TELK OMNIKA T elecommunication, Computing, Electr onics and Contr ol V ol. 19, No. 1, February 2021, pp. 27 35 ISSN: 1693-6930, accredited First Grade by K emenristekdikti, No: 21/E/KPT/2018 DOI: 10.12928/TELK OMNIKA.v19i1.16138 r 27 In v estigation on ener gy har v esting enabled de vice-to-de vice netw orks in pr esence of co-channel interfer ence Thanh-Luan Nguy en 1 , Dinh-Thuan Do 2 1 F aculty of Electronics T echnology , Industrial Uni v ersity of Ho Chi Minh City (IUH), Ho Chi Minh City , V ietnam 2 W ireless Communications Research Group, F aculty of Electrical and Electronics Engineering, T on Duc Thang Uni v ersity , Ho Chi Minh City , V ietnam Article Inf o Article history: Recei v ed Mar 25, 2020 Re vised Jul 7, 2020 Accepted Sep 24, 2020 K eyw ords: Amplify-and-forw ard Co-channel interference Ener gy harv esting Er godic capacity Outage capacity ABSTRA CT Ener gy harv esting from ambient radio-frequenc y (RF) sources has been a no v el ap- proach for e xtending the lifetime of wireless netw orks. In this paper , a cooperati v e de vice-to-de vice (D2D) system with the aid of ener gy-constrained relay is considered. The relays are assumed to be able to harv est ener gy from information signal and co- channel interference (CCI) signals broadcasted by nearby traditional cellular users and forw ard the source’ s signal to its desired destination (D2D user) utilizing amplify-and- forw ard (AF) relaying protocol. T ime switching protocol (TSR) and po wer splitting protocol (PSR) are proposed to assist ener gy harv esting and information processing at the relay . The proposed a pproaches are applied in a model with three nodes in- cluding the source (D2D user), the relay and the destination (D2D user), the system throughput is in v estig ated in terms of the er godic capacity and the outage capacity , where the analytical results are obtained approximately . Our numerical results v erify the our deri v ations, and also points out the impact of CCI on system performance. Fi- nally , this in v estig ation pro vide fundamental design guidelines for selecting hardw are of ener gy harv esting circuits that s atisfies the requirements of a practical cooperati v e D2D system. This is an open access article under the CC BY -SA license . Corresponding A uthor: Dinh-Thuan Do W ireless Communications Research Group F aculty of Electrical and Electronics Engineering T on Duc Thang Uni v ersity , Ho Chi Minh City , V ietnam Email: dodinhthuan@tdtu.edu.vn 1. INTR ODUCTION Recent adv ances in ener gy harv esting technology has indicated that f ar -field wireless po wer trans fer can also pro vide interesting aspects in wireless communication systems [1–7]. Notice that the source sig- nals carry both ener gy and information at the same time. Hence, a h ypothesis recei v er which can process the information and harv est ener gy simultaneously is required [8, 9]. Ho we v er , such de vice is dif ficult to im- plement since the limitation of circuitry . Furthermore, harv esting protocols for information processing and ener gy harv esting separately ha v e been mentioned in man y scient ific papers [10, 11]. In cooperati v e de vice- to-de vice (D2D) netw orks, an intermediate relay is deplo yed between D2D users to enhance the co v erage rate and throughput of communication systems [12, 13]. F or both time-switching relaying (TSR) and po wer -splitting relaying (PSR) protocols, the co-channel J ournal homepage: http://journal.uad.ac.id/inde x.php/TELK OMNIKA Evaluation Warning : The document was created with Spire.PDF for Python.
28 r ISSN: 1693-6930 interference (CCI) signals act as unnecessary signals, i.e. noises, in information processing phase; on the contrary , supply ener gy for forw arding information signal in the ener gy harv esting phase. More importantly , ener gy harv esting (EH) can be implemented in modern netw orks such as de vice-to-de vice (D2D) netw orks, small cell netw orks as man y recent w orks in [14–18]. The authors in [14] e xamined joint optimization problem to maximize the ener gy ef ficienc y e v aluation related to D2D pairs together with the amount of harv ested po wer by cellular user equipment. W e need more comple x technologies for D2D communications in some w ay in cellular bands [15–18]. The impacts of CCI signals are also considered in [19–24]. Moti v ated by these recent w orks, we continue to fill g ap in the system performance under considering ener gy harv esting protocols in D2D scenario under impact of CCI by traditional cellular users. In this paper , the TSR and PSR recei v er architectures and the corresponding protocols are also adopted. A three-node model of amplify-and-forw ard (AF) relaying is proposed for both protocols, where the source node can only communicate with destination node with the aid of an intermediate ener gy-constrained relay node. 2. SYSTEM MODEL Figure 1 illustrates the system model for the underlay D2D in which tw o de vices, name ly UED S and UED D , participate in the communication through a controlling base station (BS). Assuming hea vily block ed line-of-sight (LOS) path from UED S to UED D , the EH-D2D relay is then deplo yed to ass ist the transmission. In addition, the relay harv ests ener gy from the RF-signal emitted from the UED S and each interferer UEC i , i = 1 ; :::; M . Both the source-to-relay link and relay-to-destination link transmission e xperience quasi-static independent flat Rayleigh f ading with the a v erage g ain E fj h S j 2 g = S and E fj h D j 2 g = D , respecti v ely , in which E fg specifies e xpectation operator . It is pre viously stated that the CUEs are the cross-mode interferers and can be treated as CCIs at the relay in the proposed model. The CCIs deteri orate the system performance b ut surprisingly aid the ener gy harv esting process at the relay . R e l a y 1 U EC M U EC B S i n t e r f e r e n c e I n f o r m a t i o n   l i n k e n e r g y   h a r v e s t i n g S U ED D U ED c o n t r o l   l i n k Figure 1. System model of D2D netw ork under impact of the co-channel interferences 3. TIME SWITCHING-B ASED RELA YING PR O T OCOL Complying with the TSR-assisted relay architecture, after recei ving the RF-signal broadcasted by UED S , the relay passes the signal to the ener gy harv esting recei v er for a duration of r T block time and then to the information recei v er for that of (1 r ) T = 2 block time [12]. Accordingly , the relay performs ener gy harv esting process and then information process, respecti v ely . Under the presences of the UECs, i.e., the cellular users, the recei v ed signal at node R is modeled as y R ( t ) = h S s ( t ) + P M i =1 l i s i ( t ) ( t ) + ~ n R [ a ] ( t ) ; (1) where s ( t ) is the information signal with po wer of P S , E fj s ( t ) j 2 g , h S 2 C is the comple x channel f actor from UED S to R , s i ( t ) specifies the i -th interference signal with the po wer of P i , E fj s i ( t ) j 2 g , l i 2 C TELK OMNIKA T elecommun Comput El Control, V ol. 19, No. 1, February 2021 : 27 – 35 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control r 29 denotes comple x channel f actor from UEC i to R , the number of CCIs is denoted by M and ~ n R [ a ] ( t ) is the corrupted narro w band Gaussian noise observ ed at the recei ving antenna. Subsequently , the recei v ed signal is con v erted to a basebanded comple x signal after the do wn con v ersion process, which results the sampled baseband signal y R ( k ) gi v en by y R ( k ) = h S s ( k ) + P M i =1 l i s i ( k ) + n R [ a ] ( k ) + n R [ c ] ( k ) | {z } , n TSR R ( k ) ; (2) where s i ( k ) and s ( k ) denote the signals induced afte r sampling the i th interfererence signal and the source signal, respecti v ely , n R [ a ] ( k ) is the baseband additi v e white Gaussian nois e (A WGN) at the recei ving antenna, and n R [ c ] ( k ) is the sampled A WGN induced after being con v erted to baseband, s i ( k ) and s ( k ) ha v e zero means and v ariance of N R [ a ] and N R [ c ] , respecti v ely , and n TSR R ( k ) is defined as the total Gaussian noise at node R introduced from adopting the TSR architecture. The relay then utilizes r T block time to harv est ener gy from the recei v ed signals. Hence, the ener gy harv ested at the relay is gi v en by E h = e P S j h S j 2 + P M i =1 P i j l i j 2 r T ; (3) where e , with 0 e 1 represents the ef ficienc y of the ener gy harv ester , its v alue depends on the manu- f acturer . Assuming that the relay fully absorbs the harv ested ener gy to forw ard the recei v ed signal to the other D2D user , i.e., the UED D node. Accordingly , the transmit po wer at the relay can be obtained as P R = E h (1 r ) T = 2 = 2 r e 1 r P S j h S j 2 + P M i =1 P i j l i j 2 ; (4) As a priority , the relay amplifies the recei v ed signal with a g ain f actor G and then forw ards y R ( k ) to UED D . The g ain f actor G is gi v en by G = p P R q P S j h S j 2 + P M i =1 P i j l i j 2 + N TSR R ; (5) where N TSR R = N R [ a ] + N R [ c ] denotes the total Gaussian noise po wer observ ed at the relay under TSR protocol. Secondly , the recei v ed signal at UED D after being sampled, y D ( k ) is gi v en by y D ( k ) = h S s ( k ) h D G + P M i =1 l i s i ( k ) + n TSR R ( k ) h D G + n D ( k ) : (6) 4. PO WER SPLITTING-B ASED RELA YING PR O T OCOL Let P be the recei v ed po wer at R and , 0 1 , denote the ener gy harv esting ratio of the PSR protocol, thus P specifies the amount of po wer inputted into the ener gy harv ester . The remaining po wer , i.e., (1 ) P , inputs the information transmission to forw ard the UED S ’ s signal to UED D . Under the presences of cross-mode CCIs, the recei v ed signal observ ed at the relay antenna is y R ( t ) = h S s ( t ) + P M i =1 l i s i ( t ) ( t ) + ~ n R [ a ] ( t ) : (7) The sampled baseband signal at the relay node, y R ( k ) , is gi v en by y R ( k ) = p (1 ) h S s ( k ) + p 1 M P i =1 l i s i ( k ) + p 1 n R [ a ] ( k ) + n R [ c ] ( k ) | {z } = n PSR R ( k ) ; (8) in which n PSR R ( k ) denotes the total Gaussian noise introduced by the PSR-assisted relay . At the relay , an amount of recei v ed signal, is adopted for ener gy harv esting. Hence, the ener gy har - v ested at the node R is E h = e P S j h S j 2 + P M i =1 P i j l i j 2 r T : (9) In vestigation on ener gy harvesting enable de vice-to-de vice networks... (Thanh-Luan Nguyen) Evaluation Warning : The document was created with Spire.PDF for Python.
30 r ISSN: 1693-6930 Assume that the harv ested ener gy is perfectly consumed by the relay . As a result, the transmit po wer at the node R is e xpressed as P R = E H T = 2 = e P S j h S j 2 + P M i =1 P i j l i j 2 : (10) Similarly , the relay firstly amplifies the recei v ed signal with the g ain f actor , G , which can be e xpressed as G = p P R q (1 ) P S j h S j 2 + (1 ) P M i =1 P i j l i j 2 + N PSR R ; (11) where N PSR R = (1 ) N R [ a ] + N R [ c ] . Accordingly , the recei v ed signal after the being sampled at the desti- nation node, y D ( k ) , is gi v en by y D ( k ) = h S s ( k ) h D G + P M i =1 l i s i ( k ) + n PSR R ( k ) h D G + n D ( k ) : (12) 5. GENERAL AN AL YSIS W e find that the TSR and PS R protocols ha v e similar mechanisms, deri ving a general form for the signal-to-noise-plus-interence ratio (SINR) can be feasible. In order to deri v e an unified result, we define n Y R ( k ) as the total Gaussian noise at the relay with v ariance of N Y R ( k ) for the Y 2 f TSR ; PSR g protocol, the e xpressions of n TSR R ( k ) and n PSR R ( k ) are defined in the pre vious section. Therefore, the unified form of the achie v able end-to-end SINR under the adoption of the protocol Y , denoted by Y g en , can be e xpressed as Y g en = 1 I N F + 1 Y g 1 + N Y R 1 + I N F + N Y R ; (13) in which 1 , P S j h S j 2 , I N F , P M i =1 P i j l i j 2 , TSR g , 2 r e 1 r j h D j 2 N D and PSR g , e j h D j 2 N D . Hereafter , we define SNR , P S S =N D as the a v erage signal-to-noise ratio (SNR). 5.1. Outage pr obability In this paper , considering the whole system, an outage e v ent occurs whene v er Y g en drops belo w an acceptable threshold, th (dB). Accordingly , the outage probability is defined as P Y out = Pr Y g en < th = F Y g en ( th ) : (14) It is not tractable to deri v e the e xact outage probability in closed-form from (14). Hence, to simplify the calculation, we apply the high SNR approximation. At high SNR, where the UED S transmits with relati v ely high po wer le v el, the term ” N Y R = ( 1 + I N F ) ” in the denominator of (13) ca n be ne gligible. As a result, the approximated SINR at the relay is gi v en by Y g en 1 I N F + 1 Y g + N Y R : (15) Therefore, the approximated outage probability , P Y out , in (14) is then re written as F Y g en ( th ) Z 1 0 Z 1 0 Pr 1 < th z + 1 y + N Y R  f I N F ( y ) f Y g ( z ) dy dz : (16) Note that TSR g and PSR g are random v ariables ha ving e xponential distrib ution. Subsequently , the probability density function (PDF) of Y g is gi v en by f Y g ( z ) , 1 Y g exp z Y g ; z > 0 ; (17) TELK OMNIKA T elecommun Comput El Control, V ol. 19, No. 1, February 2021 : 27 – 35 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control r 31 where TSR g , 2 r e 1 r D N D and PSR g = e D N D . In addition, the CDF of 1 is F 1 ( x ) = 1 exp n x 1 o , where 1 , P S S and the PDF of I N F is gi v en by f I N F ( y ) = v ( D ) X i =1 i ( D ) X j =1 i;j ( D ) j h i i ( j 1)! y j 1 exp y h i i ; y > 0 ; (18) in which D = diag ( 1 ; 2 ; :::; M ) specifies a diagonal matrix with the eigen v alues of i = P i N D i , ( D ) denotes the number of distinct diagonal elements, h 1 i > h 2 i > ::: > h ( D ) i are the distinct diagonal elements in descending order , i ( D ) is the multiplicity of h i i , and i;j ( D ) is the ( i; j ) -th characteristic coef ficient of the matrix D [24]. Substituting (18) and (17) into (16), t he approximated P Y out e xpressed in the inte gral-form is gi v en by P Y out 1 1 Y g exp N Y R th 1 v ( D ) X i =1 i ( D ) X j =1 i;j ( D ) ( j 1)! 1 j h i i Z 1 0 y j 1 exp th 1 + 1 h i i  dy | {z } I 1 Z 1 0 exp th 1 z Y g dz | {z } I 2 : (19) Subsequently , t he abo v e inte grals ( I 1 and I 2 ) can be deri v ed in closed-form with the help of [25, (2.3.3.1)] and [25, (2.3.16.1)] as I 1 = 1 Z 0 y j 1 exp th 1 + 1 h i i y dy = ( j ) th 1 + 1 h i i j ; (20) I 2 = 1 Z 0 exp th 1 z z Y g dz = 2   th Y g 1 ! 1 = 2 K 1   2 r th 1 Y g ! ; (21) respecti v ely , where K v ( ) denotes the v -th order modified Bessel function of the second kind and ( x ) specifies the Gamma functi on. Hence, the outage probability P Y out can be approximated by using (21), (20) and (19) which then results the follo wing equation after some algebraic steps P Y out 1 s 4 th Y g 1 K 1   s 4 th Y g 1 ! exp N Y R th 1 v ( D ) X i =1 i ( D ) X j =1 i;j ( D ) 1 + h i i th 1 j : (22) When the interfering signals are statistically independent and identically distrib uted (i.i.d.), i.e., i = ; i = 1 ; 2 ; :::; M ; then ( D ) = 1 and i ( D ) = M , the outage probability , P Y out , is then reduced to P Y out =1 s 4 th Y g 1 K 1   s 4 th Y g 1 ! exp N Y R th 1 1 +  th 1 M : (23) 5.2. Outage capacity and achie v able thr oughput The outage capacity for the AF cooperati v e D2D system under consideration is gi v en by C Y O = 1 P Y out log 2 (1 + th ) (24) The achie v able throughput is defined in terms of ef fecti v e transmission block time, which is the block time utilized for relay-to-destination transmission. According to [24], the achie v able throughput of a cooperati v e system is gi v en by Y O = 8 > < > : (1 r ) T = 2 T C TSR O ; Y TSR T = 2 T C PSR O ; Y PSR = (1 r ) C TSR O = 2 ; Y TSR C PSR O = 2 ; Y PSR (25) In vestigation on ener gy harvesting enable de vice-to-de vice networks... (Thanh-Luan Nguyen) Evaluation Warning : The document was created with Spire.PDF for Python.
32 r ISSN: 1693-6930 5.3. Er godic capacity and achie v able thr oughput In this subsection, the throughput achie v ed by e v aluating the Er godic capacity in t he unit of bits/Hz is deri v ed as the third important metrics to e v aluate the system performance. In the AF-cooperati v e D2D communication, using Y g en in (8), the recei v ed SINR at the relay , C E is gi v en by C Y E = E 1 2 log 2 (1 + Y g en ) = Z 1 0 log 2 (1 + $ ) f Y g en ( $ ) d$ ; (26) where f Y g en ( $ ) stands for the PDF of the random v ariable Y g en . Applying the inte gration by parts for the inte gral in (32), the abo v e e xpression becomes C Y E = h log 2 (1 + $ )( F Y g en ( $ ) 1) i 1 0 1 ln 2 Z 1 0 1 1 + $ [ F Y g en ( $ ) 1] d$ (27) = 1 ln 2 Z 1 0 1 1 + $ (1 F Y g en ( $ )) d$ ; (28) where f f ( x ) g b a = f ( b ) f ( a ) . Similarly as in 5.2, the throughput at the destination depends only on the ef fecti v e transmission time, (1 r ) T = 2 for TSR protocol and T = 2 for PSR protocol, and can be e xpress ed as Y E = (1 r ) C TSR O = 2 ; Y TSR C PSR O = 2 ; Y PSR (29) 6. NUMERICAL RESUL TS In this section, the simulation results and the approximated analytical results are deri v ed. T o e v aluate the ef fects of the interference on the system throughput we define SIR = P S S P M i =1 P i i as the a v erage signal-to- interference ratio. The v ariances are assumed to the identical and k ept fix ed, that is N D = 1 , N R [ a ] = N R [ c ] = 1 and the SINR threshold, is set to 8 dB unless stated otherwise. In Figures 2-5, we assume a single interferer ( M = 1 ). In addition, the ener gy con v ersion ef ficienc y is set to 1 ( e = 1 ). Importantly , in order to e v aluate the impact of the interference on the throughput, we define k k = f I N F g as the normalized po wer distri b ution, where = h 1 i ; h 2 i ; :::; h M i . 0 0.2 0.4 0.6 0.8 1 0 0.5 1 1.5 ξ r Throughput (bit/s/Hz) Y = TSR     Analytical SIR = 10 dB Analytical SIR = 20 dB Simulation τ E PSR τ O PSR Figure 2. Throughput as a function of the ener gy harv esting ratio with tw o v alues of the a v erage SIR, the a v erage SNR is set to 20 dB TELK OMNIKA T elecommun Comput El Control, V ol. 19, No. 1, February 2021 : 27 – 35 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control r 33 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 β Throughput (bit/s/Hz) Y = PSR     Analytical SIR = 10 dB Analytical SIR = 20 dB Simulation τ E PSR τ O PSR Figure 3. Throughput TSR E , TSR O , PSR E and PSR O as a function of the a v erage SNR, in which SIR = 10 dB Figure 2 sho ws throughput TSR E and TSR O v ersus the ener gy harv esting r atio r for dif ferent v alues of a v erage SIR where SNR is set to 20 dB. The simulation results of TSR E are e v aluated, where C Y E and Y g en are obtained. The solid curv es are the corresponding approximated analytical results of TSR E which deri v ed in (33). The dashed curv es are the corresponding approximated analytical results of TSR O deri v ed. It is observ ed in Figure 3 that the throughput increases as the ener gy harv esting ratio, r increases from 0 to some optimal v alue b ut later as r continues increasing, the relay w astes more time on ener gy harv esting rather than information transmission resulting that the throughput of the system starts dropping do wn from its maximum v alue. 0 5 10 15 20 25 30 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 SNR (dB) optimal  ξ r  (Y = TSR)     τ O TSR  SIR = 10 dB τ O TSR  SIR = 20 dB τ E TSR  SIR = 10 dB τ E TSR  SIR = 20 dB Figure 4. Optimal r v ersus the a v erage SNR for dif ferent v alues of the a v erage SIR 0 5 10 15 20 25 30 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 SNR (dB) optimal  β Y = PSR     τ O PSR  SIR = 10 dB τ O PSR  SIR = 20 dB τ E PSR  SIR = 10 dB τ E PSR  SIR = 20 dB Figure 5. Optimal v ersus the a v erage SNR for dif ferent v alues of the a v erage SIR Figure 4 and Figure 5 sho ws the optimal r and optimal , respecti v ely , the corresponding optimal throughputs where the a v erage SIR is set to 10 dB are illustrated in Figure 3. It is seen that, in TSR protocol, as the a v erage SNR increases the optimal r decreases. This implies that the system performance can ef fecti v ely be enhanced and the time spent for ener gy harv esting ( r T ) can also be reduced by increasing the transmit po wer of the source, P S . In addition, the optimal ratios to achie v e the optimal throughput TSR E increases as the a v erage SIR increases. Ho we v er , the similar trend does not apply to optimal TSR O , in this case, the optimal r does not change as the a v erage SIR increases. The con v erse happened in PSR protocol, where the optimal increases as the a v erage SNR increases. Furthermore, the optimal to achie v e the optimal throughput PSR E decreases as the a v erage SIR increases. This implies that, in PSR protocol, more po wer is used for ener gy harv esting as the a v erage SNR increases and less po wer can be needed if there is an increasing in the po wer of the interference. The impact of CCI po wer distrib ution to the system throughput is illustrated in Figure 6 and Figure 7 for system with TSR and PSR protocol, respecti v ely . The ener gy harv esting ratio r and are set to 0.2 and 0.8, respecti v ely . Though the po wer distrib utions are dif ferent, e.g. k 1 k = (1 : 0 ; 0 ; 0 ; 0) ; k 2 k = In vestigation on ener gy harvesting enable de vice-to-de vice networks... (Thanh-Luan Nguyen) Evaluation Warning : The document was created with Spire.PDF for Python.
34 r ISSN: 1693-6930 (0 : 5 ; 0 : 5 ; 0 ; 0 ; 0) and k 1 k = (1 : 0 ; 0 ; 0 ; 0) , b ut the total po wer of interferers remains the same v alue. It is observ ed that, the ac h i e v able throughput decreases as the normalized po wer distrib ution are changed from k 1 k to k 2 k and from k 2 k to k 3 k . 0 5 10 15 20 25 30 0 0.2 0.4 0.6 0.8 1 1.2 1.4 SNR (dB) Throughput (bit/s/Hz) Y = TSR     Ana.  µ  = (1.0 0 0 0) Ana.  µ  = (0.5 0.5 0 0) Ana.  µ  = (0.25 0.25 0.25 0.25) Simulation τ E TSR τ O TSR Figure 6. Throughput TSR E and TSR O v ersus the a v erage SNR under dif ferent CCI po wer distrib ution where the a v erage SIR is set to 10 dB 0 5 10 15 20 25 30 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 SNR (dB) Throughput (bit/s/Hz) Y = PSR     Ana.  µ  = (1.0 0 0 0) Ana.  µ  = (0.5 0.5 0 0) Ana.  µ  = (0.25 0.25 0.25 0.25) Simulation τ E PSR τ O PSR Figure 7. Throughput PSR E and PSR O v ersus the a v erage SNR under dif ferent CCI po wer distrib ution where the a v erage SIR is set to 10 dB 7. CONCLUSION In this paper , an AF cooperati v e D2D system w as proposed where the EH-assisted relay is af fected by co-channel interferences (CCI) from the CUEs. The ener gy-constrained relay absorbs the harv ested ener gy from the recei v ed source signal and CCI signals to support the transmission between D2D users. The system performance can be deteriorated if the po wer of t he CCI signals increases. One can ef fecti v ely increase the system throughput by increasing the a v erage SNR, this can be achie v ed by increasi ng the transmit po wer of D2D users. Lastly , dif ferent po wer distrib ution can also af fect to the system throughput. REFERENCES [1] T .-L. Nguyen, D.-T . Do, ”Po wer Allocation Schemes for W ireless Po wered NOMA Systems with Imperfect CSI: System model and performance analysis, ” International J ournal of Communication Systems , v ol. 31, no. 15, 2018. [2] D.-T . Do, et al. , “W ireless po wer transfer enabled NOMA relay systems: tw o SIC modes and performance e v alua- tion, ” TELK OMNIKA T elecommunication Computing Electr onics and Contr ol , v ol. 17, no.6, pp. 2697-2703, 2019. [3] D.-T . Do and C. B. Le, ”Exploiting Outage Performance of W ireless Po wered NOMA, ” TELK OMNIKA T elecom mu- nication Computing Electr onics and Contr ol , v ol. 16, no. 5, pp. 1907-1917, 2018. [4] D.-T . Do, M.-S. V an Nguyen, T . A. Hoang, B. M. Lee, “ Exploiting Joint Base Station Equipped Multiple Antenna and Full-Duple x D2D Users in Po wer Domain Di vision Based Multiple Access Netw orks, ” Sensor s ,v ol. 19, no. 11, pp. 2475-2494, 2019. [5] D.-T . Do and C. B. Le, “ Application of NOMA in W ireless System with W ireless Po wer T ransfer Scheme: Outage and Er godic Capacity Performance Analysis, ” Sensor s , v ol. 18, no. 10, pp. 3501-3517, 2018. [6] D.-T . Do, M.-S. V . Nguyen, “Outage probability and e r godic capacity analysis of uplink NOMA cellular netw ork with and without interference from D2D pair , ” Physical Communication , v ol. 37, 2019. [7] R. Rajesh, V . Sharma, and P . V isw anath, ”Information capacity of ener gy harv esting sensor nodes, ” Pr oc. 2011 IEEE Int. Symp. Inf . Theory , pp. 2363-2367, July 2011. [8] L. R. V arshne y , ”T ransporting information and ener gy simultaneously , ” Pr oc. 2008 IEEE Int. Symp. Inf . Theory , pp. 1612-1616, July 2008. [9] P . Gro v er , A. Sahai, ”Shannon meets T esla: W ireless information and po wer transfer , ” Pr oc. 2010 IEEE Int. Symp. Inf . Theory , pp. 2363-2367, July 2010. [10] R. Zhang and C. K. Ho, ”MIMO broadcasting for simultaneous wireless information and po wer transfer , ” IEEE T r ans. W ir el. Commun. , v ol. 12, no. 5, pp. 1989-2001, May 2013. [11] B. Medepally and N. B. Mehta, ”V oluntary ener gy harv esting relays and selection in cooperati v e wireless netw orks, ” IEEE T r ans. W ir el. Commun. , v ol. 9, no. 11, pp. 3543-3553, No v ember 2010. [12] A. A. Nasir , X. Zhou, S. Durrani, and R. A. K ennedy , ”Relaying protocols for wireless ener gy harv esting and infor - mation processing, ” IEEE T r ans. W ir el. Commun. , v ol. 7, no. 12, pp. 3622-3636, No v . 2013. TELK OMNIKA T elecommun Comput El Control, V ol. 19, No. 1, February 2021 : 27 – 35 Evaluation Warning : The document was created with Spire.PDF for Python.
TELK OMNIKA T elecommun Comput El Control r 35 [13] D. Y ang, ”W ireless Information and Po wer T ransfer: Optimal po wer control in one-w ay and tw o-w ay relay system, ” W ir eless P er sonal Commun. , v ol. 84, no. 1, pp. 1-14, June 2015. [14] Z. Zhou, C. Gao, C. Xu, T . Chen, D. Zhang and S. M umtaz, ”Ener gy-Ef ficient Stable Matching for Resource Alloca- tion in Ener gy Harv esting-Based De vice-to-De vice C ommunications, ” IEEE Access , v ol. 5, pp. 15184-15196, May 2017. [15] S. Lee, R. Zhang, and K. Huang, ”Opportunis tic wireless ener gy harv esting in cogniti v e radio netw orks, ” IEEE T r ans. W ir eless Commun. , v ol. 12, no. 9, pp. 4788-4799, September 2013. [16] S. Gupta, R. Zhang and L. Hanzo, ”Ener gy Harv esting Aided De vice-to-De vice Communication Underlaying the Cellular Do wnlink, ” IEEE Access , v ol. 5, pp. 7405-7413, 2017. [17] L. Jiang et al. , ”Social-a w are ener gy harv esting de vice-to-de vice communications in 5G netw orks, ” IEEE W ir el. Commun. , v ol. 23, no. 4, pp. 20-27, 2016. [18] M. L. K u, J. W . Lai, ”Joint Beamforming and Resource Allocation for W ireless-Po wered De vice-to-De vice Commu- nications in Cellular Netw orks, ” IEEE T r ans. W ir el. Commun. , v ol. 16, no. 11, pp. 7290-7304, 2017. [19] D. W . K. Ng, R. Schober , Spectral ef ficient optimization in OFDM syste ms with wire less information and po wer transfer , Pr oc. 21st Eur . Signal Pr ocess. Conf . , pp. 1-5, September 2013. [20] G. Y ang, C. K. Ho, and Y . L. Guan, Dynamic resource allocation for multiple-antenna wireless po wer transfer , IEEE T r ans. Signal Pr ocess. , v ol. 62, no. 14, pp. 3565-3577, Jul. 2014 [21] D.-T . Do, H. S. Nguyen, ”A T ractable Approach to Analyze the Ener gy-A w are T w o-w ay Relaying Netw orks in Presence of Co-channel Interference, ” EURASIP J ournal on W ir eless Communications and Networking , v ol. 2016, no. 271, 2016. [22] T .-L. Nguyen, D.-T . Do, ”A ne w look at AF tw o-w ay relaying netw orks: ener gy harv esting architecture and impact of co-channel interference”, Annals of T elecommunications , v ol. 72, no. 11, pp. 669-678, 2017. [23] J. A. Hussein, S. Boussakta and S. S. Ikki, ”Performance Study of a UCRN Ov er Nakag ami- m F ading Channels in the Presence of CCI, ” IEEE T r ans. on Co gnitive Communications and Networking , v ol. 3, no. 4, pp. 752-765, December 2017. [24] Y . Gu and S. Aissa, ”Interference aided ener gy harv esting in decode-and forw ard relaying systems, ” in Proc. IEEE Int. Conf . Commun. , pp. 5378-5382, Jun. 2014. [25] A. P . Prudnik o v , Y . A. Brychk o v , and O. I. Mariche v , Inte grals and Series, v ols. 1-2. Ne w Y ork, Gor don and Br eac h Science Publisher s , 1986. In vestigation on ener gy harvesting enable de vice-to-de vice networks... (Thanh-Luan Nguyen) Evaluation Warning : The document was created with Spire.PDF for Python.