etwork. Three iitial power values which are radomly geerated by computer are set to O 2,O ad 1 O respectively (i reality, the trasmittig power values for differet base statios are ot idetical commoly). I this case, based o the algorithm, the geometric figure of this etwork is show i Fig. 4. Iside of it, O 2,O are two adjacet odes of 1 O ad T is the itersectio of O ad 2 O ; furthermore, with the algorithm ad through calculatio, the power utilizatio efficiecy ca be improved to 97.4% after simulatio. Figure 4: Three Base Statios The illustratio metioed above is oly able to partly show the stregth of the provided algorithm ad for wireless etwork with more tha three BSs. Through simulatio, the outcomes are all over 96.2%; but recet researches o this topic ca hardly achieve the figure over 95.0%. [13] Hece this result idicates the sigificace of savig power resource of the Redudacy-Optimizatio arithmetic as well as the stability of this arithmetic to some extet; however, i fact, the power utilizatio efficiecy i real life is cosiderably lower tha this value due to specific eviromets at differet places; therefore, further practical testig is still idispesable before it is employed i real life ad certai modificatios for the algorithm whe facig diverse situatios are also ecessitated. 5. Coclusio I this paper, the Node Self-Adaptio Model ad the Path-Loss Model are firstly established, providig wireless etwork coverage-shortage optimizatio with more feasible ad precise solutio; at meawhile, i order to reduce the coverage redudacy ad optimize the overall power utilizatio efficiecy of the base statios i a certai regio, the two models are utilized thoroughly ad the favorable algorithm is desiged. Uder the premise of meetig the 6

7 coverage eeds, the provided coverage redudacy optimizatio process is helpful i solvig the overlappig problem ad is tested theoretically, which will be more efficacious i practice. Although the whole researchig procedure is still i theory, it is meaigful for improvig the coverage capability of wireless etwork i the future. Refereces [1] L.Roeles, S. Va De Bulcke, W. Joseph, G. Vermeere, L. Martes, Path loss model for wireless arrowbad commuicatio above flat phatom,electroics Letters, vol.42,2006, 1, pp10-11 [2] M. Zhou, Y. Su, Y. B. Xu, Z.A. Deg, W. X. Meg, WLAN idoor locatio method based o artificial eural etwork, High Techology Letters, vol.16, 3, p , September 2010 [3] T. Bi, L. Zi, G.L. Wag, W. S. Zhag, How wireless power chargig techology affects sesor etwork deploymet ad routig, Istitute of Electrical ad Electroics Egieers Ic. Jue 21,pp , 2010, ICDCS [4] X.Y. Zhag, J. Tag, M. Li, The desig of mobile termial security maagemet system based o digital certificate, 3rd Iteratioal Coferece o Advaced Egieerig Materials ad Architecture Sciece, Tras Tech Publicatios Ltd, vol , 2014, pp [5] B. She, X.M. Gu, C.D. Yag, J.H. Zhao, R.H. Yua, No-stadard parts maagemet based o Browser\Server, 2014 Iteratioal Coferece o Advaces i Materials Sciece ad Iformatio Techologies i Idustry, Tras Tech Publicatios, vol , 2014, pp [6] J.Z. Wag, G.F. Yi, X.B. Hu, T. Che, Implemetig method of remote cooperative desig system based o browser/server model, Jisuaji Jicheg Zhizao Xitog/Computer Itegrated Maufacturig Systems, vol.6, 2000, pp [7] Y.H. Yag, Y. P. Su, D.Y. She, Y.G. Xie, study of the pass loss models based o measuremet ad simulatio, 2012 Iteratioal Coferece o Mechaical Egieerig ad Itelliget Systems, Tras Tech Publicatios, vol , 2012, pp [8] H. Wu, B. Che, L. Gua, W. Le, Quatitative evaluatio ad optimized selectio of power supply reliability ehacemet measures, Diali Zidoghua Shebei/Electric Power Automatio Equipmet, vol.35, 5, pp , 2015, May 10. [9] Y. Jig, Z. Fu, M. Qi, J. Wag, Predictio research based o the optimized BP eural etwork for coal calorific value, 2012 Sprig World Cogress o Egieerig ad Techology, May 27, May 30, [10] R. Ha, C. Feg, H. Xia, Y. Wu, Coverage optimizatio for dese deploymet small cell based o at coloy algorithm, IEEE Vehicular Techology Coferece, 2014-November, November, November 24, 2014, 2014 IEEE 80th Vehicular Techology Coferece. [11] J. Jia, J. Che, G. Chag, LiLiag Zhao, GuagXig Wag, Kogzhi yu Juece/Cotrol ad Decisio, vol.22, 11,2007,pp [12] J. Hu, Xiufag Feg, A coverage optimizatio scheme for wireless sesor etworks, Joural of Computatioal Iformatio Systems, vol.9, 20, pp , 2015, October 15 [13] Y. Zheg, Lombard, Athoy, Kellerma Walter, A improved combiatio of directioal BSS ad a source localizer for robust source separatio i rapidly time-varyig acoustic scearios, IEEE Computer Society, p58-63,

Application of Improved Genetic Algorithm to Two-side Assembly Line Balancing

Application of Improved Genetic Algorithm to Two-side Assembly Line Balancing 206 3 rd Iteratioal Coferece o Mechaical, Idustrial, ad Maufacturig Egieerig (MIME 206) ISBN: 978--60595-33-7 Applicatio of Improved Geetic Algorithm to Two-side Assembly Lie Balacig Ximi Zhag, Qia Wag,