Inter national J our nal of Electrical and Computer Engineering (IJECE) V ol. 11, No. 3, June 2021, pp. 2350 2359 ISSN: 2088-8708, DOI: 10.11591/ijece.v11i3.pp2350-2359 r 2350 A h ybrid objecti v e function with empirical stability awar e to impr o v e RPL f or IoT applications Abdelhadi Eloudrhiri Hassani, Aicha Sahel, Abdelmajid Badri, El Mourabit Ilham EEA and TI laboratory , F aculty of Sciences and T echniques, Hassan II Uni v ersity , Casablanca, Morocco Article Inf o Article history: Recei v ed Aug 16, 2020 Re vised Sep 6, 2020 Accepted Okt 1, 2020 K eyw ords: Combined metrics Contiki OS IoT Objecti v e function RPL WSN ABSTRA CT The di v erse applications of the internet of things (IoT) require adaptable routing pro- tocol able to cope with se v eral constraints. Thus, RPL protocol w as designed to meet the needs for IoT netw orks cate gorized as lo w po wer and lossy netw orks (LLN). RPL uses an objecti v e function based on specific metrics for preferred parents selection through these pack ets are sent to root. The single routing metric issue generall y doesn’ t satisfy all routing performance requirements, whereas s ome are impro v ed others are de graded. In that purpose, we propose a h ybrid objecti v e function with empirical sta- bility a w are (HOFESA), implemented in the netw ork layer of the embedded operating system CONTIKI, which combines linearly three weighty metrics namely hop count, RSSI and node ener gy consumption. Als o, T o remedy to frequent preferred parents changes probl ems caused by taking into account more than one metric, our proposal relies on static a nd empirical thresholds. The designed HOFESA, e v aluated under COOJ A emulator ag ainst Standard-RPL and EC-OF , sho wed a pack et deli v ery ratio impro v ement, a decrease in the po wer consumption, the con v er gence time and DIO control messages as well as it gi v es netw ork stability through an adequate churn. This is an open access article under the CC BY -SA license . Corresponding A uthor: Abdelhadi Eloudrhiri Hassani EEA and TI laboratory F aculty of Sciences and T echniques, Hassan II Uni v ersity , Morocco Email: eloudrhiri.abdelhadi@gmail.com 1. INTR ODUCTION The internet of things (IoT) is a wireless communication technology with a great potential for human- ity which pro vides access f aciliti es to the ph ysical w orld e v erywhere all the time. Thi n gs term in IoT mak e direct reference to netw ork ed embedded de vices with sensors and actuators [1]. These netw orks embrace a lar ge number of battery-po wered de vices with constraints i n term of processing limitations and storage capac- ity [2]. Systems based on the Internet of Things ha v e countless applications such as smart cities [3], smart home [4], healthcare [5, 6], industries [7] and smart grids [8]. Thereby , IoT netw orks are e xpected to ensure ef ficienc y and reliability in the future e v en if the y are deplo yed in a harsh en vironments. These netw ork ed de vices must be apt to handle data processing, pack et transmission, and ener gy consumption according to their limited capacities. Thus, The MA C and routing protocols are required to respond to those challenging tasks. In that purpose, a routing protocol for lo w-po wer and lossy netw orks (RPL) w as designed by the internet engineering task force (IETF) [9]. The RPL is a routing protocol destinated for limited resource IoT platforms,based on IPv6 and uses the IEEE802.15.4 at the PHY and MA C layer [10]. W ith RPL as a proacti v e routing protocol, the paths are constructed once the netw ork is initialized. The nodes in the netw ork use the RPL in order to set up a tree-lik e routing topology which is a destination-oriented directed ac yclic graph (DOD A G), J ournal homepage: http://ijece .iaescor e .com Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 2351 based on four principal ICM Pv6 messages: DOD A G information object (DIO) holds information that enable nodes to kno w the instance, configuration and select the preferred parent, DOD A G information solicitation (DIS) message as a DIO request from neighbours, DOD A G adv ertisement object pack ets (D A O) used to collect topology informations and D A O-A CK as a response to a D A O message. The aim of this kind of topology is to steer all the data pack ets to one or more sink nodes . These routing paths are created using a specific objecti v e function (OF). In the core of RPL, tw o OF’ s are proposed namely the minimum rank with h ysteresis objecti v e function (MRHOF) [11] based on e xpected transmission counts (ETX) as a routing metric and OF0 [12], which is based on hop count. The paths in MRHOF are based on the link quality metric calculated by broadcasting probe pack ets between the sender and recei v er nodes at time interv als. On the other hand, the routing paths in OF0 are based on node metric that aims to each node in the netw ork to kno w its position ag ainst the sink. Ho we v er , the tw o objecti v e functions tend to minimize the cost of their metrics which causes non-optimized routes due to taking into account a single constraint. Considering those dra wbacks, we propose a designed h ybrid objecti v e function with e mpirical stability a w are (HOFESA), based on a ne w method of rank processing using linear combination of radio signal strength indicat or (RSSI), hop count (HC) and ener gy consumption (EC) metrics in order to select the optimal preferred parent taking into account se v eral constraints and stability through an static or empirical threshold. The main contrib utions of this paper are summarized as follo ws: Enhance the RPL by a ne wly designed objecti v e function HOFESA. Our proposal combines linearly three dif ferent metrics, with dif ferent weights, chosen to respect the objecti v e function con v er gence, cope with dif ferent constraints Consider the stability in routing by tw o diferrent thresholds A simulation under Cooja of HOFESA compared to Standard-RPL and EC-OF i n term of P ack et Deli v ery Ratio, Ener gy consumption, Con v er gence time, churn and number of DIO control messages The rest of this paper is or g anized as follo w . In section 2 we present the related w orks with RPL routing protocol. In section 3 we present the proposed objecti v e function HOFESA. In section 4 we report the e xperimental results and discussion, finally a conclusion is gi v en in section 5. 2. RELA TED W ORKS The RPL protocol has been the subject of se v eral researches aiming to enhance or adapt it to dif fere nt requirements since only three metrics are implemented in RPL core i.e. e xpected transmission count, ener gy consummed and hop count. Authors were interested first on these objecti v e functions in order to study their adv antages and limits [13–15]. Ho we v er , in term of single m etrics, Xiao et al. [16] proposed the a v erage Expected T ransmission Count of the path to w ards the sink. This proposal address the problem of single long hops introduced in high densities b ut can’ t cope with po wer consumption of the netw ork. Remaining ener gy as a node metric for n e xt hop selection has been defined by kamgueu et al. in [17] to choose parents with the most residual ener gy . This ne w metric has pro v en to be ef fecti v e in terms of e xtending the netw ork lifetime and di strib ute ener gy e v enly among nodes, b ut do not considers the quality of links which leads to choose lossy paths. In order to cope with bottlenecks issue generated by long hops in high densities, San Martin et al. in [18] proposed Sigma-ETX metric based on the standard de viation of ETX in each route. The best path is that with the minimum number of de viations. Ho we v er , the ener gy consumption also is not considered by the authors which conducts to f ast nodes depletion. The delay issue w as considered by Gonizzi et al. in [19] with a no v el objecti v e function based on A V GDELA Y metric which b uilds the netw ork arborescence to minimize the routing a v erage delay between senders and root. The proposal sho wed a significant decrease in terms of latenc y b ut can’ t cope with reliability performances. The requirements of each application dif fers which lea v es the single routing metr ic not the ideal im - plementation to of fer the best quality of routing services. In this conte xt, se v eral w orks ha v e been proposed based on metrics amalg amation. RPL adaptation for smart grid applications are proposed by Nassar et al. in [20] who designed OFQS objecti v e function which combines ETX, del ay and po wer state metrics. It uses the concept of h ysteresis in order to ensure path stability selection and reduce the churn. It sho wed a lifetime, end to end delay and pack et deli v ery ratio impro v ements as results. In [21], Gao et al. proposed ETEN-RPL a h ybrid routing metric that combine ETX and rema ining ener gy additi v ely injected into an objecti v e function. It remediates to the problems of unbalanced ener gy and bad data reliability , impro v es the stability and reduces the po wer consumption in the netw ork. Ho we v er , the proposal is limited to performance results only in v ery A hybrid objective function with empirical stability awar e to... (Abdelhadi Eloudrhiri Hassani) Evaluation Warning : The document was created with Spire.PDF for Python.
2352 r ISSN: 2088-8708 lo w densities. Mishra et al. proposed le xicographic and additi v e approachs in [22] to combine ETX, a v ailable ener gy and hop count routing metrics into EHA objecti v e function for rank processing and select an optimized pref fered parent. The results sho wed better performance in terms of ener gy consumption, netw ork latenc y and pack et deli v ery ratio compared to MRHOF-ETX and OF0. In [23], Al-Kashoash et al. ha v e been interested to the paths conges tion caused by the b uf fer nodes occupanc y . In that purpose, Congestion-A w are Objecti v e Function CA-OF were be proposed which consider ETX metric at lo w data rate whi le the b uf fer occupanc y is considered at high data rate. The proposal is limited to pack et deli v ery ratio, po wer consumption and don’ t e v aluate other performance parameters also don’ t consider a lar ge scale of densities. Man y researches used the fuzzy logic as an approch for combined metrics. Indeed, in [24], Araujo et al. proposed a ne w objecti v e f u nc tion called DQCA-OF that combines three metrics, i.e. ETX, hop count and ener gy consumed. DQCA-OF pro vides a pack et deli v ery ratio o v er 95%, reduce end to end delay and the number of e xcepted transmission count. Ho we v er , the proposal is simulated with a topology of 20 nodes which need other tests for higher densities. Also in [25], Sankar et al. designed FLEA-RPL objecti v e function based on ETX, load and residual ener gy , it is used for calculating the step parameter for rank assignment. The proposal sho ws a rise of PDR around 2% to 5% and an increase of lifetime around 10%. In [26], Lamaazi et al. proposed a ne w objecti v e function based on fuzzy logic system EC-OF that combines tree metrics: ETX, Hop Count and ener gy consumption. The results sho wed that the EC-OF k eeps the routing protocol RPL ef ficient and impro v e its performances in term of PDR, netw ork life time, con v er gence time, latenc y and po wer consumption ag ainst MRHOF . 3. RESEARCH METHOD 3.1. Pr oblem statement By its def ault definition, RPL uses a single routing metric such as ETX, ener gy consumption or hop count for selecting the preferred parents, which leads to sho w some limits and poorly performs in applications where dif ferent constraints must be tak en into account. Indeed, RPL based on one metric minimized by the objecti v e function, conduct to select non-optimized or static routing paths which greatly af fects the netw ork quality of service performances. Also, its note w orth y that when the density of netw ork increase, the frequent preferred parents change phenomenon increases too which destabilizes the netw ork. T o o v ercome these issues, we propose a h ybrid objecti v e function with empirical stabili ty a w are (HOFESA) based on linear combination of se v eral metrics namely RSSI, Ener gy consumption and Hop count to cope with dif ferent constraints while an empirical Threshold is implemented to gi v e more netw ork stabil- ity . Accordingly , HOFESA utilizes di v erse policies from the DIO amendment to rank computing and parent selection procedures. T o ha v e an optimized parent selection with HOFESA, we ha v e equipped the DIO mes- sage with a hop count metric as sho wn in Figure 1, while RSSI and Ener gy consumption metrics are locally computed by each node. Our proposed objecti v e function selects the best parent based on certain priority in- terpreted by weights. Thus, the metric with the highest weight has the most influence in the nodes parent’ s selection. Ho we v er , the design of a no v el objecti v e function using composite routing metrics must respect the monotonicity property defined by (1) to be loop-free directed to sink. Figure 1. Amended DIO message with hop cout metric Int J Elec & Comp Eng, V ol. 11, No. 3, June 2021 : 2350 – 2359 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 2353 ( g 1( ) ; g 2( )) < ( g 1() ; g 2()) , g 1( ) + g 2( ) < g 1() + g 2() ( g 1( ) ; g 2( )) > ( g 1() ; g 2()) , g 1( ) + g 2( ) > g 1() + g 2() (1) Where , are tw o dif ferent routing paths and g1, g2 are the functions which define the primary routing metrics. 3.2. Metrics of inter est In our case, the primary metrics must be chosen in such a w ay that all must be minimized for HOFESA con v er gence adn respect (1). 3.2.1. Hop count metric: It is a metric that sho ws the number of nodes in a routing path, minimizable by the objecti v e function to find the shortest path to sink and processed as in (2). H opC ount ( i ) = M inH C I ncr ement if cp = sink H C ( cp ) + M inH C I ncr ement if cp 6 = sink (2) Where cp is the candidate parent susceptible to be the preferred parent of node i and the v alue of MinHC Increment is 256. 3.2.2. Ener gy consumption metric: It is a metric that compute the current po wer consumption by each node. It is must be minimizable by the objecti v e function in order to choose preferred parents with lo wer ener gy consumption. It is processed as in (3). E C ( i ) = C P U 5 : 4 + T r ans 58 : 5 + List 64 : 5 + LP M 0 : 1635 32768 (3) Where CPU,T rans,List,LPM are respecti v ely the number of ticks when the node is in CPU le v el pro- cessing, transmitting, listening or going to lo w po wer mode [27], while numerical parameters are the nominal v alues pro vided in the Sk ymote Datasheet. 3.2.3. RSSI metric: it is CC2420 radio metric based on the signal strength and pro vided through an RSSI re gister adjusted with antenna v ariation named of fset modeled during the system de v elopment. The RSSI is computed as (4). It is a maximizable metric in order to gi v e the best link between node and neighbor node since it is measured with a log arithmic scale in dBm , typically ranges between 0 dBm and -110 dBm respecti v ely for a v ery strong and lo w signal le v els. It is a metric to maximaze while hop count and ener gy consumption metrics must be minimized. Thus, to respect the designed objecti v e function monotonicity , we used the in v erse of RSSI. R S S I ( i ) = R S S I r eg ister ( i ) 45 (4) 3.3. Design of h ybrid objecti v e function with empirical stability awar e (HOFESA) The HOFESA is based on no v el method of rank processing. Indeed, at a DIO reception from candidate parent, the node measure the RSSI at the MA C layer follo wing (4), it ener gy consumption follo wing (3) and add the candidate parent hop count v alue adv ertised in the DIO message to the MinHC Increment as in (2). After all these computa tions, since the metric priorit y is interpreted by weights, the increment of rank is calculated as in (5). Finally , the node compute its susceptible ne w rank refering to (6). R ank I ncr ement = H C ( i ) + R S S I ( i ) + E C ( i ) (5) Where and are the weights influence of each metric comprised between 0 and 1, while the hop count ha v e a constant weight equal to 1. Also, the we ights of and must be complementary such a w ay that their summation is equal to 1. A hybrid objective function with empirical stability awar e to... (Abdelhadi Eloudrhiri Hassani) Evaluation Warning : The document was created with Spire.PDF for Python.
2354 r ISSN: 2088-8708 R ank ( i ) = R ank ( cp ) + R ank I ncr ement (6) Where the Rank(cp) is the rank e xtracted from the candidate parent DIO. Once the node had process its susceptible ne w rank, if it don’ t already ha v e a preferred parent, the node opt for the ne w calculated rank. Otherwise, a comparison of the susceptible ne w rank with the preferred parent rank is ine vitable, if it is higher , then the candidate parent is discarded. Else, for stability a w areness, if it is lo wer than the preferred parent rank with a certain threshold, then the candidate parent is retained. The threshold is essentiel for reducing the preferred parent changes. F or that purpose, the Static Threshold v alue defined in (7) is used as proper HOFESA threshold while the Empirical threshold defined in (8) is utilized to optimize the performances of our proposal by gi ving more stability . S taticT h r eshol d = M inH C I n c r ement + M inH C I ncr ement 2 (7) E mpir ical T hr eshol d = S taticT hr eshol d + E v al ue (8) Where the Ev alue is defined during the simulations which gi v es the best performance optimization. At this stage, if the preferred parent change condition is fulfilled, the node update it metric container and rank in the DIO message and broadcast a ne w one. T able 1 describes the proposed HOFESA algorithm. T able 1. Proposed HOFESA algorithm When node i recei v e a DIO message from cp IF (cp ! = NULL) base rank = cp.dio.rank; RSSI = -1 * (rssi measure()); EC = ener gy measure(); HC = cp.dio.hc + MinHC Increment; RankIncrement = HC + *RSSI + *EC; IF ((base rank + RankIncrement) < base r ank ) r etur n inf inite r ank ; ELSE IF (Preferred parent == NULL) Best parent = cp; rank = base rank + RankIncrement; ELSE IF ((base rank + RankIncrement) < (Preferred parent.rank + StaticThreshold) Preferred parent = cp; rank = base rank + RankIncrement; hc = HC; ELSE Pr eferr ed par ent do not c hang e exit; ENDIF ENDIF /* Gener ate a ne w DIO messa g e dio.mc.hc = HC; dio.rank = rank; Br oadcast a ne w DIO messa g e ENDIF ENDIF 4. RESUL TS AND DISCUSSION This section pro vide a discussion about the outcome obtained from the proposed implementati on of h ybrid objecti v e function with empirical stability a w are. In order to e v aluate our proposal, we ha v e e xploited the simulation en vironment COOJ A. It is considered as an emulator of netw ork ed embedded platforms running Int J Elec & Comp Eng, V ol. 11, No. 3, June 2021 : 2350 – 2359 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 2355 contiki as an embedded operating system. Our simulations are established with the IoT sk ymotes platforms that use the lo w po wer T e xas instruments MSP430 micro-controll er as CPU and emplo ys for its wireless com- munications the radio module chipcon CC2420. In this w ork, the sk ymotes are randomly deplo yed in an area of 200x200 m. Dif ferent densities are considered from 25 to 100 senders with a single sink that collects all the data of the netw ork. The ef ficienc y of our proposed method has been in v estig ated through its comparison ag ainst standard RPL and EC-OF in terms of fi v e performance metrics namely pack et deli v ery ratio, po wer consumption, con v er gence time, churn and number of DIO control pack ets. Also, in order to gi v e an idea about the impact of and parameters in the rank calculation by our objecti v e function, we took three dif ferent v alues 0 : 3 , 0 : 5 and 0 : 7 as weights. The simulation is p e rformed o v er 600 s for e v ery en vironmental setup. Simulation param eters are s u m marized in T able 2. Ho we v er , this section is di vided on tw o parts, the firs t where we analyse and discuss the performances pro vided by our proposal in term of netw ork densit y , while the second we focus more on the HOFESA stability using the empirical threshold. T able 2. Simulation parameters Netw ork simulator Cooja Embedded operating system Contiki 2.7 Radio en vironment Unit disk graph medium - DL Emulated nodes Sk y motes Netw ork area 200 x 200 m Deplo yment of nodes Random Number of senders 25,50,100 Ev alue 200 Number of sinks 1 T ransmission / interference ranges 70/100 m Objecti v e functions RPL-standard, EC-OF Simulation time 600s 4.1. P erf ormance e v aluation of HOFESA 4.1.1. Analysis of pack et deli v ery ratio The netw ork reliability of HOFESA is assessed by comparing it to Standard RPL and EC-OF in terms of PDR for dif ferent densities. As noticed in Figure 2, the EC-OF pro vide a lo w PDR due to it tend to choose nodes with lo wer po wer consumption despite of poor link quality . Re g arding standard-RPL that tends to minimize the ETX metric aiming to of fer routing paths with good link quality , it cannot perform better ag ainst our proposed method. It can be e xplained by the f act that the se v eral routing metrics h ybridization increases the number of routing paths, thus a v oiding the bottlenecks responsible in pack ets loss. Another aspect that e xplains this impro v ement is the number of preferred parent changes indicated by the churn. Indeed, e v en if our proposal considers more metrics to be optimized compared to the othe r tw o objecti v e functions, it allo ws to reduce this number of changes which gi v es better routes stability then as a result less pack ets loss. On the other hand, re g arding the influence of the weights on the combined metrics especially when the ener gy metric weight is greater than RSSI weight, our proposition sho ws a decrease in the PDR since that the ener gy metric is instantaneous which influences more the rank processing then leads to more routing path changes. Figure 2. P ack et deli v ery ratio for dif ferent netw ork densities A hybrid objective function with empirical stability awar e to... (Abdelhadi Eloudrhiri Hassani) Evaluation Warning : The document was created with Spire.PDF for Python.
2356 r ISSN: 2088-8708 4.1.2. Analysis of po wer consumption A glance on Figure 3 sho ws that HOFESA considerably decrease the po wer consumption since it is based on combining hop count, ener gy and RSSI. Indeed, the first metric leads to minimize the hops to the sink which induces technically less ener gy consumption in retransmission. The second metric tend to of fer a path with lo w po wer consumption and the third metric gi v es a good link quality in routing path which is good to a v oid the retransmission caused by poor links. On the other hand, as we can see in section 4.1.5 our proposal induces less DIO control pack ets which a v oids the e xcess ener gy consumption in the processi ng, sending, and reception of these pack ets. Re g arding the influence of the weights on HOFESA, when the superiority is gi v en to the ener gy metric, there is a slight increase in ener gy consumption because, as mentioned before, it continually changes which leads to more preferred parents changes. Figure 3. Po wer consumption for dif ferent netw ork densities 4.1.3. Analysis of chur n The churn is a essential parameter that se v eral research papers ne glect whereas it is v ery im portant for e v aluating the routing protocol performances. The goal is to minimize i t with non-zero v alues in order to not af fect the netw ork stability and optimize the routing paths continously with less ener gy consumption loss in changing the prefered parent. As noticed in Figure 4, our proposition with a dominant weight on the ener gy metric t ends to frequentl y change its preferred parents, which leads to a performance deterioration in terms of PDR and ener gy consumption as we could see pre viously . Ho we v er , our proposal with a dominant weight for the RSSI metric re v eals an appropriate churn reflecting thus a good PDR with the lo wer po wer consumption. Figure 4. Churn for dif ferent netw ork densities 4.1.4. Analysis of con v er gence time Con v er gence time reflects one of the properties of real-time performance that a routing protocol can pro vide. A short con v er gence time means that the protocol allo ws the nodes to quickly construct the netw ork arborescence. As can be seen in Figure 5, the con v er gence time increases as the density is higher . The results Int J Elec & Comp Eng, V ol. 11, No. 3, June 2021 : 2350 – 2359 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 2357 re v eals that HOFESA can con v er ge f aster notably when increasing the number of nodes in the netw ork. This can be e xplained by the f act that each node, in its rank processing, refers to its o wn metrics without w aiting for those of its candidate parent. Figure 5. Con v er gence time for dif ferent netw ork densities 4.1.5. Analysis of DIO contr ol pack ets The afteref fect of netw ork stability is pro vided by the number of DIO control pack ets. Indeed, as sho wn in Figure 6, our proposal can pro vide a lo w number of DIO pack ets compared to Standard-RPL and OF- EC due to the op t imized paths selected using combined metrics. Concerning t he weight’ s ef fect, we can notice that the instability induced when the ener gy metric ha v e the higher weight, the number of DIO automatically increases. Figure 6. DIO control pack ets for dif ferent netw ork densities 4.2. Stability effect on HOFESA perf ormances This part focuses on the the stability ef fect induced by empirical threshold defined pre viously in (8). F or this, we were interested in our proposal where the churn is v ery lar ge, namely HC + 0.3rssi + 0.7ener gy . Indeed, we obtained this threshold of stability empiri cally in order to find the optimal performances. From the T ables 3, 4 and 5, it can be seen that empirical threshold can decrease the churn, the con v er gence time as well as a pro vide good PDR with less po wer consumption and DIO pack ets. T able 3. Empirical threshold ef fect on HOFESA in density of 25 nodes Stability PDR Po wer(mW) churn DIO Con v er gence(s) Static threshold 0,995 1,207 0,48 6864 15,769 Empirical threshold 1 1,166 0,12 5700 15,490 A hybrid objective function with empirical stability awar e to... (Abdelhadi Eloudrhiri Hassani) Evaluation Warning : The document was created with Spire.PDF for Python.
2358 r ISSN: 2088-8708 T able 4. Empirical threshold ef fect on HOFESA in density of 50 nodes Stability PDR Po wer(mW) churn DIO Con v er gence(s) Static threshold 0,980 1,599 0,88 18130 18,679 Empirical threshold 0,988 1,421 0,26 12205 16,110 T able 5. Empirical threshold ef fect on HOFESA in density of 100 nodes Stability PDR Po wer(mW) churn DIO Con v er gence(s) Static threshold 0,965 2,07 1,12 36956 33,680 Empirical threshold 0,966 2 0,66 34342 27,789 5. CONCLUSION In this paper , an impro v ement of RPL routing protocol based on its objecti v e function w as proposed. Our approach called HOFESA relies on a no v el m ethod for rank processing using three metrics namely Hop count, RSSI and Ener gy consumption to surmount the single routing metric limits. The metrics amalg amation is based on weights that identifies the more influencing in the rank calculation. The netw ork stability is also tak en into account by introduced static and empirical threshold to limit the number of parents changes and impro v e the proposal performances in terms of quality of services. The designed HOFESA, e v aluated ag ainst Standard- RPL and EC-OF , sho wed an enhancement in pack et deli v ery rat io, lo wer po wer consumption, con v er gence time and DIO control messages as well as it ensure netw ork stabili ty through an adequate churn. In the future w ork, we will focus on combination more metrics as traf fic management metric, ETX, b uf fer of nodes and delay , to bring out better performances. A CKNO WLEDGEMENT This w ork has been supported by the T echnology of Information and Communication Center of uni- v ersity Hassan II Casablanca as a part of the Big data and Connected objects research project, and the National Center for Scientific and T echnical Research in Morocco (CNRST). REFERENCES [1] I. Y aqoob, et al., ”Internet of Things architecture: Recent adv ances, taxonomy , requirements, and open challenges, ” IEEE W ireless Communications , v ol. 24, no. 3, pp. 10-16, 2017. [2] A. Musaddiq, et al., ”A surv e y on resource management in IoT operating systems, ” IEEE Access , v ol. 6, pp. 8459–8482, 2018. [3] M. A. Naeem, et al., ”A periodic caching strate gy solution for the smart city in information-centric Internet of Things, ” Sustainability , v ol. 10, no. 7, pp. 2576, 2018, doi: https://doi.or g/10.3390/su10072576. [4] K. Y oshigoe, et al., ”Ov ercoming in v asion of pri v ac y in smart home en vironment with synthetic pack et injection, ” TR ON Symposium (TR ONSHO W) , 2015. [5] Y . A. Qadri, et al., ”The Future of Healthcare Internet of Things: A Surv e y of Emer ging T echnologies, ” IEEE Communications Surv e ys and T utorials , v ol. 22, no. 2, pp. 1121–1167, 2020. [6] R. Ali, et al., ”Q-learning-enabled channel access in ne xt-generation dense wireless netw orks for IoT - based eHealth systems, ” EURASIP Journal on W ireless Communications and Netw orking , v ol. 2019, pp. 1-12, 2019. [7] L. D. Xu, W u He, and Shancang Li ”Internet of Things in industries: A surv e y , ” IEEE T ransactions on Industrial Informatics , v ol. 10, no. 4, pp. 2233–2243, 2014. [8] V . C. Gungor , et al., ”Smart grid technologies: Communication technologies and standards, ” IEEE T rans- actions on Industrial Informatics , v ol. 7, no. 4, pp. 529-539, 2011. [9] R. Ale xander , et al., ”RPL: IPv6 Routing Protocol for Lo w-Po wer and Lossy Netw orks. Internet Engi- neering T ask F orce(IETF), ” RFC 6550 , pp. 1-157, 2012. [10] A. E. Hassani, Aicha Sahel, and Abdelmajid Badri ”A Ne w Objecti v e Function Based on Additi v e Combi- nation of Node and Link Metrics as a Mechani sm P ath Selection for RPL Protocol, ” International Journal of Communication Netw orks and Information Security , v ol. 12, no. 1, pp. 63-68, 2020. [11] O. Gna w ali and P . Le vis, ”The Minimum Rank with Hystere sis Objecti v e Function. Internet Engineering T ask F orce(IETF), ” RFC 6719 , 2012. Int J Elec & Comp Eng, V ol. 11, No. 3, June 2021 : 2350 – 2359 Evaluation Warning : The document was created with Spire.PDF for Python.
Int J Elec & Comp Eng ISSN: 2088-8708 r 2359 [12] P . Thubert, ”Objecti v e Function Zero for the Routing Protocol for Lo w-Po wer and Lossy Netw orks (RPL). Internet Engineering T ask F orce(IETF), ” RFC 6552 , 2012. [13] A. E. Hassani, A ¨ ıcha Sahel, and Abdelmajid Badri ”Impact of RPL Objecti v e Functions on Ener gy Con- sumption in Ipv6 based W ireless Sensor Net w o r ks, ” Colloque International sur les Objets et Syst ` emes Connect ´ es , 2019. [14] S. A. Hak eem, Shimaa A., Anar A. Hady , and HyungW on Kim, ”RPL routing protocol performance in smart grid applications based wireless sensors: Experimental and simulated analysis, ” Electronics , v ol. 8, no. 2, pp. 1-23, 2019. [15] A. E. Hassani, et al., ”Assessment of a proac ti v e routing protocol RPL in Ipv6 based wireless sensor netw orks, ” Third International Conference on Intelligent Computing in Data Sciences (ICDS) , pp. 1-7, 2019. [16] W . Xiao, et al., ”An optimization of the object function for routing protocol of lo w-po wer and Lossy netw orks, ” 2nd International Conference on Systems and Informatics , pp. 515-519, 2015. [17] P . O. Kamgueu, et al., ”Ener gy-based routing metric for RPL, ” HAL archi v es , v ol. 14, 2013. [18] P . Sanmartin, et al., ”Sigma routing metric for rpl protocol, ” Sensors , v ol. 18, no. 4, pp. 1277, 2018. [19] P . Gonizzi, Riccardo Monica, and Gi anluigi Ferrari ”Design and e v aluation of a delay-ef ficient RPL rout- ing metric, ” 9th International W ireless Communications and Mobile Computing Conference , pp. 1573- 1577, 2013. [20] J. Nassar , et al., ”Multiple instances QoS routing in RPL: Application to smart grids, ” Sensors , v ol. 18, no. 8, pp. 1-16, 2018. [21] L. Gao, Zengwei Zheng, and Meimei Huo ”Im pro v ement of RPL protocol algorithm for smart grid, ” International Conference on Communication T echnology Proceedings , pp. 927-930, 2019. [22] S. N. Mishra, et al., ”Eha-rpl: A composite routing technique in iot application netw orks, ” Adv ances in Intelligent Systems and Computing , v ol. 1045, pp. 645-657, 2020. [23] H. Al-Kashoash, et al., ”Congestion-A w are Routing Protocol for 6LoWP ANs, ” Congestion Control for 6LoWP AN W ireless Sensor Netw orks: T o w ard the Internet of Things , pp. 95-107, 2020. [24] H. Araujo, et al., ”A proposal for IoT dynamic routes selection based on conte xtual information, ” Sensors , v ol. 18, no. 2, pp. 1-16, 2018. [25] S. Sankar , and P . Srini v asan, ”Fuzzy logic based ener gy a w are routing protocol for internet of things, ” International Journal of Intelligent Systems and Applications , v ol. 10, pp. 11-19, 2018. [26] H. Lam aazi, and N. Benamar , ”OF-EC: A no v el ener gy consumption a w are objecti v e function for RPL based on fuzzy logic, ” Journal of Netw ork and Computer Applications , v ol. 117, pp. 42-58, 2018. [27] H. Lamaazi, and N. Benamar , ”RPL enhancement using a ne w objecti v e function based on combined metrics, ” 13th International W ireless Communications and Mobile Computing Conference (IWCMC), 2017. A hybrid objective function with empirical stability awar e to... (Abdelhadi Eloudrhiri Hassani) Evaluation Warning : The document was created with Spire.PDF for Python.