Analysisof power consumption and throughput using joint-bit ADCs in massive MIMO AbstractThe sensibleimplementation of massive multiple-inputmultiple-output (MIMO) inside the fifthgeneration (5G)wirelesscommunication structures is challenging due to its high hardware cost and powerconsumption. This paperinvestigated the uplink of massive multi-input multi-output (MIMO) systems witha joint analog-to-digital converter (ADC) receiver structure, in which someantennas are equipped with expensive full-resolution ADCs , a few antennas areequipped with intermediate-resolution ADCs and some antennas with much lessexpensive low-resolution ADCs.
A closed-shape approximation of the achievablesum rate with the maximum-ratio combining (MRC) detector is derived . weperform a parametric energy efficiency analysis of MIMO uplink for the wholebase station receiver system with joint ADC resolutions. The analysis showsthat, for a wide variety of system parameters, ADCs with joint- bitresolutions are most beneficial inenergy efficiency sense, and that the usage of very low bit resolutionsoutcomes in degradation of energy efficiency. We also achieve the power scalinglaw that the transmit power of every consumer can be scaled down proportionallyto the inverse of the wide variety of base station (BS) antennas for perfectCSI. Furthermore, we show the trade-off among the achievable rate and theenergy efficiency with respect to key system parameters, such as the quantizationbits, wide variety of BS antennas and consumer transmit power.
in the end,results are provided to show that the joint-ADC architecture can obtain ahigher energy-rate trade-off in comparison with high-resolution and low-resolution ADC architectures at a significantly lower hardware cost.1. IntroductionThe research on upcoming technology of wireless communication fifth generation(5G) has been undertaken to achieve 1000time more data rate then the present technology (4G) 1. To meet this requirement, Massive multiple input multipleoutput (MIMO) is key technology. By using the hundred of the antenna at thebase station(BS) to achieve batter datarate and energy efficiency.
Differentsignal processing scheme as maximum ratio combining(MRC) and zero forcing(ZF) arereduced the inter-user interference 2. Large number of antenna at (BS) havemore power consumption and complexity. It has costly hardwire implementation.
we cannot provide a specific radio frequency(RF) chain to each antenna, different technique are proposed in recent past toovercome these problems. To reduce the(RF) chain different analog and digital approaches are used as like switches,phase shifters and lenses 4-6 of 2.One basic problem is the powerconsumption related with theanalog-to-digitalconversion(ADC). In conventional (MIMO) receiver architecture,the analog-to-digitalconvertors (ADCs) are supposedto have high resolution. In channels with largerbandwidths, the correlated samplingrate of the ADCs scales up.Unlikely, high-resolution ADCs arecostly and power-hungry forportable devices 3.
For example,in an ideal x-bit ADC withflash design, there are2x-1 comparatorsand therefore the power consumption increaseexponentially with theresolution 4-5. Furthermore,The most direct answers to thepower assumption problem are to reduce the sampling rateand/or the quantizationresolution of ADCs.6 There is a massiveMIMOarchitecture at the receiver to reducethe high resolutionof the ADCs to low resolution while keeping the number of RF chainsunchanged. However,the many nonlinear distortion caused by low-resolutionADCsnecessarily causes different problems,along with capacity loss athigh SNR ,high pilot overhead forchannel estimationand complex precoder design 2.Related WorkIn Massive MIMO using the large numberantenna can enhance the capacity and the energy efficiency. But powerconsumption and the hardwire cost is also increased.
In (RF) chain analog todigital convertor (ADC) is most power hungry 7. The power consumption of ADC is increased exponentially when the quantization bits increase.8. Toovercome this problem low resolutionADCs get attention in last few years. The 1-bit ADCis most attractive due to the lowerhardware complexity. In this scenario,the in-phase and quadraturecomponents of thecontinuous-valued received signals aredistinctly quantized usingsimple zero-threshold comparators;so that, there is no use for anautomatic gain controller(AGC) 9. while all ADCs have low resolution,time-frequency syn-chronization and channelestimation are difficult andrequire immoderate overhead 10In 9a mixed-ADCreceiver structure was proposedfor massive MIMO sys-tems, wherein a fraction of theADCs has complete-resolutionto promote system overallperformance, and the others have low-resolution in consideration ofthe hardware price and powerconsumption.
In 10 the mixed-ADC receiverarchitecturewas developed for multiusermassive MIMO systems, inwhich a circle of relatives ofBayes detectors were developed by means of con-ducting Bayesian estimate of theconsumer indicators with preferred approximate message passing algorithm. byway of analyzing the general mutualinformation (a beneficial lower bound on the channel capability with mismatcheddecoding)11. in12the evaluation of generalizedmutualinformation has been emphasizedto illustratethat the mixed-ADC architectureis capable of obtain a massivefraction of the channelcapability of best ADC architecturesover frequency-flat channels. Theauthors derived a closed-formapproximation of themanageable rate of mixed-ADCmassive MIMO structures withthe MRC detector.
These initial works validate the deserves ofthe mixed-ADC structure formassive MIMO systems 13.System model In this paper, we suppose R number ofantennas at BS station of single cell Massive MIMO system and U number of userswith single antenna 14. Fig. 1.System model of joint-ADC massive MIMO receivers with high-resolution ADCs, Intermediate-resolutionADCs and low-resolution ADCs. The power consumption and the hardwire costis linearly with number of antenna’s. In fig.1 we introduce a joint MassiveMIMO architecture rather than using high resolution ADCs.
In theJoint-ADC architecture,there are three different groups of RF chains. One group is connected with highresolution ADCs, second with intermediate resolution and third with lowresolution ADCs.the power consumption of the receiver isprecipitated by means of the RF chain and the baseband processor. The localoscilator (LO) that is used with a mixer to alternate the frequency of a signalis shared by using all RF chains.
furthermore, each RF chain is composed of alow-noise amplifier (LNA), and the LO buffer. The in-phase and quadraturecircuit consist of a mixer which includes a quadrature hybrid coupler, an ADC,and an automatic gain control (AGC), respectively, observe that the AGC is usedto amplify the obtained signal to use the overall dynamic range of the ADC. theusage of one-bit ADCs can simplify the RF chain, e.g., an AGC isn’timportant due to the fact only the signal ofthe input signal is output 2 Receiver has W antennaelements. However, only H antennasare connected to the full-resolution ADCs, I antennas are connectedto the intermediate-resolutionADCs and L antennas are connected to thelow-resolution ADCs. Let G be the W×U channel matrixfrom the users to the BS.
The channel matrix is modeledas 1 (1)Where contains the fast-fadingcoefficients, whose entries that are independent For ease of expression, wedenote G = T , where is the × u channel matrixfrom the users to the BS antennas withfull-resolution ADCs, is the × u channel matrixfrom the users to the BS antennas withfull-resolution ADCs and is the ×u channel matrixfrom the users to the BS antennas with low-resolution ADCs.Let pu be the averagetransmitted power of each user and x be the u×1 vector ofinformationsymbols. The received signals atthe full-resolution ADCs can be given bySignal at the receiver is given as (2) (3) Signal at AGC is defined as (4)Amplitude AGC gains pi can beconveniently collected in adiagonal matrix (5) The q represented by thequantization noise ,where b is represent quantization bits.
Signal after ADCs of three different groups is received as = (6) = (7) = (8) The overall signal received after passingthrough maximum ratio combining (RMC) is given by (9)For prefect channel state information CSIand signal received after MRC at BS R= By (9) itcan be expressed as R= x+ ( ) + ( (10) The receivedsignal-to-interference-plus-noise ratio (SINR) of the kth stream for MRCis given by expression: (11) ¥= + + (12)Theachievable sum rate is expressed as (13) By using eq (11 ) Spectral efficiency is derived asi (14)Powerconsumption model First consider a general power consumptionmodelof the joint-ADC architecture. Theenergy efficiency (EE) isdefined as bit / Joule, (15)where denotes the sum rate, W denotes the transmissionbandwidth assumed to be 1 GHz and Ptotal is the totalconsumed (15)Where PLO,PLNA,PH,PM,PAGC,PH ADC,PLADC,PBBdenotethe power consumptionof local oscillator, mixer, AGC, high-resolution, intermediate-resolution and low resolution ADCs respectively. In addition, f isthe flag associated to the quantization bit oflow-resolution ADCsand is given by (16) Furthermore, the power consumed in theADCs can beexpressed in terms of the number ofquantization bits as , (17) where fs is the Nyquistsampling rate, b is the number ofquantization bits, and is Walden’s figure-of-merit forevaluating the power efficiency withADC’s resolution andspeed.
With state-of-the-arttechnology, the power consumptionvalue of FOMWat 1GHzbandwidth is = 5 15 fJ/conversion-step 15.Results: Fig:2 powerconsumption of joint-ADC massive MIMOsystems over Rician fading channelsfor K = 10 dB, and pu = 10 dB. Fig:3 Energy efficiency of joint-ADC massive MIMOsystems against different ADC quantization bits for K =10dB, and pu= ?10 dB.The first insight obtained from Fig.2 isthat the joint- ADC structure requires more power than the only with purelow-resolution ADCs, however with a exceedingly massive increase within theachievable rate for small quantization bits. in addition, the distance of power consumption decreaseswith the increasing ADC’s resolution. For the joint-ADC structure, theattainable rate significantly will increase while the quantization bit willincrease from 1 to 4, with a negligible increase in the receiver powerconsumption.
but, the power consumption will increase significantly whilst thequantization bit going from 4 to 11 bits. This means the joint-ADC architecturecanobtain a better trade-off between thepower consumption andachievable rate via the usage ofdifferent bits.It’s far clean from Fig. 3 that thepremier ADC quantization bit is encouraged by using both the part of high-resolutionADCs and the Rician k-element. For the joint-ADC architecture, the energyefficiency simplest increases up to a certain wide variety of quantizationbits, while it decreases at higher ADC quantization bits.
this is because ofthe reality that the plausible charge is a sub-linear increasing feature of thequantization bits, while the power consumption of ADCs will increaseexponentially with b. Fig:4 Sum achievable rate ofjoint-ADC massive MIMO systems over Rician fading channels against the ADCquantization bits and pu = 10 dB. Fig: 5 Energy efficiency of joint-ADC massive MIMO systems overRician fading channels againstdifferent numbers of antennas for K=10 dB and pu= 0 dB.
Fig. 4 gives the simulated and sumachievablerate of joint-ADC massive MIMOstructures for diversenumbers of ADC quantization bits. Withoutlack of generality,we count on that all customers havethe identical Rician k-aspect asKn = k.
The simulation outcomesextensively coincide with thederived outcomes for all ADCquantization bits. We also locatefrom Fig. 4 that better sum charge canbe performed by way of usingextra high-resolution ADCs. moreover, theachievablerate increases with the range ofquantization bits (b), andconverges to a constrained manageablerate with high-resolution ADCs. For Rayleigh fading channels (k= zero), the joint-ADC structure can achieve the same sumviable charge byusing 5 bits, at the same time as forRician fading channels (k = 10),extra ADC quantization bits arewished, that’s in settlementwith statement 1. This is genuine dueto the fact the quantization noise is large with greater received power in LoSdominating scenarios. In Fig.
5, we analyze the impact ofthe quantity of receiveantennas on the strength efficiency.because of the truth that each antenna is attached with one RF chain, it may bewastefulto apply greater antennas at the BS.The Energy Efficiency(EE) oflow-resolution ADCs is biggest amongothers. indeed, if theBS has simplest around 30 antennas,the joint-ADC structureachieves its most appropriate powerperformance for the deciding on system. conclusion:in this paper, the overall performanceof joint-ADC massive MIMO systems over Ricin fading channels is investigated.We derive closed-shape approximate expressions for the achievable rate formassive-antenna limit.
The cases of perfect are incorporated in our analysis.With similar hardware cost, the joint-ADC structure can acquire larger sum ratethan the ideal-ADC architecture. we conclude that realistic massive MIMO canachieve a considerable performance with small power consumption by adopting thejoint-ADC structure for 5G. We conclude that the joint-ADCstructure can carry most of the favored overall performance enjoyed throughmassive MIMO receivers with complete perfect-resolution ADCs. A parameterizedevaluation of energy efficiency within the uplink of a massive MIMO system withvarying ADC bit resolutions at the base station has been carried out.
systemsetup and models have been selected with the purpose of being close topractical system implementations. consequences suggest that using ADCs withvery low bit resolutions is not an premiere technique from energy efficiencyfactor of view, except for highly unique system architectures. instead, for awide form of systems, ADCs with joint bit resolutions (1 – 11 bits) are shownto maximize system energy efficiency. Further, the joint-ADC architecture may have alarge working area with few high-decision ADCs. more energy efficiency may beexecuted when operating over stronger LoS situations.Reference: 1M. Agiwal, A.
Roy, and N. Saxena, “Next generation 5G wireless networks: Acomprehensive survey,” IEEE Commun. Surveys Tuts., Volume: 18 , Issue: 3, thirdquarter 2016 2 Jiayi Zhang,LinglongDai, Ziyan He, Shi Jin, and Xu Li” Performance Analysis of Mixed-ADCMassiveMIMO Systems Over Rician Fading Channels” IEEE JOURNAL ON SELECTED AREASIN COMMUNICATIONS, VOL. 35, NO.
6, JUNE 2017 13273 R. Walden,”Analog-to-digital converter survey and analysis,” IEEE Sel.Areas Commun.,vol. 17, no. 4, pp.
539–550, 1999.4 B. Murmann, “ADCperformance survey 1997-2014.
” Online.Available:http://www.stanford.edu/_murmann/adcsurvey.html5 B. Le, T.
Rondeau, J. Reed,and C. Bostian, “Analog-to-digital converters,” IEEE Signal Process. Mag., vol.
22, no.6, pp. 69–77, 2015.6S.
Vitali, G. Cimatti, R.Rovatti, and G. Setti, “Adaptive time-interleaved ADC offset compensation bynonwhite data chopping,” IEEE Trans. CircuitsSyst. II,vol. 56, no. 11, pp.
820–824, 2009. 7Song-NamHong_, Seonho Kim_, and NamyoonLeey”Uplink Multiuser Massive MIMOSystems with Low-Resolution ADCs:A Coding-Theoretic Approach Ajou University,Suwon, Korea.8B.Murmann, “ADC Performance Survey 1997-2015,” Online.Available:http://web.stanford.edu/ murmann/adcsurvey.html.
9 N. Liang and W. Zhang “A mixed-ADCreceiver architecture for massive MIMOsystems.”. (Nov. 2015). 10 WEIQIANG TAN1, , SHI JIN1,”SpectralEfficiency of Mixed-ADC Receiversfor Massive MIMO Systems “August 25, 2016, DigitalObject Identifier 10.
-C. Zhang, C.-K.
Wen, S.Jin, and T. Jiang. “Mixed-ADC massive MIMO detectors: Performance analysis anddesign optimization. (Dec.
2015).12 N. Liang and W.Zhang, “Mixed-ADC massive MIMO,” IEEE J. Sel. Areas Commun., vol. 34, no.
4, pp.983–997, Apr. 2016.13 W.
Tan, S. Jin, C.Wen, and Y. Jing. “Spectral efficiency of mixed-ADC receivers for massive MIMOsystems.
” (Aug. 2016). 14 F. Boccardi, R.
W. Heath, Jr., A.Lozano, T. L. Marzetta, and P.
Popovski,”Five disruptive technology directionsfor 5G,” IEEECommun. Mag.,vol.52, no. 2, pp.
74–80, Feb. 2014. 15 B. Murmann.
ADC Performance Survey 1997-2016, accessed on Mar. 2017. 16 R. Méndez-Rial, C.Rusu, N.
González-Prelcic, A. Alkhateeb, and R. W. Heath, Jr., “Hybrid MIMO architecturesfor millimeter wave communications:Phase shifters or switches?” IEEE Access, vol. 4,pp.