CN101227226A - Method for multi-business scheduling based on communication system real buffer memory - Google Patents

Abstract

The invention relates to a method for modulating multi-service, which is based on the actual cache amount of a communication system and comprises: firstly calculating the weight priority of each user who will be modulated only according to the actual cache amount without differentiating the weight priority of each user who will be modulated, distributing sub-channels of the communication system to each user who will be modulated according to the weight priority, then judging if all communication channels of the communication system are distributed to users who will be modulated with high professional work priority according to the professional work priority of each user which will be modulated or not, if all communication channels of the communication system are not distributed to users which will be modulated with high professional work priority, further judging if the users with high professional work priority are waiting for distributing communication channels or not, if all communication channels of the communication system are distributed to users which will be modulated with high professional work priority, redistributing the communication channels which are corresponded to the users which will be modulated and are distributed with the communication channels with low professional work priority and low weight priority to the users which will be modulated with high professional work priority and are waiting for distributing communication channels, thereby the diversity efficiency of the users can be intensified, the filling probability can be reduced, the increase of system throughput can be realized, and multi-service treating principles can be satisfied.

Description

Carry out the method for multi-business scheduling based on communication system real buffer memory

Technical field

The present invention relates to a kind of method of carrying out multi-business scheduling based on communication system real buffer memory.

Background technology

In recent years, OFDM inserts (OFDMA:Orthogonal frequency division multiple access) technology owing to possess and can make full use of multiple diversity effect, can eliminate in the sub-district multiple advantage such as interference, and by international telecommunication standardization body---third generation affiliate (3GPP) is decided to be the descending air interface technologies of future mobile communication system.

The OFDM access technology is the synthetic subchannel of a plurality of subcarriers and gives a user at every turn, a plurality of subchannels composition channel sets, and a plurality of users select different subchannels according to channel conditions with business demand in channel set.Obviously, how correct allocated sub-channels is the key that influences the OFDMA systematic function.Therefore, people also more and more pay attention to the dispatching technique of specializing in channel allocation.

And the main target of dispatching technique is exactly when satisfying QoS of customer (QoS) requirement and fairness, improves throughput of system as much as possible.Up to the present, people have proposed a variety of dispatching algorithms at radio communication, and wherein, algorithm can provide the highest throughput of system to max carrier to interference (max C/I), but fairness is the poorest; Roulette (RR:Round Robin) algorithm is given each user's allocated channel successively, fairness is preferably also the simplest, so become a benchmark algorithm of dispatching algorithm, but owing to do not consider characteristics of radio channels and user's actual need, so throughput of system is lower; Direct ratio justice (PF:Proportional Fair) algorithm that Holtzman proposes is owing to taken into account the demand of throughput and fairness two aspects, and overall performance is best, so become the main flow dispatching technique of radio communication very soon.At present, the researcher has proposed some again and has improved algorithm at some defectives of conventional P F algorithm.For example, Shin S joins delay restrictive condition to adapt to the transmission of real-time media in the PF algorithm, Patrick Svedman adapts to the different rate requirement of user by introducing a fairness adjustment parameter, Hsiao-ChiangChuang is divided into two class business to the user area: real-time and non real-time, and utilize user's real buffer memory to calculate the wait time delay, what time delay was big distributes earlier, if the non real-time user then also needs to carry out two times scheduling with conventional P F algorithm.

These improved PF dispatching algorithms are owing to being academic article mainly, so have do not conform to the actual conditions at 2.The one, these algorithms all are based on subcarrier to be considered, then is based on subchannel in the reality and dispatches.Like this,, when user's real buffer memory is very little, will produce a lot of fillings, cause bigger waste owing to be the subchannel that a plurality of subcarriers are formed.So should in dispatching algorithm, consider the influence of user's real buffer memory; The 2nd, these algorithms do not consider that multiservice environment or multi-service treatment principle do not conform to the actual conditions.Existing clearly definition in the relevant criterion of 3GPP, must carry out channel allocation by service priority, after promptly having served high-priority service, serve low priority traffice again, and top these algorithms are not considered the requirement of this respect, so press the algorithm of Hsiao-Chiang Chuang, the phenomenon that ftp business ties up the VOIP business can appear, and this is flagrant in practice.

Therefore, in practice, people still adopt traditional PF algorithm elder generation schedules high priority business, schedules low priority business more usually.Like this, but user that data volume very little very high and some service priority are lower but when user that data volume is bigger coexists when some service priority, as VOIP user and FTP user, will the channel that VOIP user occupies earlier appear because of considering the real buffer memory size, but concerning VOIP user, taking very good channel very wastes, because it does not have too much data to transmit, and the big FTP user of data volume only can occupy poor channel, and this will reduce the throughput of system.

In sum, how to solve that many shortcomings that prior art exists are real to have become the technical task that those skilled in the art need to be resolved hurrily.

Summary of the invention

The object of the present invention is to provide and a kind ofly carry out the method for multi-business scheduling, fill probability, realize the raising of throughput of system, can also satisfy the multi-service treatment principle simultaneously to strengthen the multi-user diversity effect, to reduce based on communication system real buffer memory.

In order to achieve the above object, the invention provides a kind of method of carrying out multi-business scheduling based on communication system real buffer memory, it comprises: 1) do not distinguish and respectively treat the dispatched users service priority, only calculate the described weight priority of respectively treating dispatched users based on the described real buffer memory of dispatched users of respectively treating; 2) adopt the PF algorithm that each channel allocation of described communication system is given according to the described weight priority of respectively treating dispatched users and treat dispatched users accordingly; 3) according to the service priority of respectively treating dispatched users judge all channels of described communication system whether all distributed to service priority high treat dispatched users; 4) when all channels of described communication system be not all distribute to service priority high treat dispatched users the time, judge whether that the dispatched users for the treatment of that service priority is high in addition waiting for allocated channel; 5) if there is the high dispatched users for the treatment of of service priority waiting for allocated channel, with service priority low and weight priority is low be assigned channel treat the pairing channel of dispatched users redistribute to described wait for allocated channel and service priority high treat dispatched users.

Wherein, basis in the described step 1) $P_i\left(t\right)=\frac{{\mathrm{DRC}}_i\left(t\right)}{\stackrel{\‾}{R_i\left(t\right)}}$ Calculate the weight priority of respectively treating dispatched users, wherein $\stackrel{\‾}{R_i\left(t\right)}=(1-\frac{1}{t_c})*\stackrel{\‾}{R_i(t-1)}+\frac{1}{t_c}*{\mathrm{DRC}}_i(t-1),$ And work as DRC _i(t)≤R _{Buffer, i}(t) time, DRC _i(t)=DRC _i(t), work as DRC _i(t)＞R _{Buffer, i}(t) time, DRC _i(t)=R _{Buffer, i}(t), DRC _i(t) for treating the channel speed of dispatched users i, R _{Buffer, i}(t) be the described real buffer memory for the treatment of dispatched users i, R _i(t) be the described Mean Speed for the treatment of dispatched users i, this Mean Speed is with t _cLong time window upgrades.

In sum, method of carrying out multi-business scheduling based on communication system real buffer memory of the present invention is at first carried out the channel allocation first time according to the weight priority of respectively treating dispatched users, and then will the high dispatched users of respectively treating carry out the channel allocation second time according to its service priority again at the service priority that do not obtain dispatching in the channel allocation first time, can strengthen the multi-user diversity effect thus, reduce and fill probability, realize the raising of throughput of system, can also satisfy the multi-service treatment principle simultaneously.

Description of drawings

Fig. 1 is the operating process schematic diagram that carries out the method for multi-business scheduling based on communication system real buffer memory of the present invention.

Fig. 2 compares schematic diagram for throughput of system.

Fig. 3 is that FTP user's channel allocation fairness compares schematic diagram.

Embodiment

See also Fig. 1, in method of carrying out multi-business scheduling based on communication system real buffer memory of the present invention, execution in step S10 at first, do not distinguish and respectively treat the dispatched users service priority, only calculate the described weight priority of respectively treating dispatched users according to the real buffer memory of respectively treating dispatched users, it is according to formula $P_i\left(t\right)=\frac{{\mathrm{DRC}}_i\left(t\right)}{\stackrel{\‾}{R_i\left(t\right)}}$ Calculate the weight priority of respectively treating dispatched users, wherein $\stackrel{\‾}{R_i\left(t\right)}=(1-\frac{1}{t_c})*\stackrel{\‾}{R_i(t-1)}+\frac{1}{t_c}*{\mathrm{DRC}}_i(t-1),$ And work as DRC _i(t)≤R _{Buffer, i}(t) time, DRC _i(t)=DRC _i(t), work as DRC _i(t)＞R _{Buffer, i}(t) time, DRC _i(t)=R _{Buffer, i}(t), DRC _i(t) for treating the channel speed of dispatched users i, R _{Buffer, i}(t) be the described real buffer memory for the treatment of dispatched users i, R _i(t) be the described Mean Speed for the treatment of dispatched users i, this Mean Speed is with t _cLong time window upgrades, for example, if remain dispatched users A and B in the described communication system, described communication system has two channels of x1, x2, take the x1 channel if treat dispatched users A, through calculating its weight priority a1 is 1, take the x2 channel if treat dispatched users A, through calculating its weight priority a2 is 3, taking the x1 channel if treat dispatched users B, is 2 through calculating its weight priority b1, takies the x2 channel if treat dispatched users B, through calculating its weight priority b2 is 2, then execution in step S11.

In step S11, adopt the PF algorithm that each channel allocation of described communication system is given according to the described weight priority of respectively treating dispatched users and treat dispatched users accordingly, for example, according to calculating resulting each weight priority for the treatment of dispatched users A and B, the x1 channel allocation given treat dispatched users B, the x2 channel allocation to treating dispatched users A, is followed execution in step S12.

In step S12, according to the service priority of respectively treating dispatched users judge all channels of described communication system whether all distributed to service priority high treat dispatched users, if finish, otherwise execution in step S13, in the present embodiment, treat that dispatched users B is the FTP user, its service priority is rudimentary.

In step S13, judge whether to also have the high dispatched users for the treatment of of service priority waiting for allocated channel, if execution in step S14, otherwise finish.For example, be still waiting dispatched users c in the described communication system, it is the voip user, and obviously service priority is senior.

In step S14, with service priority low and weight priority is low be assigned channel treat the pairing channel of dispatched users redistribute to described wait for allocated channel and service priority high treat dispatched users, therefore, having distributed to the x1 channel for the treatment of dispatched users B redistributes to treating dispatched users c, in addition, as a plurality of service priority for the treatment of dispatched users that have been assigned channel all be lower than wait for allocated channel and service priority high treat dispatched users, usually the minimum and pairing channel priority allocation of dispatched users for the treatment of that be assigned subchannel of weight priority gives the grade of service high unallocated user to subchannel, then return the judgement of carrying out a new round among the step S12 again again.

Below will further specify and of the present inventionly carry out the superior function of the method for multi-business scheduling based on communication system real buffer memory, at first hypothesis has 30 users, is divided into two kinds of business of VOIP and FTP.With adopting the present invention and existing two kinds of dispatching methods commonly used is that RR (VOIP)+PF (FTP) and PF (VOIP)+PF (FTP) compare, wherein, RR (VOIP)+PF (FTP) refers to and uses PF algorithmic dispatching FTP user again with RR algorithmic dispatching VOIP user earlier, PF (VOIP)+PF (FTP) uses PF algorithmic dispatching FTP user again with PF algorithmic dispatching VOIP user earlier, two kinds of algorithms all are after having dispatched VOIP user earlier, dispatch the FTP user again, simulated conditions as shown in Table 1:

Table 1 is a simulated conditions

Parameter	Numerical value
		System bandwidth
	5 megahertzes
		Cell diameter	1500 meters
Scene	The suburb is grand
		Number of users	30
The contained sub-carrier number of every subchannel	25
		Scheduling duration (TTI)	0.5 millisecond
Business model	VOIP，FTP
		Modulation coding mode	Quadriphase keying (QPSK) (R=1/3,1/2,2/3,3/4,4/5), 16QAM (R=1/2,2/3,3/4,4/5)
Mix request retransmission (HARQ)	Do not have
		The thermal noise spectrum density	-174dBm/Hz
The emulation duration	20000 TTI

See also Fig. 2 and Fig. 3, wherein, Fig. 2 compares schematic diagram for throughput of system, and transverse axis is the number of users of VOIP, mostly is 9 VOIP users most, and total number of users is 30, so this moment, the FTP user was 21.The longitudinal axis is a throughput, and unit is a kilobits/second.As can be seen from Figure, the present invention is under any circumstance than existing two kinds of method throughput height.Whether Fig. 3 is that FTP user's channel allocation fairness compares schematic diagram, represent the chance of each FTP user's allocated channel impartial.Transverse axis is the number of users of FTP, and the longitudinal axis is the fairness of channel allocation, is up to 1.As can be seen from Figure, the present invention is also good than existing two kinds of method outlines on fairness.

In sum, method of carrying out multi-business scheduling based on communication system real buffer memory of the present invention reduces and fills probability, when having guaranteed user fairness by improving the multi-user diversity effect, improve throughput of system, and satisfied the multi-service treatment principle.

Claims (2)

1. one kind is carried out the method for multi-business scheduling based on communication system real buffer memory, it is characterized in that comprising step:

1) only calculates the described weight priority of respectively treating dispatched users based on the real buffer memory of respectively treating dispatched users;

2) adopt the PF algorithm that each channel allocation of described communication system is given according to the described weight priority of respectively treating dispatched users and treat dispatched users accordingly;

3) according to the service priority of respectively treating dispatched users judge all channels of described communication system whether all distributed to service priority high treat dispatched users;

4) when all channels of described communication system be not all distribute to service priority high treat dispatched users the time, judge whether that the dispatched users for the treatment of that service priority is high in addition waiting for allocated channel;

5) if there is the high dispatched users for the treatment of of service priority waiting for allocated channel, with service priority low and weight priority is low be assigned channel treat the pairing channel of dispatched users redistribute to described wait for allocated channel and service priority high treat dispatched users.

2. as claimed in claim 1ly carry out the method for multi-business scheduling, it is characterized in that described step 1) basis based on real buffer memory $P_i\left(t\right)=\frac{\mathrm{DR}{C}_i\left(t\right)}{\stackrel{\‾}{R_i\left(t\right)}}$ Calculate the priority of respectively treating dispatched users, wherein $\stackrel{\‾}{R_i\left(t\right)}=(1-\frac{1}{t_c})*\stackrel{\‾}{R_i(t-1)}+\frac{1}{t_c}*\mathrm{DR}{C}_i(t-1),$ And work as DRC _i(t)≤R _{Buffer, i}(t) time, DRC _i(t)=DRC _i(t), work as DRC _i(t)＞R _{Buffer, i}(t) time, DRC _i(t)=R _{Buffer, i}(t), DRC _i(t) for treating the channel speed of dispatched users i, R _{Buffer, i}(t) be the described real buffer memory for the treatment of dispatched users i, R _i(t) be the described Mean Speed for the treatment of dispatched users i, this Mean Speed is with t _cLong time window upgrades.

Images