CN100574175C - The method and apparatus that obtains the method and system of high-speed uplink packet scheduling capacity and obtain link effective data rate - Google Patents

Abstract

The invention discloses a kind of method and apparatus that obtains the method and system of high-speed uplink packet scheduling capacity and obtain link effective data rate, relate to wireless communication field; The invention solves prior art and carrying out power system capacity when estimation, a taking into account system layer and cause estimating that the result does not meet the defective of actual conditions.The present invention under different rapid mixing re-transmission mode automatically, obtains link effective data rate according to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer; Determine the chip signal-noise ratio threshold according to system's residue frame error rate, draw the corresponding relation of link effective data rate and chip signal to noise ratio; Utilize described link effective data rate to obtain power system capacity.Analysis-by-synthesis of the present invention the influence of the automatic retransmission technique of link adaptation techniques, rapid mixing of link aspect and system layer fast dispatch to high-speed uplink packet scheduling capacity, thereby make properer the tallying with the actual situation of estimation result.

Description

The method and apparatus that obtains the method and system of high-speed uplink packet scheduling capacity and obtain link effective data rate

Technical field

The present invention relates to wireless communication field, relate in particular to high-speed uplink packet scheduling (HSUPA) capacity estimation method and system, effective data rate evaluation method and the device of a kind of WCDMA of being used for system.

Background technology

The develop rapidly of wireless data service has proposed new requirement to communication system.Gradually, can not satisfy people's requirement merely based on the mobile communication of speech, therefore mobile communication system in the future must be on the basis that guarantees voice service, provide and transmit picture file, receiving and dispatching mail, surf the web, even multimedia service such as movies on demand, to satisfy the demand business of user to high-speed data.

In order to provide on the existing network basis more at a high speed and more advanced wireless data communication service, the various enhancement techniques that are used for mobile data communication have appearred.As present HSUPA (High Speed Uplink PacketAccess, high-speed uplink packet scheduling) is a kind of Enhanced Technology that 3GPP proposes for the demand that satisfies upstream data service in Release 6 release protocol.The new technology that HSUPA adopts comprises H-ARQ (Hybrid Automatic Repeat Request, rapid mixing retransmits automatically), Adaptive Modulation and Coding (Aadaptive Modulation and Coding, AMC), fast dispatch (Fast Scheduling), the short frame transmission of 2msTTI (Transmission Time Interval, Transmission Time Interval) etc.

H-ARQ is meant that the recipient under the situation of decoding failure, preserves the data that receive, and requires the transmit leg data retransmission, and the recipient made up data that retransmit and the data that before received before decoding.H-ARQ can improve systematic function, and can adjust effective code element speed neatly, can also compensate because the error code that adopts link adaptation to bring.H-ARQ has two kinds of operational modes, is respectively to catch up with merging (chasecombining) and steadily increase redundancy (IR, Incremental Redundancy).Based on the method for catching up with merging, transmitting terminal is after the judgement of making re-transmission, the all or part of of Frame that last time sends resend, result after receiving terminal then merges all Frames that receive decodes, the Frame that receives is many more, the probability of correct decoding is just big more, though the Frame that at every turn receives may be all correct decoding separately, but the result of a plurality of Frames merging still might carry out correct decoding; The mechanism of steadily increase redundancy method with catch up with merge basic identical, it is the coding redundancy degree that each repeating transmission all can increase Frame, correspondingly code rate can reduce, its starting point is exactly that the coding redundancy degree of thinking Frame has caused sending failure inadequately, and when retransmitting next time, transmission conditions can not improved, and therefore need constantly to increase the redundancy of coding, to guarantee that Frame can correctly send as early as possible.

The basic principle of Adaptive Modulation and Coding (AMC) is based on channel status adjustment modulation and coded system.The assessment of channel status is from the feedback of receiver.In the system of configuration AMC, user near cell site generally uses high order modulation and high-rate coded (such as 64QAM, R=3/4Turbo sign indicating number), and adopts low-order-modulated and low rate coding (such as QPSK, R=1/2Turbo sign indicating number) near the user of cell boarder.The major advantage of AMC is: thus a) user of correct position has higher data rate to increase the average throughput of sub-district; B) method of minimizing interference variations is to change modulation and coded system, rather than changes through-put power.

Suppose that A is that base station end, B are that user side (UE) (1) A held before certain sends, and at first predicts the current signal to noise ratio of A → B channel (average signal-to-noise ratio that transmitting element is interior); (2) the A end removes to look into the AMC table of prior formulation with the signal to noise ratio of prediction, learns suitable coding and modulation system; When (3) the AMC tabular has gone out given channel signal to noise ratio,, select best modulation, the coded system of spectrum efficiency guaranteeing that error rate is not higher than under the prerequisite of certain particular value.The AMC table generally is to set up by prior emulation.Table 1 is the example of ANC table, and this table is by 5% target frame error rate design.The hypothesis signal to noise ratio snr only rounds numerical value in the table.Because the SNR that prediction obtains can not be quite accurate, so it is just enough that SNR is expressed as integer.For example Yu Ce signal to noise ratio is 10dB, and 1 modulation system that is fit to as can be known of tabling look-up so is 16QAM, and encoding rate is 1/3.

Table 1

SNR scope (dB)	≤2	3～5	6～7	8～11	12～15	16～20	21～22	＞＝23
									Modulation system	QPSK	QPSK	16QAM	16QAM	64QAM	64QAM	64QAM	64QAM
Encoding rate	1/4	1/3	1/4	1/3	1/2	1/2	2/3	3/4
									Spectrum efficiency (bps/Hz)	1/2	2/3	1	4/3	2	3	4	4.5

Dispatching algorithm mainly data volume to be launched and professional situations such as priority level such as considers simultaneously based on channel condition, with the mean data rate and the resistance to overturning of maximization system.The index of estimating a dispatching algorithm quality is: throughput and fairness, throughput comprise cell throughout and user throughput, and fairness is commonly considered as the statistics of each user or the professional busy channel resource of different grouping.Dispatching algorithm commonly used has RR scheduling (Round Robin, polling dispatching), Max C/I scheduling (MaximumCarrier-to-Interference Rate, max carrier to interference scheduling), PF scheduling (ProportionalFair, Proportional Fair) etc.The RR dispatching algorithm adopts polling mode, service is provided for each user according to the position of each user in fifo queue.This algorithm can guarantee the fairness between the user, but does not consider the characteristic of subscriber channel, may cause the throughput of system low.Max C/I dispatching algorithm is carried out the ordering of priority from high to low according to user's carrier/interface ratio to the user, thus at any time always the best user priority of carrier/interface ratio accept service.This dispatching algorithm can make the throughput of system and the availability of frequency spectrum obtain maximization, but lacks fairness, makes the user near cell boarder can not get sufficient service.The PF dispatching algorithm is the compromise of RR and Max C/I, it is based on the skew of long-term average carrier/interface ratio and instant carrier/interface ratio, the characteristic of channel, data volume and service state according to the user are calculated a priority for each user, serve for the user of sub-district medium priority maximum at any time, so both can guarantee the throughput of system, can provide certain user fairness again.In order to make system's short term variations of adaptive channel better, in high-speed wireless link, dispatching algorithm is positioned at Node-B (base station) rather than is positioned at RNC (Radio NetworkController radio network controller).

(Dedicated channel, TTI DCH) are 10ms, 20ms, 40ms and 80ms to special uplink channel in Release 99 release protocol.In HSUPA, allow to continue to use original 10ms TTI, also can further introduce shorter 2ms TTI simultaneously to reduce transmission delay.

For continuous channel, if B is a channel width, under the interference of additive white Gaussian noise, according to Shannon information theory, its channel capacity C=Blog ₂(1+S/N) (b/s) wherein, N is the average power of noise; S is the average power of signal; S/N is a signal to noise ratio.Channel capacity C is meant the maximum information speed (being the maximum transmitted ability that channel can reach) that channel may transmit.Can draw to draw a conclusion by following formula:

If there is a kind of method in theory in 1 information rate R＜=C, can pass through Channel Transmission with arbitrarily small error probability; If R＞C, error free transmission is impossible in theory.

2,, can transmit with the combination of different bandwidth and signal to noise ratio for given C.If reduce bandwidth, then must send bigger power, promptly increase signal to noise ratio S/N; If there is the then available less through-put power of bigger transmission bandwidth (being less S/N) to transmit.This shows that broadband system shows better anti-interference.Therefore, when signal to noise ratio is too little, in the time that communication quality can not being guaranteed, can adopt broadband system, to improve communication quality.But the exchange of bandwidth and signal to noise ratio is not automatically, must make it to have desired bandwidth by figure signal.In fact this is finished by various types of modulation and coding.Modulation and cataloged procedure are exactly the means that realize that bandwidth and signal to noise ratio are exchanged.

What (being code channel number) of code channel are relevant with the spreading factor that is adopted in the time slot of frame structure; Spreading factor is the dimension of OSVF sign indicating number, can equal 4,8,16,32,64,128,256,512.The more numbers of users many more (under the perfect conditions) that mean that system can hold of dimension, different channels carries out Code And Decode with similar and different OSVF sign indicating number more.

After the physical channel mapping, the data on the channel will carry out spread spectrum and scrambler is handled.So-called spread spectrum multiplies each other with the Serial No. and the channel data that are higher than data bit rates exactly, and multiplied result has been expanded the bandwidth of signal, has converted the data flow of bit rate to spreading rate data flow.

Document [C.Rosa, J.Outes, T.B.Sorensen, J.Wigard, and P.E.Mogensen.Combined time and code division scheduling for enhanced uplink packetaccess in WCDMA (WCDMA strengthens the upstream packet access time-division and sign indicating number divides in conjunction with dispatching) [J] .IEEE60th VTC.2004,9 (2): 851 855.] studied HSUPA enhanced services channel (EnhancedData Channel, E-DCH) the three kinds of time on and the assessment of rate scheduling mode, three kinds of blind fair throughput scheduling of dispatching algorithm have been analyzed, the scheduling of maximum transmission power efficient, based on the dispatching criterion of the fair throughput scheduling of channel status, and provided simulation result.But there is following defective in this technical scheme: only provided the criterion and the simulation result of dispatching algorithm, do not derived the throughput of system expression formula when adopting dispatching algorithm.Do not have and the combination of link layer performance.

Document [S.Das, H.Viswanatha.On the reverse link interferencestructure for next generation cellular systems (the return link interferences structure of cellular system of future generation) [J] .IEEE GLOBECOM.2004,12 (5): 3,068 307.] different transmission mechanisms have been analyzed based on sub-district internal and external interference situation.The sub-district internal and external interference comprises that three kinds of situations (a) inside, sub-district is noiseless, the outside average interference in sub-district; (b) inside, sub-district is noiseless, the outside bursty interference in sub-district; (c) the inner average interference in sub-district, the outside average interference in sub-district.No matter whether ul transmissions is whether time division multiplexing, each TTI pass between one or more user's data, a plurality of user quadrature, and the area interference structure must fall among the three above generic categories.Document analysis the throughput performance under three kinds of situations when adopting H-ARQ, and under the restriction of RoT (Rise of Thermal, noise rise), compared the performance of various mechanism.But there is following defective in this technical scheme: do not consider concrete cell topology when analyzing interference profile, as hexagon serving sector structure; Only considered the polling dispatching algorithm, do not considered of the influence of other dispatching algorithms throughput of system; Do not have and the combination of link layer performance.

In the present invention, capacity is to weigh with the throughput of system, and throughput has the branch of user throughput and throughput of system, and throughput of system is appreciated that into capacity.User throughput is meant the peak transfer rate that the user can reach.Effective data rate can be regarded as throughput in this case.

Summary of the invention

Embodiments of the invention provide a kind of can be with the volume calculation step of link aspect and the system level method and system that obtains high-speed uplink packet scheduling capacity of combination organically, and the method and apparatus that obtains link effective data rate.

Embodiments of the invention provide a kind of method of obtaining high-speed uplink packet scheduling capacity, comprise the steps:

According to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, draw the link effective data rate EfflinkRate under different rapid mixing re-transmission mode automatically;

Determine the chip signal-noise ratio threshold according to the residue frame error rate;

According to described link effective data rate and chip signal-noise ratio threshold, draw the corresponding relation of link effective data rate and chip signal to noise ratio;

According to the corresponding relation of link effective data rate and chip signal to noise ratio, utilize described link effective data rate to obtain power system capacity.

Embodiments of the invention also provide a kind of system that obtains high-speed uplink packet scheduling capacity, comprising:

The link effective data rate deriving means is used for spreading rate, modulation coding mode, spreading factor and code channel number according to physical layer, draws the link effective data rate under different rapid mixing re-transmission mode automatically;

Chip signal-noise ratio threshold calculation element is used for determining the chip signal-noise ratio threshold according to system's residue frame error rate;

Link effective data rate and chip signal to noise ratio corresponding relation are determined device, are used for drawing the corresponding relation of link effective data rate and chip signal to noise ratio according to described link effective data rate and chip signal-noise ratio threshold;

The power system capacity deriving means according to the corresponding relation of link effective data rate and chip signal to noise ratio, utilizes described link effective data rate to obtain power system capacity.

Embodiments of the invention provide a kind of method of obtaining the high-speed uplink packet scheduling link effective data rate simultaneously, comprising:

According to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, under different rapid mixing re-transmission mode automatically, obtain link effective data rate.

Embodiments of the invention also provide a kind of high-speed uplink packet scheduling effective data rate device that obtains to comprise: bit error probability computing unit, be used for spreading rate, modulation coding mode, spreading factor and code channel number, obtain the bit error probability P of physical layer setting _b

Transmit the frame error rate computing unit first, be used for the bit error probability P that obtains according to described bit error probability computing unit _b, utilize formula $P_e=1-{(1-P_b)}^{N_p}$ Draw data frame transfer frame error rate P first _e

Merge the frame error rate computing unit, be used for according to transmitting the frame error rate of the data frame transfer first P that the frame error rate computing unit obtains first _e, under different rapid mixing re-transmission mode automatically, obtain the merging frame error rate P after twice Frame merges _s

The number of transmissions probability distribution computing unit is used for according to the described frame error rate of data frame transfer first P _eWith merging frame error rate P _s, obtain data frame transfer number of times probability distribution Pr (j);

Average transmission number of times computing unit is used for obtaining Frame average transmission number of times according to data frame transfer number of times probability distribution Pr (j) $N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\Σ}}}\mathrm{jPr}\left(j\right);$

Residue frame error rate computing unit is used for the maximum retransmission N that is allowing _MaxIn, obtain residue frame error rate FER according to Frame average transmission number of times _r:

${\mathrm{FER}}_r=1-\underset{j=1}{\overset{N_{\max }}{\mathrm{\Σ}}}\mathrm{Pr}\left(j\right)$

The effective data rate computing unit is used for according to described residue frame error rate FER _r, obtain link effective data rate EffinkRate

$\mathrm{EfflinkRate}=\frac{W\·{\log }_{2}M\·R_c\·N_c\·(1-{\mathrm{FER}}_r)}{\mathrm{SF}\·N_s},$

Wherein W is spreading rate, R _cBe code rate, N _cBe code channel number, M is the QAM modulation index, and SF is a spreading factor, N _SBe Frame average transmission number of times, FER _rBe the residue frame error rate.

Embodiments of the invention have been considered the corresponding relation and the link effective data rate of link effective data rate and chip signal to noise ratio in obtaining the power system capacity process, organically combine the link layer performance, analyze the HSUPA performance more all sidedly.

Description of drawings

Fig. 1 obtains the realization flow figure of the method for high-speed uplink packet scheduling capacity for the embodiment of the invention;

Fig. 2 obtains the schematic diagram that hexagon serving sector normalization diameter in the method for high-speed uplink packet scheduling capacity is 1,0≤r≤1/2 o'clock for the embodiment of the invention, r is the polar coordinates radius of hexagon serving sector;

Fig. 3 obtains for the embodiment of the invention that hexagon serving sector normalization diameter is 1 in the method for high-speed uplink packet scheduling capacity, $1/2<r\&le;\sqrt{3}/2$ The time schematic diagram, r is the polar coordinates radius of hexagon serving sector;

Fig. 4 obtains for the embodiment of the invention that hexagon serving sector normalization diameter is 1 in the method for high-speed uplink packet scheduling capacity, $\sqrt{3}/2<r\&le;1$ The time schematic diagram, r is the polar coordinates radius of hexagon serving sector;

Fig. 5 obtains the system construction drawing of high-speed uplink packet scheduling capacity for the embodiment of the invention;

Fig. 6 obtains the realization flow figure of the method for high-speed uplink packet scheduling effective data rate for the embodiment of the invention;

Fig. 7 obtains the structural representation of high-speed uplink packet scheduling effective data rate device for the embodiment of the invention.

Embodiment

Defective at prior art taking into account system layer when carrying out the cell capacity estimation, embodiments of the invention have been taken all factors into consideration the various key technologies of HSUPA (rapid mixing re-transmission automatically, fast dispatch, the short frame transmission of 2ms TTI) and to the influence of HSUPA capacity, have been comprised two aspects of link and system.

At link layer, embodiments of the invention have provided the mathematic(al) representation of the link effective data rate (formula (9)) when Frame average transmission number of times (formula (5)), residue frame error rate (formula (7)), given modulation coding mode, spreading factor and code channel number make up.At this moment, the place, base station does not receive the chip signal to noise ratio when considering that probability that portable terminal is scheduled and portable terminal are scheduled, and drops on the probability in signal to noise ratio interval, MCS mode n place.

In system layer, take into full account probability and (2) that (1) portable terminal is scheduled when calculating the portable terminal throughput and received the chip signal to noise ratio in the place, base station when portable terminal is scheduled, drop on the probability in signal to noise ratio interval, MCS mode n place.

(1) probability that is scheduled of portable terminal

Consider weighting type dispatching algorithm.Under such dispatching algorithm, user's service priority computing formula is:

PRIORITY _i＝ω _i·r _i(t) (1)

Wherein, ω _iBe the dispatch weight that system distributes for user i, r _i(t) be the instantaneous signal-to-noise ratio of user i at moment t.For Max C/I dispatching algorithm, ${\mathrm{\&omega;}}_i^{\mathrm{dB}}=0;$ For the PF dispatching algorithm, ${\mathrm{\&omega;}}_i^{\mathrm{dB}}=-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}.$

Consider N user, its position is respectively (r _i, θ _i) (1≤i≤N), then to receive i user's chip signal to noise ratio be γ to the place, base station _i ^DB, and i user by the probability density function of base station scheduling is

$f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)=f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)\&CenterDot;\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)---\left(2\right)$

Wherein

$f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_T}\exp [-\frac{{({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_i,{\mathrm{\&theta;}}_i))}^2}{{2\mathrm{\&sigma;}}_T^2(r_i,{\mathrm{\&theta;}}_i)}]$

According to different dispatching algorithms, different Pr (UEiisschedule) value is arranged.

The place, base station received the probability that the chip signal to noise ratio drops on signal to noise ratio interval, MCS mode n place when (2) portable terminal was scheduled.

Embodiments of the invention have provided the relation (signal-noise ratio threshold is meant and carries out the signal-noise ratio threshold value that the MCS mode is switched) of modulation coding mode, spreading factor and code channel number and the signal-noise ratio threshold of physical layer:

γ ₀＝0

γ _n=f (P _Loss, MCS _n) n=1,2 ..., N _Mcs, γ wherein _nBe that whole chip signal to noise ratio scope is divided into

${\mathrm{\&gamma;}}_{N_{\mathrm{mcs}}+1}=+\&infin;$

N _McsInterval separation behind+1 nonoverlapping continuum is designated as

The interval corresponding modulation coding mode of each chip signal to noise ratio is guaranteeing that the residue frame error rate is not higher than the packet loss P that system allows _LossPrerequisite under, modulation, coded system that base station selected spectrum efficiency is best.

In system layer, the place, base station received chip signal to noise ratio γ when i portable terminal was scheduled _i ^DBThe probability that drops on the signal to noise ratio interval at MCS mode n place is

$\mathrm{Pr}\left(n\right)={\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{f}_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)d{\mathrm{\&gamma;}}_i^{\mathrm{dB}}---\left(3\right)$

Average repeat time under the MCS mode n is simultaneously

$\stackrel{\&OverBar;}{N}\left(n\right)=\frac{{\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{N}_s{f}_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right){\mathrm{d\&gamma;}}_i^{\mathrm{dB}}}{\mathrm{Pr}\left(n\right)}---\left(4\right)$

Residue frame error rate under the MCS mode n is

${\mathrm{FER}}_r\left(n\right)=\frac{{\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{\mathrm{FER}}_r\&CenterDot;f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right){\mathrm{d\&gamma;}}_i^{\mathrm{dB}}}{\mathrm{Pr}\left(n\right)}---\left(5\right)$

Then N portable terminal position is (r _i, θ _i) (i user's throughput is during 1≤i≤N)

$\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i)=W\&CenterDot;\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{{\log }_{2}M\left(n\right)\&CenterDot;R_c\left(n\right)\&CenterDot;N_c\left(n\right)\&CenterDot;(1-{\mathrm{FER}}_r\left(n\right))\&CenterDot;\mathrm{Pr}\left(n\right)}{\mathrm{SF}\left(n\right)\&CenterDot;\stackrel{\&OverBar;}{N}\left(n\right)}---\left(6\right)$

Embodiments of the invention obtain the method for high-speed uplink packet scheduling capacity, comprise the steps:

According to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, under different rapid mixing re-transmission mode automatically, obtain link effective data rate;

Draw the corresponding relation of link effective data rate and chip signal to noise ratio;

According to the corresponding relation of link effective data rate and chip signal to noise ratio, utilize described link effective data rate to obtain power system capacity.

The present invention at hexagon serving sector structure, analyzes portable terminal receives chip signal to noise ratio and uplink noise lifting at the place, base station distribution situation when system level computing system capacity;

Then,, utilize described link effective data rate, draw sector throughput, user throughput fixedly the time at all activated mobile terminal locations according to described signal to noise ratio and noise rise distribution situation;

According to described user throughput, require and the activation of service factor for given service rate, the system of drawing satisfies user's ratio;

According to described sector throughput and user throughput, the average sector throughput when drawing all activated portable terminal and in serving sector, evenly distributing, each user's average throughput;

Average throughput and system according to described average sector throughput, each user satisfy user's ratio, draw average system and satisfy user's ratio.

With reference to accompanying drawing 1, below embodiments of the invention are described in further detail.

A. link aspect

The restriction of taking into account system greatest admissible retransmission value and the error of transmission of backward channel acknowledgement frame when analyzing the H-ARQ performance.Derive data frame transfer number of times probability distribution and average the number of transmissions under certain chip signal to noise ratio.Can retransmit residue frame error rate afterwards in the hope of carrying out maximum times simultaneously.Based on this, analyze the link effective data rate that obtains under certain modulation coding mode, spreading factor (SF) and the code channel number combination.According to the residue frame error rate requirement of system, determine the chip signal-noise ratio threshold of MCS mode (being modulation coding mode) again.Obtain the corresponding relation of link effective data rate and chip signal to noise ratio at last.

A1. data frame transfer frame error rate first

If each bit in the grouping bag has identical bit error rate, and the bit mistake is separate, then transmits frame error rate first to be

$P_e=1-{(1-P_b)}^{N_p}---\left(7\right)$

$N_p=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;\mathrm{TTI}}{\mathrm{SF}}---\left(8\right)$

Wherein, W is a spreading rate, N _pBe grouping packet length, R _cBe code rate, N _cBe code channel number.P _bBe the bit error probability, under different modulation coding modes, itself and chip signal to noise ratio γ _cCorresponding relation determine by the error rate result of physical layer.As physical layer modulation mode DPSK, when demodulation mode is differential coherence, $P_b=\frac{1}{2}{e}^{-{\mathrm{\&gamma;}}_c}.$

A2. the frame error rate after twice Frame merges

When a) the H-ARQ mode merges for catching up with

Catch up with and merge H-ARQ employing Rake receiver high specific folding, the error rate P after then the adjacent two data frame merges _sWith P _eSimilar expression formula is arranged, as long as the chip signal to noise ratio is become γ _c'=2 γ _cGet final product P _sExpression formula.Supposing equally has under the DPSK situation:

$P_b^{\&prime;}=\frac{1}{2}{e}^{-{\mathrm{\&gamma;}}_c^{\&prime;}}=\frac{1}{2}{e}^{-2{\mathrm{\&gamma;}}_c}$

$P_s=1-{(1-P_b^{\&prime;})}^{N_p}$

B) steadily increase redundancy H-ARQ

The situation of twice transferring data frames before and after considering to merge.Twice data frame transfer can be equivalent to the once transmission of carrying out with the original code length of twice, 1/2 original encoding speed, promptly makes N in formula (1) _p'=2N _p, at P _bWith γ _cCorresponding relation computational process in make $R_c^{\&prime;}=\frac{R_c}{2}$ Once transmission, get final product P _sExpression formula.Also be example with DPSK:

$P_s=1-{(1-P_b)}^{N_p^{\&prime;}}$

$N_p^{\&prime;}={2N}_p=2\frac{W\&CenterDot;{\log }_2^M\&CenterDot;R_c^{\&prime;}\&CenterDot;N_c\&CenterDot;\mathrm{TTI}}{\mathrm{SF}}=\frac{W\&CenterDot;{\log }_2^M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;\mathrm{TTI}}{\mathrm{SF}}$

A3. data frame transfer number of times probability distribution

After receiving terminal receives the transmitting terminal signal, need at once to send ACK (correctly receiving return signal) or NACK (mistake receive or do not have reception return signal) to transmitting terminal, whether the indication transmitting terminal is retransmitted.During calculated data frame the number of transmissions probability distribution, need to consider the desirable and imperfect two kinds of situations of feedback acknowledgment channel.

Ideal feedback

Suppose feedback acknowledgment channel ideal, promptly the feedback acknowledgment signal error probability from receiving terminal that receives of transmitting terminal is 0, the probability of the inferior successful decoded information piece of j (j＞1) then, and promptly the number of transmissions is that the probability of j is:

Pr(j)＝P _e(P _eP _s) ^j-2(1-P _eP _s) (9)

When j=1, Pr (j)=(1-P _e).

A) imperfect feedback

Suppose that the feedback acknowledgment channel is imperfect, 3 kinds of situations may appear in the feedback information ACK of receiving terminal or NACK in transmission course: correct transmission, erroneous transmissions and lose.Making system allow maximum retransmission is N _Max, the probability of the inferior successful decoded information piece of j (j＞1) then, promptly the number of transmissions is that the probability of j is:

Wherein, P _fBe the probability that receiving terminal feedback information ACK or NACK lose, the probability that receiving terminal feedback information ACK loses is generally 0.01, and the probability that receiving terminal feedback information NACK loses is generally 0.1; P _AeExpression receiving terminal feedback information ACK the probability of being misinformated for NACK, the span of this probability is generally 0.01-0.04; P _NeThe receiving terminal feedback information NACK probability for ACK of being misinformated, the span of this probability is generally 0.01-0.04.

A4. Frame average transmission number of times

The average of the number of transmissions is

$N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{jPr}\left(j\right)---\left(11\right)$

Under desirable feedback, the average transmission number of times can directly be expressed as

$N_s=\frac{1+P_e-P_e{P}_s-P_e{\left(P_e{P}_s\right)}^{N_{\max }-1}-N_{\max }{P}_e{\left(P_e{P}_s\right)}^{N_{\max }-1}+N_{\max }{P}_e{\left(P_e{P}_s\right)}^{N_{\max }}}{1-P_e{P}_s}---\left(12\right)$

A5. remain frame error rate

After the residue frame error rate was meant that the maximum times of carrying out system's permission retransmits, Frame did not have the successfully ratio of transmission.

System at first defines the maximum transmission times (N of permission _Max), obtain the residue frame error rate then

${\mathrm{FER}}_r=1-\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(j\right)---\left(13\right)$

Under desirable feedback, the residue frame error rate can directly be expressed as

${\mathrm{FER}}_r=P_e{\left(P_e{P}_s\right)}^{N_{\max }-1}---\left(14\right)$

System is to FER _rCertain requirement is arranged, must be no more than P _Loss, i.e. FER _r≤ P _Loss

Link effective data rate when A6. given modulation coding mode, spreading factor and code channel number make up

When system adopts M-QAM modulation, code rate is R _cTurbo code the time, link effective data rate is

$\mathrm{EfflinkRate}=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;(1-{\mathrm{FER}}_r)}{\mathrm{SF}\&CenterDot;N_s}---\left(15\right)$

Wherein W is spreading rate, R _cBe code rate, N _cBe code channel number, M is the QAM modulation index, and SF is a spreading factor, N _SBe Frame average transmission number of times, FER _rBe the residue frame error rate.

The chip signal-noise ratio threshold of A7.MCS mode

Design object is a maximum data speed, satisfies the requirement of residue frame error rate simultaneously.Suppose that through-put power remains unchanged.Cutting apart whole chip signal to noise ratio scope is N _Mcs+ 1 nonoverlapping continuum, interval separation is designated as

Work as γ _c∈ [γ _n, γ _N+1) time, select corresponding MCS pattern n.For fear of deep fade, work as γ _c∈ [γ ₀, γ ₁) time, there are not valid data to propagate.

Provide the chip signal to noise ratio separation γ of MCS mode below _nDefinite method.γ _nBe to guarantee under the scene of transmission mode n, to make FER _r=P _LossThe chip signal to noise ratio, can be expressed as

γ ₀＝0

γ _n＝f(P _loss，MCS _n) n＝1，2，...，N _mcs (16)

${\mathrm{\&gamma;}}_{N_{\mathrm{mcs}}+1}=+\&infin;$

The γ that through type (16) is tried to achieve _n, can make the MCS mode of selection meet the requirement of residue frame error rate.Under the prerequisite that keeps target residue frame error rate to require, this choice mechanism can maximize the availability of frequency spectrum.

A8. the corresponding relation of link effective data rate and chip signal to noise ratio

Convolution (15) (16) can obtain the corresponding relation of link effective data rate and chip signal to noise ratio.Work as γ _c∈ [γ _n, γ _N+1) time, select MCS mode n, obtain corresponding effective data rate by formula (15).For fear of deep fade, work as γ _c∈ [γ ₀, γ ₁) time, there are not valid data to propagate.

At link layer,, can obtain data frame transfer frame error rate first by the error rate of this layer for modulation coding mode, spreading factor and the code channel number of given physical layer; Can determine to be operated under two kinds of working methods of H-ARQ the frame error rate after twice the number frame merges according to data frame transfer frame error rate first; According to the frame error rate after data frame transfer frame error rate and twice number frames merge first, can obtain and carry out the residue frame error rate of maximum times after retransmitting, pass number of times number of times probability distribution and average the number of transmissions, thus modulation coding mode, spreading factor (SF) and the code channel number of the trying to achieve given physical layer link effective data rate under making up.According to the residue frame error rate requirement of system, determine the chip signal-noise ratio threshold of MCS mode (being modulation coding mode) again.Obtain the corresponding relation of link effective data rate and chip signal to noise ratio at last.Consider the restriction of greatest admissible retransmission value and the error of transmission of backward channel acknowledgement frame at link layer, provide the corresponding relation of link effective data rate and chip signal to noise ratio, overcome the limitation of independent analysis link adaptation techniques and H-ARQ performance, thereby make properer the tallying with the actual situation of result.

B. system level

B1. the current mobile terminal chip signal to noise ratio probability distribution that receives of base station place

Suppose: (1) considers the shadow fading of logarithm normal distribution, and (2) are used polar coordinates (r, θ), as limit, horizontal line is as pole axis this cell base station, (a 3) k portable terminal (r _k, θ _k) to the propagation loss of base station be

$L(r_k,{\mathrm{\&theta;}}_k)=r_k^{-l}{10}^{\frac{X_k}{10}}=r_k^{-l}{e}^{Y_k}=r_k^{-l}{K}_k---\left(17\right)$

Wherein l is path loss exponent (representative value is 3 or 4), X _kFor average is zero, standard deviation is

Gaussian distribute K _kFor the lognormal stochastic variable, represent shadow fading.

In addition, i _rOther sub-districts that receive, expression base station and the ratio of this cell power, with the base station receive from the power sum of peripheral cell portable terminal divided by calculating of receiving from this cell mobile terminal power sum:

$i_r=\frac{I_{\mathrm{oth}}}{I_{\mathrm{own}}}---\left(18\right)$

Because the receiver by the base station calculates i in up link _rSo, i _rInfluence to all connections of a sub-district is all similar.According to simulation result, i _rThe typical means scope generally from 0.15 (more independently Microcell) to 1.2 (relatively poor wireless network plannings).

Portable terminal (the r that the place, base station receives _k, θ _k) the chip signal to noise ratio be

${\mathrm{\&gamma;}}_k={\left(\frac{E_c}{N_t}\right)}_k=\frac{P_{k}L(r_k,{\mathrm{\&theta;}}_k)}{N_{0}W+(I_{\mathrm{own}}-P_{k}L(r_k,{\mathrm{\&theta;}}_k))+i_r{I}_{\mathrm{own}}}$

$=\frac{P_{k}L(r_k,{\mathrm{\&theta;}}_k)}{N_{0}W+\underset{j=1,j\&NotEqual;k}{\overset{N}{\mathrm{\&Sigma;}}}{P}_j{L}_j(r_j,{\mathrm{\&theta;}}_j)+i_r\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_j{L}_j(r_j,{\mathrm{\&theta;}}_j)}---\left(19\right)$

$=\frac{P_k{K}_k{r}_k^{-l}}{N_{0}W+(1+i_r)\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{P}_j{K}_j{r}_j^{-l}-P_k{K}_k{r}_k^{-l}}$

Wherein W is a spreading rate, N ₀Be thermal noise density, P _kBe the transmitting power of k portable terminal, compare that the background heat The noise can be ignored with total transmitted signal power.Then

${\mathrm{\&gamma;}}_k=\frac{K_k}{(1+i_r)\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{P_j}{P_k}\&CenterDot;{\left(\frac{r_j}{r_k}\right)}^{-l}{K}_j-K_k}---\left(20\right)$

Wherein $K_k=e^{Y_k}$ Be logarithm normal distribution, Y _kFor average is

Standard deviation is

About Gaussian stochastic variable.K then _kAverage and variance be respectively

$E\left[K_k\right]=e^{m_{Y_k}+{\mathrm{\&sigma;}}_{Y_k}^2/2}---\left(21\right)$

$\mathrm{Var}\left(K_k\right)=\exp \{{2m}_{Y_k}+{\mathrm{\&sigma;}}_{Y_k}^2\}\&CenterDot;\left(\exp \right\{{\mathrm{\&sigma;}}_{Y_k}^2\}-1)---\left(22\right)$

$(1+i_r)\frac{P_j}{P_k}\&CenterDot;{\left(\frac{r_j}{r_k}\right)}^{-l}{K}_j$ Be the lognormal stochastic variable, whole denominator can be approximated to be the lognormal stochastic variable, is designated as I, and its average and variance can export as

$E\left[I\right]=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}(1+i_r)\frac{P_j}{P_k}\&CenterDot;{\left(\frac{r_j}{r_k}\right)}^{-l}\&CenterDot;E\left[K_j\right]-E\left[K_k\right]---\left(23\right)$

$=(1+i_r)\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{P_j}{P_k}\&CenterDot;{\left(\frac{r_j}{r_k}\right)}^{-l}\&CenterDot;e^{m_{Y_j}+{\mathrm{\&sigma;}}_{Y_j}^2/2}-e^{m_{Y_k}+{\mathrm{\&sigma;}}_{Y_k}^2/2}$

$\mathrm{Var}\left(I\right)=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{[(1+i_r)\&CenterDot;\frac{P_j}{P_k}\&CenterDot;{\left(\frac{r_j}{r_k}\right)}^{-l}]}^2\&CenterDot;\mathrm{Var}\left[K_j\right]-\mathrm{Var}\left[K_k\right]$

$=\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{[(1+i_r)\&CenterDot;\frac{P_j}{P_k}\&CenterDot;{\left(\frac{r_j}{r_k}\right)}^{-l}]}^2\&CenterDot;\exp \{{2m}_{Y_j}+{\mathrm{\&sigma;}}_{Y_j}^2\}\&CenterDot;\left(\exp \right\{{\mathrm{\&sigma;}}_{Y_j}^2\}-1)---\left(24\right)$

$-\exp \{{2m}_{Y_k}+{\mathrm{\&sigma;}}_{Y_k}^2\}\&CenterDot;\left(\exp \right\{{\mathrm{\&sigma;}}_{Y_k}^2\}-1)$

Because the I obeys logarithm normal distribution is then established $I={10}^{\frac{X_I}{10}},$ X then _IAverage and variance can obtain by the Wilkinson method

$\mathrm{Var}\left(X_I\right)=\frac{1}{{\mathrm{\&beta;}}^2}\ln (\frac{\mathrm{Var}\left(I\right)}{E{\left[I\right]}^2}+1)---\left(25\right)$

$E\left[X_I\right]=\frac{1}{\mathrm{\&beta;}}\ln \left(E\right[I\left]\right)-0.5\mathrm{\&beta;}\&CenterDot;\mathrm{Var}\left(X_I\right)---\left(26\right)$

β=0.1ln10 wherein, the then portable terminal (r that receives of place, base station _k, θ _k) the chip signal to noise ratio

${\mathrm{\&gamma;}}_k=\frac{K_k}{I}=\frac{1}{{10}^{\frac{X_I-X_k}{10}}}=\frac{1}{{10}^{\frac{X_T}{10}}}---\left(27\right)$

X wherein _T=X _I-X _kBe that average is m _T, variance is σ _T ²The Gaussian stochastic variable.

E[X _T]＝m _T＝E[X _I-X _k]＝E[X _I]-E[X _k] (28)

$\mathrm{Var}\left[X_T\right]={\mathrm{\&sigma;}}_T^2=\mathrm{Var}(X_I-X_k)=\mathrm{Var}\left(X_k\right)+\mathrm{Var}\left(X_T\right)---\left(29\right)$

Portable terminal (the r that receives of base station place then _k, θ _k) the dB value of chip signal to noise ratio is

${\mathrm{\&gamma;}}_k^{\mathrm{dB}}={\left(\frac{E_c}{N_t}\right)}_k^{\mathrm{dB}}=-X_T---\left(30\right)$

Can get γ thus _k ^DBProbability density function be

$f_{{\mathrm{\&gamma;}}_k^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_k^{\mathrm{dB}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_T}\exp [-\frac{{({\mathrm{\&gamma;}}_k^{\mathrm{dB}}+m_T)}^2}{{2\mathrm{\&sigma;}}_T^2}]---\left(31\right)$

B2. up link RoT probability distribution

RoT can be expressed as

$\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}=\frac{I_{\mathrm{own}}+I_{\mathrm{oth}}+N_{0}W}{N_{0}W}$

$=\frac{(1+i_r)I_{\mathrm{own}}+N_{0}W}{N_{0}W}$

$=1+(1+i_r)\frac{\underset{j=u_1,u_2,...,u_s}{\mathrm{\&Sigma;}}{P}_j{L}_j(r_j,{\mathrm{\&theta;}}_j)}{N_{0}W}---\left(32\right)$

$=1+(1+i_r)\frac{\underset{j=u_1,u_2,...,u_s}{\mathrm{\&Sigma;}}{P}_j{K}_j{r}_j^{-l}}{N_{0}W}$

Wherein, u ₁, u ₂..., u _sRepresent each scheduled user. $K_j=e^{Y_j}$ Be logarithm normal distribution, Y _jFor average is

Standard deviation is

The Gaussian stochastic variable.K then _jAverage and variance be respectively

$E\left[K_j\right]=e^{m_{Y_j}+{\mathrm{\&sigma;}}_{Y_j}^2/2}---\left(33\right)$

$\mathrm{Var}\left(K_j\right)=\exp \{{2m}_{Y_j}+{\mathrm{\&sigma;}}_{Y_j}^2\}\&CenterDot;\left(\exp \right\{{\mathrm{\&sigma;}}_{Y_j}^2\}-1)---\left(34\right)$

Can be approximated to be the lognormal stochastic variable,

Average and variance can derive

$E\left[\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\right]=1+(1+i_r)\frac{\underset{j=u_1,u_2,...,u_s}{\mathrm{\&Sigma;}}{P}_j{r}_j^{-l}E\left[K_j\right]}{N_{0}W}---\left(35\right)$

For

$=1+(1+i_r)\frac{\underset{j=u_1,u_2,...,u_s}{\mathrm{\&Sigma;}}{P}_j{r}_j^{-l}{e}^{m_{Y_j}+{\mathrm{\&sigma;}}_{Y_j}^2/2}}{N_{0}W}$

$\mathrm{Var}\left[\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\right]=\underset{j=u_1,u_2,...,u_s}{\mathrm{\&Sigma;}}{\left[\frac{(1+i_r)}{N_{0}W}{P}_j{r}_j^{-l}\right]}^2\mathrm{Var}\left[K_j\right]---\left(36\right)$

$=\underset{j=u_1,u_2,...,u_s}{\mathrm{\&Sigma;}}{\left[\frac{(1+i_r)}{N_{0}W}{P}_j{r}_j^{-l}\right]}^2\exp \{{2m}_{Y_j}+{\mathrm{\&sigma;}}_{Y_j}^2\}\&CenterDot;\left(\exp \right\{{\mathrm{\&sigma;}}_{Y_j}^2\}-1)$

Then

The average and the variance of dB value can obtain by the Wilkinson method

$\mathrm{Var}\left({\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}\right)={\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}^2=\frac{1}{{\mathrm{\&beta;}}^2}\ln (\frac{\mathrm{Var}\left(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\right)}{E{\left[\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\right]}^2}+1)---\left(37\right)$

$E\left[{\underset{u_1,u_2,{...,u}_s}{\mathrm{RoT}}}^{\mathrm{dB}}\right]=m_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}=\frac{1}{\mathrm{\&beta;}}\ln \left(E\right[\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\left]\right)-0.5\mathrm{\&beta;}\&CenterDot;\mathrm{Var}\left(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\right)---\left(38\right)$

Wherein β=0.1ln10 can get thus

Probability density function be

$f_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}\left({\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\exp [-\frac{{({\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}-m_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}})}^2}{{2\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}^2}]---\left(39\right)$

B3. sector and the user throughput of mobile terminal locations fixedly the time

Consider weighting type dispatching algorithm.Under such dispatching algorithm, user's service priority computing formula is:

PRIORITY _i＝ω _i·r _i(t) (40)

Wherein, ω _iBe the dispatch weight that system distributes for user i, r _i(t) be the instantaneous signal-to-noise ratio of user i at moment t.For Max C/I dispatching algorithm, ${\mathrm{\&omega;}}_i^{\mathrm{dB}}=0;$ For the PF dispatching algorithm, ${\mathrm{\&omega;}}_i^{\mathrm{dB}}=-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}.$

Consider N user, its position is respectively (r _i, θ _i) (1≤i≤N), then to receive i user's chip signal to noise ratio be γ to the place, base station _i ^DB, and i user by the probability density function of base station scheduling is

$f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)=f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)\&CenterDot;\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)---\left(41\right)$

Wherein

$f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_T}\exp [-\frac{{({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_i,{\mathrm{\&theta;}}_i))}^2}{{2\mathrm{\&sigma;}}_T^2(r_i,{\mathrm{\&theta;}}_i)}]$

According to different dispatching algorithms, different Pr (UEiisschedule) value is arranged.

A) RR dispatching algorithm

Utilize the polling dispatching algorithm, when single mobile terminal was worked with different modulating coded system, spreading factor and code channel number, the probability density function that is scheduled was:

$\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)=\mathrm{Pr}(\underset{i}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\mathrm{Pr}(\underset{i,{\{i+1\}}^R}{\mathrm{RoT}}>{\mathrm{RoT}}_T)---\left(42\right)$

$+\underset{K=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{m=0}{\overset{K}{\mathrm{\&Sigma;}}}\mathrm{Pr}(\underset{{\{i-K+m\}}^R,{\{i-K+m+1\}}^R,...,{\{i+m\}}^R}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\mathrm{Pr}(\underset{{\{i-K+m\}}^R,{\{i-K+m+1\}}^R,...,{\{i+m\}}^R,{\{i+m+1\}}^R}{\mathrm{RoT}}>{\mathrm{RoT}}_T)$

Wherein,

${\left\{i\right\}}^R=\left\{\begin{array}{cc}N+i,& i<0\\ i,& 0\&le;i\&le;N\\ i-N,& i>N\end{array}\right.---\left(43\right)$

$\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)=1-Q\left(\frac{{\mathrm{RoT}}_T-m_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}{{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\right)---\left(44\right)$

$\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}>{\mathrm{RoT}}_T)=1-\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)$

Wherein, u ₁, u ₂..., u _sRepresent each scheduled user, The standard deviation (dB) that noise rise distributes, The average (dB) that noise rise distributes.

B) Max C/I dispatching algorithm

Utilize Max C/I dispatching algorithm, when single mobile terminal was worked with different modulating coded system, spreading factor and code channel number, the probability density function that is scheduled was:

$\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)=\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}\left\}\right)\&CenterDot;\mathrm{Pr}(\underset{i}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}\left({\underset{i,l}{\mathrm{RoT}}>\mathrm{RoT}}_T\right)$

$+\underset{K=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{j_1=1,j_1\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}...\underset{j_K=1,j_K>j_{K-1},j_K\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{h\&NotEqual;j_1,j_2,...,j_K}{\max }\{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}\left\}\right)\&CenterDot;\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\&CenterDot;\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K,l}{\mathrm{RoT}}>{\mathrm{RoT}}_T)---\left(45\right)$

Wherein

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}\left\}\right)=\underset{j=1,j\&NotEqual;i}{\overset{N}{\mathrm{\&Pi;}}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_j,{\mathrm{\&theta;}}_j)}{{\mathrm{\&sigma;}}_T(r_j,{\mathrm{\&theta;}}_j)}\right)]---\left(46\right)$

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{h\&NotEqual;j_1,j_2,...,j_K}{\max }\{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}\left\}\right)---\left(47\right)$

$=\underset{h\&NotEqual;j_1,j_2,...,j_K,i}{\mathrm{\&Pi;}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_h,{\mathrm{\&theta;}}_h)}{{\mathrm{\&sigma;}}_T(r_h,{\mathrm{\&theta;}}_h)}\right)]$

$\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)=1-Q\left(\frac{{\mathrm{RoT}}_T-m_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}{{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\right)---\left(48\right)$

$\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}>{\mathrm{RoT}}_T)=1-\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)$

Wherein, u ₁, u ₂..., u _sRepresent each scheduled user, The standard deviation (dB) that noise rise distributes,

The average (dB) that noise rise distributes.

C) PF dispatching algorithm

Utilize the PF dispatching algorithm, when single mobile terminal was worked with different modulating coded system, spreading factor and code channel number, the probability density function that is scheduled was:

$\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)=\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}}\left\}\right)\&CenterDot;\mathrm{Pr}(\underset{i}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\&CenterDot;\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}\left(\underset{i,l}{\mathrm{RoT}}{>\mathrm{RoT}}_T\right)$

$+\underset{K=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{j_1=1,j_1\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}...\underset{j_K=1,j_K>j_{K-1},j_K\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}=\underset{h\&NotEqual;j_1,j_2,...,j_K}{\max }\{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}}\left\}\right)\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K}{\mathrm{RoT}}>{\mathrm{RoT}}_T)---\left(49\right)$

Wherein

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}}\left\}\right)=\underset{j=1,j\&NotEqual;i}{\overset{N}{\mathrm{\&Pi;}}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_i,{\mathrm{\&theta;}}_i)}{{\mathrm{\&sigma;}}_T(r_j,{\mathrm{\&theta;}}_j)}\right)]---\left(50\right)$

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}=\underset{h\&NotEqual;j_1,j_2,...,j_k}{\max }\{{{\mathrm{\&gamma;}}_h}^{\mathrm{dB}}-\stackrel{\&OverBar;}{{{\mathrm{\&gamma;}}_h}^{\mathrm{dB}}}\left\}\right)---\left(51\right)$

$=\underset{h\&NotEqual;j_1,j_2,...,j_k,i}{\mathrm{\&Pi;}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_h,{\mathrm{\&theta;}}_h)}{{\mathrm{\&sigma;}}_T(r_h,{\mathrm{\&theta;}}_h)}\right)]$

$\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)=1-Q\left(\frac{{\mathrm{RoT}}_T-m_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}{{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\right)---\left(52\right)$

Wherein, u ₁, u ₂..., u _sRepresent each scheduled user,

The standard deviation (dB) that noise rise distributes,

The average (dB) that noise rise distributes.

$\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}>{\mathrm{RoT}}_T)=1-\mathrm{Pr}(\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)$

The place, base station received chip signal to noise ratio γ when then i user was scheduled _i ^DBThe probability that drops on the signal to noise ratio interval at MCS mode n place is

$\mathrm{Pr}\left(n\right)={\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{f}_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right){\mathrm{d\&gamma;}}_i^{\mathrm{dB}}---\left(53\right)$

Average repeat time under the MCS mode n is simultaneously

$\stackrel{\&OverBar;}{N}\left(n\right)=\frac{{\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{N}_s{f}_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right){\mathrm{d\&gamma;}}_i^{\mathrm{dB}}}{\mathrm{Pr}\left(n\right)}---\left(54\right)$

Residue frame error rate under the MCS mode n is

${\mathrm{FER}}_r\left(n\right)=\frac{{\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{\mathrm{FER}}_r\&CenterDot;f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right){\mathrm{d\&gamma;}}_i^{\mathrm{dB}}}{\mathrm{Pr}\left(n\right)}---\left(55\right)$

Then N user position is (r _i, θ _i) (i user's throughput is during 1≤i≤N)

$\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i)=W\&CenterDot;\underset{i=1}{\overset{N_{\mathrm{MCS}}}{\mathrm{\&Sigma;}}}\frac{{\log }_{2}M\left(n\right)\&CenterDot;R_c\left(n\right)\&CenterDot;N_c\left(n\right)\&CenterDot;(1-{\mathrm{FER}}_r\left(n\right))\&CenterDot;\mathrm{Pr}\left(n\right)}{\mathrm{SF}\left(n\right)\&CenterDot;\stackrel{\&OverBar;}{N}\left(n\right)}---\left(56\right)$

Wherein M (n), R _c(n), SF (n) and N _c(n) be respectively modulation index, code rate, spreading factor and code channel number under the MCS mode n.

N user position is (r _i, θ _i) (sector throughput of 1≤i≤N) is

$\mathrm{Th}(\stackrel{\&RightArrow;}{r},\stackrel{\&RightArrow;}{\mathrm{\&theta;}})=\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i)---\left(57\right)$

B4. the system of mobile terminal locations fixedly the time satisfies user's ratio

Suppose that traffic data rate requires to be R, the activation of service factor is ρ.There is N the user who activates in system, and the user throughput that provides is Th (r _i, θ _i) (i=1,2 ..., N).The requirement that the user satisfies is: user throughput is not less than the product of the traffic data rate and the activation of service factor.

Th(r _i，θ _i)≥R·ρ (58)

If satisfying the number of users of formula (58) is N ^*, then user's ratio that system is satisfied is

Percent＝N ^*/N (59)

Average sector when B5. mobile terminal locations evenly distributes and user throughput

Suppose the hexagonal serving sector structure of (1) consideration, its radius is R, and (2) N portable terminal activates simultaneously, and customer location is obeyed evenly and distributed in the sector.Suppose that normalization diameter in sector is 1, then the gross area is

With reference to accompanying drawing 2,0≤r≤1/2 A=2 π/3;

The gross area in such cases is

$S_1={\&Integral;}_0^{1/2}\frac{2\mathrm{\&pi;}}{3}\mathrm{rdr}=\frac{\mathrm{\&pi;}}{12}---\left(60\right)$

With reference to accompanying drawing 3, $1/2<r\&le;\sqrt{3}/2,$ Obtain the relation of angle B and r earlier.Utilize sine,

$\frac{r}{\sin (2\mathrm{\&pi;}/3)}=\frac{1/2}{\sin C}=\frac{1/2}{\sin (\mathrm{\&pi;}-(2\mathrm{\&pi;}/3+B))}=\frac{1/2}{\sin (2\mathrm{\&pi;}/3+B)}---\left(61\right)$

Then have $B=-\mathrm{\&pi;}/6+\arccos [\sqrt{3}/\left(4r\right)],$ The value of A is so

$A=\frac{2\mathrm{\&pi;}}{3}-2B=\frac{2\mathrm{\&pi;}}{3}-2[\frac{-\mathrm{\&pi;}}{6}+\arccos \left(\frac{\sqrt{3}}{4r}\right)]=\mathrm{\&pi;}-2\arccos \left(\frac{\sqrt{3}}{4r}\right)---\left(62\right)$

The gross area in such cases is

$S_1={\&Integral;}_{1/2}^{\sqrt{3}/2}[\mathrm{\&pi;r}-2r\arccos \left(\frac{\sqrt{3}}{4r}\right)]\mathrm{dr}=\frac{\mathrm{\&pi;}}{24}+\frac{\sqrt{3}}{8}---\left(63\right)$

With reference to accompanying drawing 4, $\sqrt{3}/2<r\&le;1,$ In like manner can get

$A=\frac{-2\mathrm{\&pi;}}{3}+2\arcsin \left(\frac{\sqrt{3}}{2r}\right)---\left(64\right)$

The gross area in such cases is

$S_1={\&Integral;}_{\sqrt{3}/2}^1[\frac{-2\mathrm{\&pi;}}{3}r+2r\arcsin \left(\frac{\sqrt{3}}{2r}\right)]\mathrm{dr}=\frac{\sqrt{3}}{4}-\frac{\mathrm{\&pi;}}{8}---\left(65\right)$

A plurality of users are uniformly distributed in the sector, and then the average throughput of sector is expressed as for all possible distribution situation is carried out statistical average

${\stackrel{\&OverBar;}{\mathrm{Th}}}_{\sec \mathrm{tor}}=({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}{\mathrm{d\&theta;}}_1{\mathrm{dr}}_1+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_1\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_1\right)\left]\right\}/2}{\mathrm{d\&theta;}}_1{\mathrm{dr}}_1+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_1\right)]\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_1\right)\left]\right\}/2}{\mathrm{d\&theta;}}_1{\mathrm{dr}}_1)$

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}{\mathrm{d\&theta;}}_2{\mathrm{dr}}_2+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_2\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_2\right)\left]\right\}/2}{\mathrm{d\&theta;}}_2{\mathrm{dr}}_2+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_2\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_2\right)\left]\right\}/2}{\mathrm{d\&theta;}}_2{\mathrm{dr}}_2)$

…

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}{\mathrm{d\&theta;}}_N{\mathrm{dr}}_N+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_N\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_N\right)\left]\right\}/2}{\mathrm{d\&theta;}}_N{\mathrm{dr}}_N+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_N\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_N\right)\left]\right\}/2}{\mathrm{d\&theta;}}_N{\mathrm{dr}}_N)$

$\mathrm{Th}(\stackrel{\&RightArrow;}{r},\stackrel{\&RightArrow;}{\mathrm{\&theta;}}){\left(\frac{8\sqrt{3}}{9}\right)}^N---\left(67\right)$

Under a this N user was evenly distributed in situation in the sector, each user's average throughput was

${\stackrel{\&OverBar;}{\mathrm{Th}}}_{\mathrm{user}}=\frac{{\stackrel{\&OverBar;}{\mathrm{Th}}}_{\sec \mathrm{tor}}}{N}---\left(68\right)$

On average satisfy user's ratio when B6. mobile terminal locations evenly distributes

A plurality of users are uniformly distributed in the sector, and then user's ratio that system is satisfied is the statistical average of all possible distribution situation, is expressed as

$\stackrel{\&OverBar;}{\mathrm{Percent}}=({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}{\mathrm{d\&theta;}}_1{\mathrm{dr}}_1+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_1\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_1\right)\left]\right\}/2}{\mathrm{d\&theta;}}_1{\mathrm{dr}}_1+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_1\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_1\right)\left]\right\}/2}{\mathrm{d\&theta;}}_1{\mathrm{dr}}_1)$

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}{\mathrm{d\&theta;}}_2{\mathrm{dr}}_2+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_2\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_2\right)\left]\right\}/2}{\mathrm{d\&theta;}}_2{\mathrm{dr}}_2+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_2\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_2\right)\left]\right\}/2}{\mathrm{d\&theta;}}_2{\mathrm{dr}}_2)$

…

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}{\mathrm{d\&theta;}}_N{\mathrm{dr}}_N+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_N\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left({4r}_N\right)\left]\right\}/2}{\mathrm{d\&theta;}}_N{\mathrm{dr}}_N+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_N\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left({2r}_N\right)\left]\right\}/2}{\mathrm{d\&theta;}}_N{\mathrm{dr}}_N)$

$\mathrm{Percent}\&CenterDot;{\left(\frac{8\sqrt{3}}{9}\right)}^N---\left(69\right)$

Wherein the system under the fixed position satisfies user's ratio Percent and sees formula (59).

Carry out considering hexagon serving sector structure when the sub-district internal and external interference is analyzed, analyze portable terminal receives chip signal to noise ratio and noise rise at the place, base station distribution situation.Based on this, obtain sector throughput, user throughput fixedly the time activating mobile terminal locations.Require and the activation of service factor for given service rate simultaneously, provide user's ratio that system satisfies.Consider user's ratio that average sector throughput, user's average throughput and system when all activated mobile terminal locations evenly distributes satisfy then in serving sector.System level is analyzed the mathematic(al) representation of multiple dispatching algorithm to the influence of systematic function, thereby has enlarged the scope of application of the inventive method embodiment; And organically combined the link layer performance, more fully analyzed the HSUPA performance.

The software of described embodiment correspondence can be stored in a computer and can store in the medium that reads.

Obtain the system of high-speed uplink packet scheduling capacity with reference to accompanying drawing 5 for embodiments of the invention embodiment, this system comprises:

The link effective data rate deriving means according to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, draws the link effective data rate under different rapid mixing re-transmission mode automatically;

Chip signal-noise ratio threshold calculation element is determined the chip signal-noise ratio threshold according to system's residue frame error rate;

Link effective data rate and chip signal to noise ratio corresponding relation are determined device, according to described link effective data rate and chip signal-noise ratio threshold, draw the corresponding relation of link effective data rate and chip signal to noise ratio;

The power system capacity deriving means according to the corresponding relation of link effective data rate and chip signal to noise ratio, utilizes described link effective data rate to obtain power system capacity.

Described link effective data rate deriving means (hereinafter in conjunction with the accompanying drawings 7 detailed description is arranged in addition)

Described power system capacity deriving means comprises:

Chip signal to noise ratio and noise rise distribution situation analytic unit at hexagon serving sector structure, are analyzed portable terminal receives chip signal to noise ratio and uplink noise lifting at the place, base station distribution situation;

The throughput calculation unit according to described signal to noise ratio and noise rise distribution situation, utilizes described link effective data rate, obtains sector throughput, user throughput fixedly the time activating mobile terminal locations;

User's ratio computing unit requires and the activation of service factor for given service rate, provides user's ratio that system satisfies;

The average throughput computing unit, user's ratio that average sector throughput, user's average throughput and the system when drawing all activated portable terminal and evenly distributing in serving sector satisfies.

Described " chip signal to noise ratio and noise rise distribution situation analytic unit " comprising:

Receive chip signal to noise ratio probability density function acquiring unit, during the fixed-site of all activated portable terminal in the regular hexagon serving sector, the probability density function that receives the chip signal to noise ratio is:

$f_{{\mathrm{\&gamma;}}_k^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_k^{\mathrm{dB}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_T}\exp [-\frac{{({\mathrm{\&gamma;}}_k^{\mathrm{dB}}+m_T)}^2}{{2\mathrm{\&sigma;}}_T^2}];$

Noise rise probability density function acquiring unit, uplink noise lifting probability density function is

$f_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}\left({\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}\right)=\frac{1}{\sqrt{2\mathrm{\&pi;}}{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\exp [-\frac{{({\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}-m_{{\underset{u_1,u_2,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}})}^2}{{2\mathrm{\&sigma;}}_{{\underset{u_1,u_2,...,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}^2}].$

Described " throughput calculation unit " comprising:

Scheduling probability density function acquiring unit, establishing the chip signal to noise ratio that the place, base station receives i portable terminal is γ _i ^DB, then i portable terminal by the probability density function of base station scheduling is

$f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)=f_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right)\&CenterDot;\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right);$

Receive the probability density function acquiring unit that the chip signal to noise ratio drops on the signal to noise ratio interval, when i portable terminal was scheduled, the place, base station received chip signal to noise ratio γ _i ^DBThe probability that drops on the signal to noise ratio interval at modulation coding mode n place is: $\mathrm{Pr}\left(n\right)={\&Integral;}_{{\mathrm{\&gamma;}}_n^{\mathrm{dB}}}^{{\mathrm{\&gamma;}}_{n+1}^{\mathrm{dB}}}{f}_{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}^s\left({\mathrm{\&gamma;}}_i^{\mathrm{dB}}\right){\mathrm{d\&gamma;}}_i^{\mathrm{dB}};$

The calculate throughput unit, N user position is (r _i, θ _i) (i user's throughput is during 1≤i≤N)

$\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i)=W\&CenterDot;\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{{\log }_{2}M\left(n\right)\&CenterDot;R_c\left(n\right)\&CenterDot;N_c\left(n\right)\&CenterDot;(1-{\mathrm{FER}}_r\left(n\right))\&CenterDot;\mathrm{Pr}\left(n\right)}{\mathrm{SF}\left(n\right)\&CenterDot;\stackrel{\&OverBar;}{N}\left(n\right)}.$

At link layer, the mathematic(al) representation of the link effective data rate when embodiments of the invention have provided given modulation coding mode, spreading factor and code channel number (formula (9)).In system layer, take into full account probability and (2) that (1) portable terminal is scheduled when calculating the portable terminal throughput and received the chip signal to noise ratio in the place, base station when portable terminal is scheduled, drop on the probability in signal to noise ratio interval, MCS mode n place.Make and estimate properer the tallying with the actual situation of result.

Embodiments of the invention also provide the method and apparatus that obtains the high-speed uplink packet scheduling effective data rate simultaneously:

With reference to accompanying drawing 6, embodiments of the invention obtain the method for high-speed uplink packet scheduling effective data rate, comprise the steps:

When initialization, set spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, the maximum retransmission that system allows, feedback information error probability and the residue frame error rate upper limit; Described feedback information error probability is generally 0.01 to 0.04, and the residue frame error rate upper limit is generally 1%; The maximum retransmission that system allows is N _MaxDescribed H-ARQ mode is for catching up with two kinds of merging and steadily increase redundancies.

According to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, under different H-ARQ modes, obtain link effective data rate;

According to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, under different H-ARQ modes, the method for obtaining link effective data rate comprises:

For spreading rate, modulation coding mode, spreading factor and the code channel number set, obtain the bit error probability P of physical layer _b

If each bit in the grouping bag has identical bit error rate, and the bit mistake is separate, then transmits frame error rate first to be

$P_e=1-{(1-P_b)}^{N_p}---\left(60\right)$

$N_p=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;\mathrm{TTI}}{\mathrm{SF}}---\left(61\right)$

Wherein, W is a spreading rate, N _pBe grouping packet length, R _cBe code rate, N _cBe code channel number.P _bBe the bit error probability, under different modulation coding modes, itself and chip signal to noise ratio γ _cCorresponding relation determine by the error rate result of physical layer.As physical layer modulation mode DPSK, when demodulation mode is differential coherence, $P_b=\frac{1}{2}{e}^{-{\mathrm{\&gamma;}}_c}.$

According to data frame transfer frame error rate P first _e, under different H-ARQ modes, obtain the merging frame error rate P after twice Frame merges _s

When the H-ARQ mode merges for catching up with

Catch up with and merge H-ARQ employing Rake receiver high specific folding, the error rate P after then the adjacent two data frame merges _sWith P _eSimilar expression formula is arranged, as long as the chip signal to noise ratio is become γ _c'=2 γ _cGet final product P _sExpression formula.Supposing equally has under the DPSK situation:

$P_b^{\&prime;}=\frac{1}{2}{e}^{-{\mathrm{\&gamma;}}_c^{\&prime;}}=\frac{1}{2}{e}^{-{2\mathrm{\&gamma;}}_c}$

$P_s=1-{(1-P_b^{\&prime;})}^{N_p}$

Steadily increase redundancy H-ARQ

The situation of twice transferring data frames before and after considering to merge.Twice data frame transfer can be equivalent to the once transmission of carrying out with the original code length of twice, 1/2 original encoding speed, promptly makes N in formula (1) _p'=2N _p, at P _bWith γ _cCorresponding relation computational process in make $R_c^{\&prime;}=\frac{R_c}{2}$ Once transmission, get final product P _sExpression formula.Also be example with DPSK:

$P_s=1-{(1-P_b)}^{N_p^{\&prime;}}$

$N_p^{\&prime;}={2N}_p=2\frac{W\&CenterDot;{\log }_2^M\&CenterDot;R_c^{\&prime;}\&CenterDot;N_c\&CenterDot;\mathrm{TTI}}{\mathrm{SF}}=\frac{W\&CenterDot;{\log }_2^M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;\mathrm{TTI}}{\mathrm{SF}}$

According to the described frame error rate of data frame transfer first P _eWith merging frame error rate P _s, obtain data frame transfer number of times probability distribution Pr (j);

After receiving terminal receives the transmitting terminal signal, need at once to send ACK (correctly receiving return signal) or NACK (mistake receive or do not have reception return signal) to transmitting terminal, whether the indication transmitting terminal is retransmitted.During calculated data frame the number of transmissions probability distribution, need to consider the desirable and imperfect two kinds of situations of feedback acknowledgment channel.

Ideal feedback

Suppose feedback acknowledgment channel ideal, promptly the feedback acknowledgment signal error probability from receiving terminal that receives of transmitting terminal is 0, the probability of the inferior successful decoded information piece of j (j＞1) then, and promptly the number of transmissions is that the probability of j is:

Pr(j)＝P _e(P _eP _s) ^j-2(1-P _eP _s) (62)

When j=1, Pr (j)=(1-P _e).

Imperfect feedback

Suppose that the feedback acknowledgment channel is imperfect, 3 kinds of situations may appear in the feedback information ACK of receiving terminal or NACK in transmission course: correct transmission, erroneous transmissions and lose.Making system allow maximum retransmission is N _Max, the probability of the inferior successful decoded information piece of j (j＞1) then, promptly the number of transmissions is that the probability of j is:

Wherein, P _fBe the probability that receiving terminal feedback information ACK or NACK lose, the probability that receiving terminal feedback information ACK loses is generally 0.01, and the probability that receiving terminal feedback information NACK loses is general 0.1; P _AeExpression receiving terminal feedback information ACK the probability of being misinformated for NACK, the span of this probability is generally 0.01-0.04; P _NeReceiving terminal feedback information NACK the probability of being misinformated for ACK, the span of this probability GenerallyBe 0.01-0.04.

According to data frame transfer number of times probability distribution Pr (j), obtain Frame average transmission number of times

$N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{jPr}\left(j\right);$

The average of the number of transmissions is

$N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{jPr}\left(j\right)---\left(64\right)$

Under desirable feedback, the average transmission number of times can directly be expressed as

$N_s=\frac{1+P_e-P_e{P}_s-P_e{\left(P_e{P}_s\right)}^{N_{\max }-1}-N_{\max }{P}_e{\left(P_e{P}_s\right)}^{N_{\max }-1}+N_{\max }{P}_e{\left(P_e{P}_s\right)}^{N_{\max }}}{1-P_e{P}_s}---\left(65\right)$

The maximum retransmission N that allows in described system _MaxIn, obtain residue frame error rate FER according to Frame average transmission number of times _r

Maximum times transmission (N is carried out in definition _Max) after the residue frame error rate

${\mathrm{FER}}_r=1-\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(j\right)---\left(66\right)$

Under desirable feedback, the residue frame error rate can directly be expressed as

${\mathrm{FER}}_r=P_e{\left(P_e{P}_s\right)}^{N_{\max }-1}---\left(67\right)$

System is to FER _rCertain requirement is arranged, must be no more than P _Loss, i.e. FER _r≤ P _Loss

According to described residue frame error rate FER _r, obtain link effective data rate EffinkRate

When system adopts M-QAM modulation, code rate is R _cTurbo code the time, link effective data rate is

$\mathrm{EfflinkRate}=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;(1-{\mathrm{FER}}_r)}{\mathrm{SF}\&CenterDot;N_s}---\left(68\right)$

Wherein W is spreading rate, R _cBe code rate, N _cBe code channel number, M is the QAM modulation index, and SF is a spreading factor, N _SBe Frame average transmission number of times, FER _rBe the residue frame error rate.

With reference to accompanying drawing 7, embodiments of the invention obtain high-speed uplink packet scheduling effective data rate device, comprising:

Apparatus for initializing is used to set spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, the maximum retransmission that system allows, feedback information error probability and the residue frame error rate upper limit;

The effective data rate calculation element, spreading rate, modulation coding mode, spreading factor and the code channel number of the physical layer of setting according to apparatus for initializing under different H-ARQ modes, are obtained link effective data rate;

Described effective data rate calculation element comprises:

Bit error probability computing unit for spreading rate, modulation coding mode, spreading factor and the code channel number that described apparatus for initializing is set, obtains the bit error probability P of physical layer _b

Transmit the frame error rate computing unit first, the bit error probability P that obtains according to described bit error probability computing unit _b, utilize formula $P_e=1-{(1-P_b)}^{N_p}$ Draw data frame transfer frame error rate P first _e

Merge the frame error rate computing unit, according to transmitting the frame error rate of the data frame transfer first P that the frame error rate computing unit obtains first _e, under different H-ARQ modes, obtain the merging frame error rate P after twice Frame merges _s

The number of transmissions probability distribution computing unit is according to the described frame error rate of data frame transfer first P _eWith merging frame error rate P _s, obtain data frame transfer number of times probability distribution Pr (j);

Average transmission number of times computing unit according to data frame transfer number of times probability distribution Pr (j), is obtained Frame average transmission number of times $N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{jPr}\left(j\right);$

Residue frame error rate computing unit, the maximum retransmission N that allows in the system that described apparatus for initializing is set _MaxIn, obtain residue frame error rate FER according to Frame average transmission number of times _r

${\mathrm{FER}}_r=1-\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(j\right)$

The effective data rate computing unit is according to described residue frame error rate FER _r, obtain link effective data rate EffinkRate

$\mathrm{EfflinkRate}=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;(1-{\mathrm{FER}}_r)}{\mathrm{SF}\&CenterDot;N_s}.$

Wherein W is spreading rate, R _cBe code rate, N _cBe code channel number, M is the QAM modulation index, and SF is a spreading factor, N _SBe Frame average transmission number of times, FER _rBe the residue frame error rate.

At link layer,, can obtain data frame transfer frame error rate first by the error rate of this layer for modulation coding mode, spreading factor and the code channel number of given physical layer; Can determine to be operated under two kinds of working methods of H-ARQ the frame error rate after twice the number frame merges according to data frame transfer frame error rate first; According to the frame error rate after data frame transfer frame error rate and twice number frames merge first, can obtain and carry out maximum times re-transmission residue frame error rate, the number of transmissions probability distribution and average the number of transmissions afterwards, thereby try to achieve the link effective data rate of modulation coding mode, spreading factor (SF) and the code channel number of given physical layer.The restriction of greatest admissible retransmission value and the error of transmission of backward channel acknowledgement frame have been considered at link layer, provide the corresponding relation of link effective data rate and chip signal to noise ratio, overcome the limitation of independent analysis link adaptation techniques and H-ARQ performance, thereby make properer the tallying with the actual situation of result.

The above; only for the preferable embodiment of the present invention, but protection scope of the present invention is not limited thereto, and anyly is familiar with those skilled in the art in the technical scope that the present invention discloses; the variation that can expect easily or replacement all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claim.

Claims (11)

1, a kind of method of obtaining high-speed uplink packet scheduling capacity is characterized in that comprising the steps:

According to spreading rate, modulation coding mode, spreading factor and the code channel number of physical layer, draw the link effective data rate EfflinkRate under different rapid mixing re-transmission mode automatically;

According to residue frame error rate FER _rDetermine the chip signal-noise ratio threshold;

According to described link effective data rate and chip signal-noise ratio threshold, draw the corresponding relation of link effective data rate and chip signal to noise ratio;

According to the corresponding relation of link effective data rate and chip signal to noise ratio, utilize described link effective data rate to obtain power system capacity;

Described corresponding relation according to link effective data rate and chip signal to noise ratio, the step of utilizing described link effective data rate to obtain power system capacity comprises:

At hexagon serving sector structure, analyze portable terminal receives chip signal to noise ratio and uplink noise lifting at the place, base station distribution situation;

According to described signal to noise ratio and noise rise distribution situation, utilize described link effective data rate, draw user throughput, sector throughput fixedly the time at all activated mobile terminal locations;

According to described user throughput, require and the activation of service factor for given service rate, the system of drawing satisfies user's ratio;

According to described sector throughput and user throughput, the average sector throughput when drawing all activated portable terminal and in serving sector, evenly distributing, each user's average throughput;

Average throughput and system according to described average sector throughput, each user satisfy user's ratio, draw average system and satisfy user's ratio;

Describedly utilize described link effective data rate, draw user throughput, sector throughput fixedly the time, comprising at all activated mobile terminal locations according to described signal to noise ratio and noise rise distribution situation:

The chip signal to noise ratio that the place, base station receives i portable terminal is γ _i ^DB, and i portable terminal by the probability density function of base station scheduling is

Wherein

During for the fixed-site of all activated portable terminal in the hexagon serving sector of rule, described i portable terminal receives the probability density function of chip signal to noise ratio,

The probability density function that Pr (UEi is scheduled) is scheduled for terminal;

The place, base station received chip signal to noise ratio γ when i portable terminal was scheduled _i ^DBThe probability that drops on the signal to noise ratio interval at MCS mode n place is

Average repeat time under the MCS mode n is

Residue frame error rate under the MCS mode n is

Then N user position is (r _i, θ _i) (i user's user throughput is during 1≤i≤N)

$\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i)=W\&CenterDot;\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{{\log }_{2}M\left(n\right)\&CenterDot;R_c\left(n\right)\&CenterDot;N_c\left(n\right)\&CenterDot;(1-{\mathrm{FER}}_r\left(n\right))\&CenterDot;\mathrm{Pr}\left(n\right)}{\mathrm{SF}\left(n\right)\&CenterDot;\stackrel{\&OverBar;}{N}\left(n\right)}$

Wherein M (n), R _c(n), SF (n) and N _c(n) be respectively modulation index, code rate, spreading factor and code channel number under the MCS mode n;

N user position is (r _i, θ _i) (sector throughput of 1≤i≤N) is $\mathrm{Th}(\stackrel{\&RightArrow;}{r},\stackrel{\&RightArrow;}{\mathrm{\&theta;}})=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i).$

2, the method for obtaining high-speed uplink packet scheduling capacity according to claim 1, it is characterized in that, described spreading rate, modulation coding mode, spreading factor and code channel number according to physical layer, the link effective data rate that draws under different rapid mixing re-transmission mode automatically comprises:

Spreading rate, modulation coding mode, spreading factor and code channel number according to setting obtain bit error probability P _b

According to described bit error probability P _b, utilize formula $P_e=1-{(1-P_b)}^{N_p}$ Draw data frame transfer frame error rate P first _e, N wherein _pBe the grouping packet length;

According to data frame transfer frame error rate P first _e, under different rapid mixing re-transmission mode automatically, obtain the merging frame error rate P after twice Frame merges _s

According to the described frame error rate of data frame transfer first P _eWith merging frame error rate P _s, obtain data frame transfer number of times probability distribution Pr (j);

According to data frame transfer number of times probability distribution Pr (j), obtain Frame average transmission number of times

$N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{jPr}\left(j\right);$

The maximum retransmission N that is allowing _MaxIn, obtain residue frame error rate FER according to Frame average transmission number of times _r

${\mathrm{FER}}_r=1-\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(j\right);$

According to described residue frame error rate FER _r, obtain link effective data rate EffinkRate

$\mathrm{EfflinkRate}=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;(1-{\mathrm{FER}}_r)}{\mathrm{SF}\&CenterDot;N_s},$

Wherein W is spreading rate, R _cBe code rate, N _cBe code channel number, M is the QAM modulation index, and SF is a spreading factor, N _SBe Frame average transmission number of times, FER _rBe the residue frame error rate.

3, the method for obtaining high-speed uplink packet scheduling capacity according to claim 2 is characterized in that:

Under the situation of ideal feedback:

Described data frame transfer number of times probability distribution Pr (j) is Pr (j)=P _e(P _eP _s) ^J-2(1-P _eP _s);

Described average transmission number of times is

$N_s=\frac{1+P_e-P_e{P}_s-P_e{\left(P_e{P}_s\right)}^{N_{\max }-1}-N_{\max }{P}_e{\left(P_e{P}_s\right)}^{N_{\max }-1}+N_{\max }{P}_e{\left(P_e{P}_s\right)}^{N_{\max }}}{1-P_e{P}_s};$

Described residue frame error rate is

${\mathrm{FER}}_r=P_e{\left(P_e{P}_s\right)}^{N_{\max }-1}.$

Under the imperfect feedback situation:

Described data frame transfer number of times probability distribution Pr (j) is

$\mathrm{Pr}\left(j\right)=\left\{\begin{array}{cc}(1-P_e)(1-P_{\mathrm{ae}}-P_f),& j=1\\ (1-P_e)(P_{\mathrm{ae}}+P_f)(1-P_{\mathrm{ae}}-P_f)+P_e(1-P_{\mathrm{ne}})(1-P_e{P}_s)(1-P_{\mathrm{ae}}-P_f),& j=2\\ \begin{array}{c}(1-P_e){(P_{\mathrm{ae}}+P_f)}^{j-1}(1-P_{\mathrm{ae}}-P_f)\\ +P_e(1-P_{\mathrm{ne}}){\left[P_e{P}_s(1-P_{\mathrm{ne}})\right]}^{j-2}(1-P_e{P}_s)(1-P_{\mathrm{ae}}-P_f)\\ +P_e(1-P_{\mathrm{ne}})(1-P_e{P}_s)(1-P_{\mathrm{ae}}-P_f){\left[P_e{P}_s(1-P_{\mathrm{ne}})\right]}^{j-3}(P_{\mathrm{ae}}+P_f)\&CenterDot;\\ \frac{1-{\left[\frac{P_{\mathrm{ae}}+P_f}{P_e{P}_s(1-P_{\mathrm{ne}})}\right]}^{j-2}}{1-\frac{P_{\mathrm{ae}}+P_f}{P_e{P}_s(1-P_{\mathrm{ne}})}},\end{array}& 3\&le;j\&le;N_{\max }-1\\ \begin{array}{c}(1-P_e){(P_{\mathrm{ae}}+P_f)}^{N_{\max }-1}\\ +P_e(1-P_{\mathrm{ne}}){\left[P_e{P}_s(1-P_{\mathrm{ne}})\right]}^{N_{\max }-2}(1-P_e{P}_s)\\ +P_e(1-P_{\mathrm{ne}})(1-P_e{P}_s){\left[P_e{P}_s(1-P_{\mathrm{ne}})\right]}^{j-3}(P_{\mathrm{ae}}+P_f)\&CenterDot;\\ \frac{1-{\left[\frac{P_{\mathrm{ae}}+P_f}{P_e{P}_s(1-P_{\mathrm{ne}})}\right]}^{N_{\max }-2}}{1-\frac{P_{\mathrm{ae}}+P_f}{P_e{P}_s(1-P_{\mathrm{ne}})}},\end{array}& j=N_{\max }\end{array}\right.$

Wherein, P _fIt is the probability that receiving terminal feedback information ACK or NACK lose; P _AeExpression receiving terminal feedback information ACK the probability of being misinformated for NACK; P _NeReceiving terminal feedback information NACK the probability of being misinformated for ACK.

4, the method for obtaining high-speed uplink packet scheduling capacity according to claim 1 is characterized in that, describedly determines that according to system residue frame error rate the chip signal-noise ratio threshold comprises: chip signal to noise ratio scope is divided into N _Mcs+ 1 nonoverlapping continuum, interval separation is designated as

When the residue frame error rate equals the packet loss P that system allows _LossThe time; According to formula

$\left\{\begin{array}{c}{\mathrm{\&gamma;}}_0=0\\ {\mathrm{\&gamma;}}_n=f(P_{\mathrm{loss}},{\mathrm{MCS}}_n)\\ {\mathrm{\&gamma;}}_{N_{\mathrm{mcs}}+1}=+\&infin;\end{array}\right.$ N=1,2 ..., N _McsDetermine chip signal-noise ratio threshold γ _n

5, the method for obtaining high-speed uplink packet scheduling capacity according to claim 4 is characterized in that, according to described link effective data rate and chip signal-noise ratio threshold, draws link effective data rate and chip signal to noise ratio γ _cCorresponding relation comprise:

Work as γ _c∈ [γ _n, γ _N+1), n=1,2 ... .N _MCSThe time, by formula $\mathrm{EfflinkRate}=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;(1-{\mathrm{FER}}_r)}{\mathrm{SF}\&CenterDot;N_s}$ Obtain the corresponding link effective data rate;

Work as γ _c∈ [γ ₀, γ ₁) time, there is not link effective data.

6, the method for obtaining high-speed uplink packet scheduling capacity according to claim 1 is characterized in that, when utilizing the polling dispatching algorithm, the probability density function that terminal is scheduled is:

$\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)=\mathrm{Pr}(\underset{i}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\mathrm{Pr}(\underset{i,{\{i+1\}}^R}{\mathrm{RoT}}>{\mathrm{RoT}}_T)$

$+\underset{K=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{m=0}{\overset{K}{\mathrm{\&Sigma;}}}\mathrm{Pr}(\underset{{\{i-K+m\}}^R,{\{i-K+m+1\}}^R,\&CenterDot;\&CenterDot;\&CenterDot;,{\{i+m\}}^R}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)(\underset{{\{i-K+m\}}^R,{\{i-K+m+1\}}^R,\&CenterDot;\&CenterDot;\&CenterDot;,{\{i+m\}}^R,{\{i+m+1\}}^R}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T),$ Wherein, RoT is a noise rise, RoT _TBe the noise rise thresholding of system's permission, Noise rise when only having user i to be scheduled; ${\left\{i\right\}}^R=\left\{\begin{array}{cc}N+i,& i<0\\ i,& 0\&le;i\&le;N\\ i-N,& i>N\end{array}\right.;$

When utilizing the max carrier to interference dispatching algorithm, the probability density function that terminal is scheduled is:

$\mathrm{Pr}\left(\mathrm{UEi\; isScheduled}\right)=\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}\left\}\right)\&CenterDot;\mathrm{Pr}\left(\underset{i}{\mathrm{RoT}<{\mathrm{RoT}}_T}\right)\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}(\underset{i,l}{\mathrm{\&Sigma;}}\mathrm{RoT}>{\mathrm{RoT}}_T)$

$+\underset{K=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{j_1=1,j_1\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}\&CenterDot;\&CenterDot;\&CenterDot;\underset{j_K=1,j_K>j_{K-1},j_K\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{h\&NotEqual;j_1,j_2,...,j_K}{\max }\{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}\left\}\right)\&CenterDot;\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K}{\mathrm{RoT}}<{\mathrm{RoT}}_T)\&CenterDot;\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K,l}{\mathrm{\&Sigma;}}\mathrm{RoT}>{\mathrm{RoT}}_T)$

Wherein, RoT is a noise rise, RoT _TBe the noise rise thresholding of system's permission,

Noise rise when only having user i to be scheduled;

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}\left\}\right)=\underset{j=1,j\&NotEqual;i}{\overset{N}{\mathrm{\&Pi;}}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_j,{\mathrm{\&theta;}}_j)}{{\mathrm{\&sigma;}}_T(r_j,{\mathrm{\&theta;}}_j)}\right)]$

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{h\&NotEqual;j_1,j_2,...,j_K}{\max }\{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}\left\}\right)$

$=\underset{h\&NotEqual;j_1,j_2,...,j_K,i}{\mathrm{\&Pi;}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_h,{\mathrm{\&theta;}}_h)}{{\mathrm{\&sigma;}}_T(r_h,{\mathrm{\&theta;}}_h)}\right)]$

$\mathrm{Pr}(\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)=1-Q\left(\frac{{\mathrm{RoT}}_T-m_{{\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}{{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\right)$

$\mathrm{Pr}(\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;,u_s}{\mathrm{RoT}}>{\mathrm{RoT}}_T)=1-\mathrm{Pr}(\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)$

Wherein, u ₁, u ₂..., u _sRepresent each scheduled user,

The standard deviation (dB) that noise rise distributes,

The average (dB) that noise rise distributes.

When the proportion of utilization fair scheduling algorithm, the probability density function that terminal is scheduled is:

$\mathrm{Pr}\left(\mathrm{UEi\; is\; scheduled}\right)=\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}\left\}\right)\&CenterDot;\mathrm{Pr}(\underset{i}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}(\underset{i,l}{\mathrm{RoT}}>{\mathrm{RoT}}_T)$

$+\underset{K=1}{\overset{N-1}{\mathrm{\&Sigma;}}}\underset{j_1=1,j_1\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}\&CenterDot;\&CenterDot;\&CenterDot;\underset{j_K=1,j_K>j_{K-1},j_K\&NotEqual;i}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}=\underset{h\&NotEqual;j_1,j_2,...,j_K}{\max }\{{\mathrm{\&gamma;}}_h^{\mathrm{dB}}\left\}\right)\&CenterDot;\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K}{\mathrm{RoT}}<{\mathrm{RoT}}_T)\&CenterDot;\underset{l\&NotEqual;i}{\mathrm{\&Pi;}}\mathrm{Pr}(\underset{i,j_1,j_2,...,j_K,l}{\mathrm{RoT}}>{\mathrm{RoT}}_T)$

Wherein, RoT is a noise rise, RoT _TBe the noise rise thresholding of system's permission, Noise rise when only having user i to be scheduled;

$\mathrm{Pr}({\mathrm{\&gamma;}}_i^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}}=\underset{j}{\max }\{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}-\stackrel{\&OverBar;}{{\mathrm{\&gamma;}}_j^{\mathrm{dB}}}\left\}\right)=\underset{j=1,j\&NotEqual;i}{\overset{N}{\mathrm{\&Pi;}}}[1-Q\left(\frac{{\mathrm{\&gamma;}}_i^{\mathrm{dB}}-m_T(r_i,{\mathrm{\&theta;}}_i)}{{\mathrm{\&sigma;}}_T(r_j,{\mathrm{\&theta;}}_j)}\right)]$

$\mathrm{Pr}({{\mathrm{\&gamma;}}_i}^{\mathrm{dB}}-\stackrel{\&OverBar;}{{{\mathrm{\&gamma;}}_i}^{\mathrm{dB}}}=\underset{h\&NotEqual;j_1,j_2,...,j_k}{\max }\{{{\mathrm{\&gamma;}}_h}^{\mathrm{dB}}-\stackrel{\&OverBar;}{{{\mathrm{\&gamma;}}_h}^{\mathrm{dB}}}\left\}\right)$

$=\underset{h\&NotEqual;j_1,j_2,...,j_k,i}{\mathrm{\&Pi;}}[1-Q\left(\frac{{{\mathrm{\&gamma;}}_i}^{\mathrm{dB}}-m_T(r_h,{\mathrm{\&theta;}}_h)}{{\mathrm{\&sigma;}}_T(r_h,{\mathrm{\&theta;}}_h)}\right)]$

$\mathrm{Pr}(\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;,u_s}{\mathrm{RoT}}\&le;{\mathrm{RoT}}_T)=1-Q\left(\frac{{\mathrm{RoT}}_T-m_{{\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;,u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}{{\mathrm{\&sigma;}}_{{\underset{u_1,u_2,\&CenterDot;\&CenterDot;\&CenterDot;u_s}{\mathrm{RoT}}}^{\mathrm{dB}}}}\right)$

Wherein, u ₁, u ₂..., u _sRepresent each scheduled user,

The standard deviation (dB) that noise rise distributes,

The average (dB) that noise rise distributes.

7, the method for obtaining high-speed uplink packet scheduling capacity according to claim 1 is characterized in that, according to described user throughput, requires and the activation of service factor for given service rate, and the system of drawing satisfies user's ratio and comprises:

Traffic data rate requires to be R that the activation of service factor is ρ; There is N the user who activates in system, and the user throughput that provides is Th (r _i, θ _i) (i=1,2 ..., N); The formula that the user satisfies is

Th(r _i，θ _i)≥R·ρ

If satisfying the number of users of following formula is N ^*, then system satisfies user's ratio and is

Percent＝N ^*/N。

8, the method for obtaining high-speed uplink packet scheduling capacity according to claim 1, it is characterized in that, described according to described sector throughput and user throughput, the average sector throughput when drawing all activated portable terminal and in serving sector, evenly distributing, each user's average throughput; Average throughput and system according to described average sector throughput, each user satisfy user's ratio, draw average system and satisfy user's ratio and comprise:

A plurality of users are uniformly distributed in the sector, and then average sector throughput is expressed as for all possible distribution situation is carried out statistical average

${\stackrel{\&OverBar;}{\mathrm{Th}}}_{\sec \mathrm{tor}}=({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}d{\mathrm{\&theta;}}_1{\mathrm{dr}}_1+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_1\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_1\right)\left]\right\}/2}d{\mathrm{\&theta;}}_{1}d{r}_1+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_1\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_1\right)\left]\right\}/2}d{\mathrm{\&theta;}}_1{\mathrm{dr}}_1)$

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}d{\mathrm{\&theta;}}_2{\mathrm{dr}}_2+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_2\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_2\right)\left]\right\}/2}d{\mathrm{\&theta;}}_{2}d{r}_2+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_2\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_2\right)\left]\right\}/2}d{\mathrm{\&theta;}}_2{\mathrm{dr}}_2)$

$\&CenterDot;\&CenterDot;\&CenterDot;$

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}d{\mathrm{\&theta;}}_N{\mathrm{dr}}_N+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_N\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_N\right)\left]\right\}/2}d{\mathrm{\&theta;}}_{N}d{r}_N+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_N\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_N\right)\left]\right\}/2}d{\mathrm{\&theta;}}_N{\mathrm{dr}}_N)$

$\mathrm{Th}(\stackrel{\&RightArrow;}{r},\stackrel{\&RightArrow;}{\mathrm{\&theta;}}){\left(\frac{8\sqrt{3}}{9}\right)}^N$

Under a this N user was evenly distributed in situation in the sector, each user's average throughput was

${\stackrel{\&OverBar;}{\mathrm{Th}}}_{\mathrm{user}}=\frac{{\stackrel{\&OverBar;}{\mathrm{Th}}}_{\sec \mathrm{tor}}}{N};$

A plurality of users are uniformly distributed in the sector, and then system satisfies the statistical average that user's ratio is all possible distribution situation, are expressed as average system and satisfy user's ratio

$\stackrel{\&OverBar;}{\mathrm{Percent}}=({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}d{\mathrm{\&theta;}}_1{\mathrm{dr}}_1+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_1\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_1\right)\left]\right\}/2}d{\mathrm{\&theta;}}_{1}d{r}_1+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_1\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_1\right)\left]\right\}/2}d{\mathrm{\&theta;}}_1{\mathrm{dr}}_1)$

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}d{\mathrm{\&theta;}}_2{\mathrm{dr}}_2+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_2\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_2\right)\left]\right\}/2}d{\mathrm{\&theta;}}_{2}d{r}_2+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_2\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_2\right)\left]\right\}/2}d{\mathrm{\&theta;}}_2{\mathrm{dr}}_2)$

$\&CenterDot;\&CenterDot;\&CenterDot;$

$\&CenterDot;({\&Integral;}_0^{1/2}{\&Integral;}_{-\mathrm{\&pi;}/3}^{\mathrm{\&pi;}/3}d{\mathrm{\&theta;}}_N{\mathrm{dr}}_N+{\&Integral;}_{1/2}^{\sqrt{3}/2}{\&Integral;}_{-\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_N\right)\left]\right\}/2}^{\{\mathrm{\&pi;}-2\arccos [\sqrt{3}/\left(4r_N\right)\left]\right\}/2}d{\mathrm{\&theta;}}_{N}d{r}_N+{\&Integral;}_{\sqrt{3}/2}^1{\&Integral;}_{-\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_N\right)\left]\right\}/2}^{\{-2\mathrm{\&pi;}/3+2\arcsin [\sqrt{3}/\left(2r_N\right)\left]\right\}/2}d{\mathrm{\&theta;}}_N{\mathrm{dr}}_N)$

$\mathrm{Percent}\&CenterDot;{\left(\frac{8\sqrt{3}}{9}\right)}^N$

Wherein, Percent=N ^*/ N.

9, a kind of system that obtains high-speed uplink packet scheduling capacity is characterized in that comprising:

The link effective data rate deriving means is used for spreading rate, modulation coding mode, spreading factor and code channel number according to physical layer, draws the link effective data rate under different rapid mixing re-transmission mode automatically;

Chip signal-noise ratio threshold calculation element is used for determining the chip signal-noise ratio threshold according to the residue frame error rate;

Link effective data rate and chip signal to noise ratio corresponding relation are determined device, are used for drawing the corresponding relation of link effective data rate and chip signal to noise ratio according to described link effective data rate and chip signal-noise ratio threshold;

The power system capacity deriving means is used for the corresponding relation according to link effective data rate and chip signal to noise ratio, utilizes described link effective data rate to obtain power system capacity;

Described power system capacity deriving means comprises:

Analysis module at regular hexagonal serving sector structure, is analyzed portable terminal receives chip signal to noise ratio and uplink noise lifting at the place, base station distribution situation;

The throughput calculation module according to described signal to noise ratio and noise rise distribution situation, is utilized described link effective data rate, draws user throughput, sector throughput fixedly the time at all activated mobile terminal locations;

System satisfies user's ratio determination module, according to described user throughput, requires and the activation of service factor for given service rate, and the system of drawing satisfies user's ratio;

The average throughput computing module, according to described sector throughput and user throughput, the average sector throughput when drawing all activated portable terminal and in serving sector, evenly distributing, each user's average throughput;

Average system satisfies user's ratio determination module, satisfies user's ratio according to described average sector throughput, each user's average throughput and system, draws average system and satisfies user's ratio;

Described throughput calculation module comprises:

The probability density function computing unit, the chip signal to noise ratio that the place, base station receives i portable terminal is γ _i ^DB, and i portable terminal by the probability density function of base station scheduling is

Wherein

During for the fixed-site of all activated portable terminal in the regular hexagonal serving sector, described i portable terminal receives the probability density function of chip signal to noise ratio, the probability density function that Pr (UEi is scheduled) is scheduled for terminal;

The place, base station received chip signal to noise ratio γ when i portable terminal was scheduled _i ^DBThe probability that drops on the signal to noise ratio interval at MCS mode n place is

Average repeat time under the MCS mode n is

Residue frame error rate under the MCS mode n is

Then N user position is (r _i, θ _i) (i user's user throughput is during 1≤i≤N)

$\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i)=W\&CenterDot;\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\frac{{\log }_{2}M\left(n\right)\&CenterDot;R_c\left(n\right)\&CenterDot;N_c\left(n\right)\&CenterDot;(1-{\mathrm{FER}}_r\left(n\right))\&CenterDot;\mathrm{Pr}\left(n\right)}{\mathrm{SF}\left(n\right)\&CenterDot;\stackrel{\&OverBar;}{N}\left(n\right)}$

Wherein M (n), R _c(n), SF (n) and N _c(n) be respectively modulation index, code rate, spreading factor and code channel number under the MCS mode n;

N user position is (r _i, θ _i) (sector throughput of 1≤i≤N) is $\mathrm{Th}(\stackrel{\&RightArrow;}{r},\stackrel{\&RightArrow;}{\mathrm{\&theta;}})=\underset{n=1}{\overset{N}{\mathrm{\&Sigma;}}}\mathrm{Th}(r_i,{\mathrm{\&theta;}}_i).$

10, the system that obtains high-speed uplink packet scheduling capacity according to claim 9 is characterized in that, described link effective data rate deriving means comprises:

Bit error probability computing unit is used for spreading rate, modulation coding mode, spreading factor and code channel number to setting, obtains bit error probability P _b

Transmit the frame error rate computing unit first, be used for the bit error probability P that obtains according to described bit error probability computing unit _b, utilize formula $P_e=1-{(1-P_b)}^{N_p}$ Draw data frame transfer frame error rate P first _e, N wherein _pBe the grouping packet length;

Merge the frame error rate computing unit, be used for transmitting the frame error rate of the data frame transfer first P that the frame error rate computing unit obtains first according to described _e, under different rapid mixing re-transmission mode automatically, obtain the merging frame error rate P after twice Frame merges _s

The number of transmissions probability distribution computing unit is used for according to the described frame error rate of data frame transfer first P _eWith merging frame error rate P _s, obtain data frame transfer number of times probability distribution Pr (j);

Average transmission number of times computing unit is used for obtaining Frame average transmission number of times according to data frame transfer number of times probability distribution Pr (j) $N_s=\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{jPr}\left(j\right);$

Residue frame error rate computing unit is used for the maximum retransmission N that is allowing _MaxIn, obtain residue frame error rate FER according to Frame average transmission number of times _r

${\mathrm{FER}}_r=1-\underset{j=1}{\overset{N_{\max }}{\mathrm{\&Sigma;}}}\mathrm{Pr}\left(j\right)$

The effective data rate computing unit is according to described residue frame error rate FER _r, obtain link effective data rate EffinkRate

$\mathrm{EfflinkRate}=\frac{W\&CenterDot;{\log }_{2}M\&CenterDot;R_c\&CenterDot;N_c\&CenterDot;(1-{\mathrm{FER}}_r)}{\mathrm{SF}\&CenterDot;N_s},$

Wherein W is spreading rate, R _cBe code rate, N _cBe code channel number, M is the QAM modulation index, and SF is a spreading factor, N _SBe Frame average transmission number of times, FER _rBe the residue frame error rate.

11, the system that obtains high-speed uplink packet scheduling capacity according to claim 9 is characterized in that, described power system capacity deriving means comprises:

Chip signal to noise ratio and noise rise distribution situation analytic unit are used at hexagon serving sector structure, analyze portable terminal receives chip signal to noise ratio and uplink noise lifting at the place, base station distribution situation;

The throughput calculation unit is used for utilizing described link effective data rate according to described signal to noise ratio and noise rise distribution situation, obtains sector throughput, user throughput fixedly the time activating mobile terminal locations;

User's ratio computing unit is used for requiring and the activation of service factor for given service rate, provides user's ratio that system satisfies;

The average throughput computing unit, user's ratio that average sector throughput, user's average throughput and the system when being used to draw all activated portable terminal and evenly distributing in serving sector satisfies.

Images