TELK OMNIKA V ol. 16, No . 3, J une 2018, pp . 1126 1134 ISSN: 1693-6930, accredited A b y DIKTI, Decree No: 58/DIKTI/K ep/2013 1126 ETDCC: Ener gy-Efficient T ransmission Sc heme f or Dynamic Climatic Conditions in WSN Sunil K umar* , Prateek Raj Gautam , T arique Rashid , Aksha y V erma , and Ar vind K umar Motilal Nehr u National Institute of T echnology Allahabad (211004) India *Corresponding A uthor , email: rel1516@mnnit.ac.in Abstract In this paper , an energy-efficient tr ansmission scheme f or dynamic climatic conditions (ETDCC) has been proposed in wireless sensor netw or ks (WSNs). This scheme is based on IEEE802.15.4 standard. In this method, open-loop and closed-loop f eedbac k systems are used f or sno wf all v ar iation. An open-loop system is utiliz ed f or sno wf all-a w are link quality compensation and estimation. Ho w e v er , closed-loop system aids to split the net w or k into tw o logical regions , resulting the o v erhead of total control pac k ets is minimiz ed. According to link quality changes due to sno wf all v ar iation, the tr ansmitting po w er is decided on the basis of current n umber of neighbor nodes and threshold po w er loss f or each region. The sim ulated results depict that the proposed scheme with reduced control pac k ets o v erhead adjusts tr ansmit ting po w er le v el ( P lev el ) to compensate link quality . This scheme based on threshold le v el is compared with the con v entional approach that compr ises the division of regions without threshold le v el. K e yw or ds: Energy-efficient; sno w; tr ansmitter po w er ; link quality; WSNs . Cop yright c 2018 Univer sitas Ahmad Dahlan. All rights reser ved. 1. Intr oduction Energy is an impor tant aspect in the design of WSNs . In order to tr ansf er data without dif- ficulty o v er wireless channels in WS Ns , an adaptiv e approach is used f or data tr ansmission to set the lo w est tr ansmitting po w er and maintain reliability . This approach either diminish unnecessar y po w er consumption or interf erence among the nodes . T o tune po w er le v el tr ansmission, a beacon message is sent per iodically through ref erence node to receiving node and respectiv ely ac kno wl- edgment(A CKs) is bac k to the ref erence node f or sending data. Through these inter action, the ref erence node retain connectivity among the nodes [1, 2]. An algor ithm, kno wn as Local Mean Algor ithm (LMA), is used b y the ref erence node to broadcast the Lif eMsg message . This message is tr ansmitted b y the neighbor nodes after receiving the Lif eMsg. Ref erence node calculates all Lif eA CKMsg and sets the tr ansmitting po w er to maintain proper connectivity . If Lif eA CKMsg is more than NodeMaxThresh, tr ansmitting po w er is decreased. In contr ast, if Lif eA CKMsg is less than Node MaxThresh, the tr ansmitting po w er is increased b y the ref erence node . Through this approach, the net w or k lif etime impro v ement is obtained sufficiently , b ut LMA estimates only connectivity among the nodes and cannot pro vide ac kno wledgment about link quality [3, 4]. Stability of netw or k based on lif etime is one of the major challenge f or WSNs . An scheme is proposed to estimate tr ansmitter po w er le v el. T r ansmitter po w er is also controlled b y the de- plo yment of node in the proper place to lessen the distance f or comm unication [5]. Since loss is directly relat ed to en vironment v ar iation, an adaptiv e tr ansmission po w er control is used according to spatial and tempor al eff ects . This scheme helps to acclimate link quality according to v ar iations b y closed-loop f eedbac k. Whereas it is v er y difficult to suppor t large WSNs due to the creation of o v erhead that is required to adjust tr ansmitting po w er of each estab lishing link [6, 7]. The e xisting scheme estimates the link quality with broadcasting a bea- con message per iodically . Fur ther f eedbac k control process helps to adapt controlled tr ansmitting po w er . In acclimation of link quality f or wireless netw or k en vironment where en vironment v ar i- ation e xists , pac k et o v erhead f or tr ansmitting po w er control w ould be diminished. An impor tant technique reduces the n umber of control pac k ets and k eeps retain reliability issues [8, 9, 10]. Receiv ed December 22, 2017; Re vised Apr il 1, 2018; Accepted Apr il 18, 2018 DOI  10.12928/TELKOMNIKA.v16i3.8513 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA ISSN: 1693-6930 1127 The contr ib ution of this paper considers as the proposed scheme is based on tw o pro- cesses kno wn as open loop and closed loop f eedbac k. The open loop is used to calculate and compensate link quality due t o diff erent w eather conditions . Ho w e v er closed-loop f eedbac k pro- cess is used to divide the whole netw or k into tw o logical regions to diminish the o v er lo ad of pac k ets . So that the tr ansmitting po w er le v el to compensate the link quality and sa v e the po w er to increase the lif etime of the sensor nodes . The str ucture of the paper is organiz ed as f ollo w: Section 2, descr ibes the proposed ETDCC reduction scheme is presented to lessen the losses and reduce the pac k et o v erhead. Sim ulation results and their discussions are presented in section 3. Finally , w e ha v e concluded the paper in section 4. 2. Pr oposed W ork In the current section, an energy efficient tr ansmission scheme (ETDCC) has been pro- posed that maintains link quality according to v ar iation in WSN. Th is scheme is designed f or tr ansmission po w er control dividing the netw or k into tw o logical regions (H, L) and using open- loop and closed-loop f eedbac k process . The adv antage of the open-loop and closed-loop process is to lessen o v erhead tr ansmitting po w er significantly which is based on sensed sno wf all r ate in- f or mation. The logical regions H and L denote high loss and lo w loss respectiv ely on the basis of threshold loss . ETDCC w or ks into tw o phases i.e . initial phase and r un-time phase . In the initial phase , a mod el is fr amed b y each node f or its respectiv e neighbor link. While r un-time phase , based on the division of netw or k, retains the link quality with respect to time in the proposed scheme . In order to maintain the sta bility of netw or k, the link quality is measured in initial phase to control o v erhead. Ne v er theless the link quality is contin u ously v ar ying in unstab le netw or k, initial phase is repeated and ser ious o v erhead tak es place . In this scheme , the some par ameters are e xpressed as f ollo ws: (i) Number of current neighbor nodes n c ( t ) (ii) Desired n umber of neighbor nodes n d ( t ) . (iii) Error e ( t ) = n d ( t ) n c ( t ) . (iv) T r ansmission po w er le v el ( p l ev el ) T o tune the tr ansmission po w er , ( p l ev el ) deter mines the connectivity with its neighbor nodes . The po w er cont roller calibr ates tr ansmission po w er le v el b y compar ing the n umber of ini- tial neighbor nodes with the desired n umber of neighbor nodes . Loss and A CKs of a node are used to deter mine estab lished link connectivity . A CK cannot deter mine the link quality b ut it can estimate the estab lished link connectivity . The atten uation is used f or connectivity estimation that estimates connectivity r ather lo w o v erhead. The tr ansmission po w er le v el has been adjusted b y analyzing sno wf all r ate and n umber of current neighbor nodes . The close-loop oper ates not only to compare the n umber of initial neighbor nodes with desired nodes b ut also using sno wf all com- pensated po w er le v el. Hence , the desired po w er le v el is obtained r apidly . If sno wf all is changing then sno wf all is compensated on the basis of connection betw een sno wf all and atten uation. The lo w o v erhead maintains netw or k c o nnectivity b y decreasing tr ansmitting po w er among the nodes . This is possib le due to division of regions in tw o par ts where link quality is changing with sno wf all v ar iation. Due to sno wf all v ar iation, tr ansmitting po w er loss is e xpressed using relationship be- tw een atten uation and sno wf all. The sno wf all atten uation has been classified into tw o categor ies kno wn as dr y and w et atten uation. The loss due to sno wf all is e xpressed as f ollo ws [11]: A snow ( l oss )[ dB =k m ] = aS b (1) Where , the par ameters a and b f or dr y sno w are a = 5 : 42 10 5 + 5 : 4958776 ; b = 1 : 38 And the par ameters a and b f or w et sno w are a = 1 : 023 10 4 + 3 : 7855466 ; b = 0 : 72 where is w a v elength. ETDCC: Energy-Efficient T r ansmission Scheme f or Dynamic Climatic ... (Sunil K umar) Evaluation Warning : The document was created with Spire.PDF for Python.
1128 ISSN: 1693-6930 T o compensate estimated atten uation from equation (1), the output po w er of the tr ans- mitter shoul d be controlled accordingly . The relationship betw een loss and required tr ansmitter po w er le v el is e xpressed b y equation (2) using least square appro ximation [12]. P l ev el [ dB m ] = ( l oss + 40) 12 2 : 91 (2) With the help of equation (1) and (2), w e ha v e obtained the po w er le v el to compensate loss due to sno wf all v ar iation. Due to distance betw een each sensor node in WSN en vironment, path loss assists to calculate actual suitab le tr ansmitting po w er . After ad dition of po w er loss due to sno wf all v ar iation and distance giv en in equation (3), w e calculate actual suitab le tr ansmitter po w er f or all sensor nodes . In free space path loss model, f ollo wing par ameters are used- Number of nodes ( N ) , spectr al efficiency ( ), required ( E b = N o ) depends upon SNR, Boltzmann constant ( k ), Bandwidth ( B ), distance betw een each sensor node ( d ) , receiv er noise figure ( R N F ), fre- quency ( f ) [13]. P t [ dB m ] =   E b N 0 k T B 4 d 2 + R N F ! + l oss (3) In this scheme , the main aim is to control the tr ansmitting po w er b y compensating loss v ar iations based on sno wf all changes sensed at each node . This scheme helps to gather inf or ma- tion from sno wf all sensor and does not need an y comm unication o v erhead which is occurred due to neighbor nodes . The closed-loop f eedbac k control comple xity is reduced b y open-loop control process significantly f or tr ansmitting po w er control. The some specific par ameters a re defined f or this scheme as (1) Threshold loss f or both regions , (2) Desired n umbers of neighbor ing nodes in both regions , n d ( t ) = n c ( t ) 5 , (3) T r ansmitting po w er le v el f or both r egions . Threshold loss should be minim um to retain link reliability . A beacon message is broadcast b y ref erence node repetitiv ely to neighbor ing nodes and A CKs is being w aited. If A CKs are successfully deliv ered from neighbor nodes then Losses ar e calculated f or both logical regions , high loss is considered in reg ion H and lo w loss is considered in region L. If ( l oss l ossthr eshol d ) and ( N cur r ent N desir ed ) then threshold tr ansmitter po w er le v el assigned if f or the similar case ( N cur r ent < N desir ed ) then similar tr ansmitter po w er assigned and if ( l oss < l ossthr e s hol d ) then b y def ault k eep same tr ansmitter po w er le v el. 3. Sim ulation Results and Discussions In this scheme , se v er al measurements h a v e been presented to estimate the a v ailability of estab lished link. The sim ulation results of proposed scheme are compared with the con v en- tional approach f or efficient po w er tr ansmission in WSNs . In Figure 1 w e ha v e sho wn v alues of meteorological sno wf all loss and tr ansmission po w er loss f or one round th at is sensed b y each sensor node in WSNs . It is considered that 100 nodes are r andomly deplo y ed in 100 x 100 m 2 area and sno wf all r ate ha v e v alues in the r ange(1 to 10)mm/hr[11] f or meteorological w eather condi- tion. The ref erence node has been placed at the edge of considered region. The Figure 1 depicts nodes and sno wf all v ar iation on x-axis and y-axis respectiv ely . The sensor nodes are r andomly placed in considered area and respectiv e v ar iation in sno wf all r ate can be seen f or diff erent nodes in WSNs . The v alue of Sno wf all r ate f or all sensor nodes , based on meteorological w eather condi- tions , is used to calculate po w er loss (dBm) that is e xpressed as tr ansmitting po w er loss . Sensed snow w et and tr ansmitt ing po w er losses ha v e been depicted in Figure 2 f or each sensor node . High (H) tr ansmitting po w er loss (dBm) means that the sno wf all r ate of sensor nodes ,placed in the considered region, is high. It represents that the estab lished links do not ha v e good quality . F or snow dr y , sno wf all r ate is in r ange (1 to 10)mm/hr , losses (dBm) ha v e v alue in r ange 0.53dBm to 11.38dBm. F or snow w et , sno wf all r ate is in r ange 0.4 mm/hr to 1.8 mm/hr , losses occur betw een 0.35dBm to 1.72dBm. The se v er al sim ulation par ameters are presented here such as round(r) 1500, distance(1-87)m, nodes(N) 100, Regions H and L, 0.0029, Signal to Noise Ratio (SNR) 0.20dB , frequency( f ) 2.45GHz, R N F 5dB , Bandwidth ( B ) 83.5MHz, and r atio of energy per bit to the spectr al noise ( E b = N o ) 8.3dB . TELK OMNIKA V ol. 16, No . 3, J une 2018 : 1126 1134 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA ISSN: 1693-6930 1129 Figure 1. Sensed sno w dr y r ate and sno w dr y loss f or nodes Figure 2. Sensed sno w w et r ate and sno w w et loss f or nodes ETDCC: Energy-Efficient T r ansmission Scheme f or Dynamic Climatic ... (Sunil K umar) Evaluation Warning : The document was created with Spire.PDF for Python.
1130 ISSN: 1693-6930 Figure 3. P l ev el (sno w dr y ) f or nodes Figure 4. P l ev el (sno w w et ) f or nodes TELK OMNIKA V ol. 16, No . 3, J une 2018 : 1126 1134 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA ISSN: 1693-6930 1131 Figure 5. P t ( snow dr y ) f or nodes Figure 6. P t ( snow w et ) f or nodes ETDCC: Energy-Efficient T r ansmission Scheme f or Dynamic Climatic ... (Sunil K umar) Evaluation Warning : The document was created with Spire.PDF for Python.
1132 ISSN: 1693-6930 Figure 7. P l ev el ( snow dr y ) using con v entional and ETDCC scheme f or nodes Figure 8. P l ev el ( snow w et ) using con v entional and ETDCC scheme f or nodes TELK OMNIKA V ol. 16, No . 3, J une 2018 : 1126 1134 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA ISSN: 1693-6930 1133 T ab le 1. Estimated par ameters f or single round P arameter V alue snow dr y snow w et N (H, L) 54, 46 52, 48 Desired neighbors (H, L) 49, 41 47, 43 Current neighbors (H, L) 47, 38 45, 40 Threshold P l ev el (H, L) 41.79, 58.19 36.65, 35.07 Nodes abo v e the threshold loss (H, L) 25, 22 24, 23 Nodes belo w the threshold loss (H, L) 18, 17 19, 16 Threshold loss (H, L) 3.07, 2.83 1.38, 0.72 F rom Figure 1 and Figure 2, it can be seen that the po w er loss and link quality ha v e an in v erse relation and when sno wf all r ate increases , loss increases . Hence , It concludes that there is lo w link quality at higher sno wf all r ate and vice v ersa. So , the link quality is better at small sno wf all r ate . After estimating the po w er loss f or each sensor , w e calculate their tr ansmitting po w er le v el to compensate the loss . Figure 3 and Figure 4 present the r ange of po w er le v el f or v ar iation of po w er loss on y-axis that is 34.53dBm to 68.88dBm f or snow dr y and 34.10dBm to 37.6dBm f or snow w et . The po w er le v el of sensor node is changing according to sno wf all v ar iation, i.e . the lo w po w er le v el is required at lo w sno wf all r ate and vice v ersa. The po w er le v el ( pow er l ev el ) of each sensor node has been compensated on the basis of sno wf all r ate to compensate the tr ansmitting po w er loss . In WSNs , the path loss occurs due to distance and sno wf all v ar iation betw een each sensor node . A free space model helps to calculate the tr ansmitter po w er ( p t ). After calculation of po w er le v el due to sno wf all v ar iation and distance , w e calculate required tr ansmitter po w er compr ising tr ansmitting po w er loss due to sno wf all r ate and free space path loss f or all nodes . Figure 5 and Figure 6 depict the tr ansmitter po w er r ange on y-axis that is -24.47dBm to -13.62dBm f or snow dr y and -24.70dBm to -23.3dBm f or snow w et . Fur ther the netw or k is divided in tw o logical regions (H,L) f or the analysis of proposed scheme . In the sim ulatio ns , w e ha v e tak en 1500 rounds f or analysis . T ab le 1 sho ws the estimated par ameters f or single rou nd based on loss and threshold loss f or each region. The threshold po w er le v els ha v e been estimated f or both logical regions . Losses are calculated and descr ibed tr ansmitting po w er loss due to sno wf all v ar iations . After sensed sno wf all at each node , w e ha v e assigned the loss to each node . On the basis of threshold loss f or both logical regions , n umber of nodes are counted f or each region. Nodes ha v e high loss in region(H) and lo w loss in region(L). T ab le 2. Compar ision of proposed technique to con v entional technique Climatic region P l ev el ( dB m ) P l ev el new ( dB m ) S now dr y (H) 58.1807 50.8787 S now dr y (L) 40.2114 31.9903 S now w et (H) 36.7412 36.6419 S now w et (L) 35.0116 34.8585 After calculating losses of nodes in both regions , the po w er le v el is calculated f or each node in each region that are clear ly sho wn in Figure 7 and Figure 8. Through Figure 7 and Figure 8 , the diff erence betw een con v entional po w er le v el and proposed po w er le v el ( P new l ev el ) can be seen clear ly f or snow dr y and snow w et . F rom Figure 7 an d Figure 8, it is clear the required p o w e r le v el decreases in both regions . The proposed scheme achie v es that the tr ansmission po w er loss is better than the con v entional tr ansmission po w er loss . ETDCC: Energy-Efficient T r ansmission Scheme f or Dynamic Climatic ... (Sunil K umar) Evaluation Warning : The document was created with Spire.PDF for Python.
1134 ISSN: 1693-6930 4. Conc lusion This paper has pro vided a scheme ETDCC to deter mine sno wf all eff ect on the node link quality . The sno wf all is the significant f actor that aff ects the v ar iation of link quality . The connection betw een atten uation and sno wf all ha v e been e v aluated f or tr ansmitting po w er control of nodes . The open-loop system is used to compensate the change of link quality according to sno wf all v ar iation. Ho w e v er , close-loop f eedbac k has minimiz ed the o v erhead of t r ansmitting po w er control in WSNs which is helpful to adjust the tr ansmitting po w er according to the v ar iation of link quality and thus , the netw or k lif etime is enhanced significantly . The sim ulation results conclude that the po w er le v el f or high and lo w region is reduced more than 13 % and 23 % respectiv ely with respect to con v entional approach. In fut ure , the po w er le v el can be estimated consider ing sensor nodes mo v ement f or diff erent w eather conditions . Ref erences [1] M. T ahir , N. J a v aid, Z. A. Khan, U . Qasim, and M. Ishf aq, “East: Energy-efficient adaptiv e scheme f or tr ansmission in wireless sensor netw or ks , in Electr ical and Computer Engineer- ing (CCECE), 2013 26th Ann ual IEEE Canadian Conf erence on . IEEE, 2013, pp . 1–4. [2] M. Alhilali, J . Din, M. Sch ¨ onhuber , and H. Y . Lam, “Estimation of millimeter w a v e atten uation due to r ain using 2d video distrometer data in mala ysia, Indonesian Jour nal of Electr ical Engineer ing and Computer Science , v ol. 7, no . 1, pp . 164–169, 2017. [3] F . La vr atti, A. Cer atti, D . Prestes , A. Pinto , L. Bolzani, F . V argas , C . Montez, F . Her nandez, E. Gatti, and C . Silv a, “A tr ansmission po w er self-optimization technique f or wireless sensor netw or ks , ISRN Comm unications and Netw or king , v ol. 2012, p . 1, 2012. [4] M. H. Misr an and S . K. A. Rahim, “Optim um tr ansmitter receiv er r atio f or maxim um wireless energy tr ansf er , Indonesian Jour nal of Electr ical Engineer ing and Computer Science , v ol. 5, no . 3, pp . 599–605, 2017. [5] V . G. Douros and G. C . P olyz os , “Re vie w of some fun damental approaches f or po w er control in wireless netw or ks , Computer Comm unications , v ol. 34, no . 13, pp . 1580–1592, 2011. [6] T . Rashid, S . K umar , and A. K umar , “Reer : Rela y based energy efficient routing f or intr a body sensor netw or k (intr a-wbsn), in Signal Processing and Integ r ated Netw or ks (SPIN), 2017 4th Inter national Conf erence on . IEEE, 2017, pp . 222–227. [7] R. Kar thik, J . Nagar aju, and M. V ucha, “Distr ib uted optimal rela y selection in wireless sensor netw or ks , Indonesian Jour nal of Electr ical Engineer ing and Computer Science , v ol. 7, no . 1, pp . 71–74, 2017. [8] N. A. Latiff , M. Ru s l ee , S . S . Y usof , M. A. Rahim, H. Sa yuti, K. M. Y usof , and M. Bahar udin, “A tr aining monitor ing system f or cyclist based on wireless sensor netw or ks , Indonesian Jour nal of Electr ical Engineer ing and Computer Science , v ol. 6, no . 1, pp . 80–87, 2017. [9] A. M. D . Masood and S . Muthusun dar , “Lo ad balance: Energy efficient routing protocol in wireless sensor netw or k, Indonesian Jour nal of Electr ical Engineer ing an d Computer Sci- ence , v ol. 9, no . 3, 2018. [10] S . Sar a v anan, “Efficient and energy scheme f or wireless rechargeab le sensor netw or k, In- donesian Jour nal of Electr ical Engineer ing and Computer Science , v ol. 9, no . 2, 2018. [11] F . Nadeem, S . Chessa, E. Leitgeb , and S . Zaman, “The eff ects of w eather on the lif e time of wireless sensor netw or ks using fso/rf comm unication. Radioengineer ing , v ol. 19, no . 2, 2010. [12] K. Bannister , G. Giorgetti, and S . K. Gupta, “Wireless sensor netw or king f or hot applications: Eff ects of temper ature on signal strength, data collection and localization, in Proceedings of the 5th W or kshop on Embedded Netw or k ed Sensors (HotEmNets 08) . Citeseer , 2008. [13] M. T ahir , N. J a v aid, A. Iqbal, Z. A. Khan, and N. Alr ajeh, “On adaptiv e energy-efficient tr ansmission in wsns , Inter national Jour nal of Distr ib uted Sensor Netw or ks , v ol. 9, no . 5, p . 923714, 2013. TELK OMNIKA V ol. 16, No . 3, J une 2018 : 1126 1134 Evaluation Warning : The document was created with Spire.PDF for Python.