CN1885748A - Mixed service resource distributing method for OFDM system - Google Patents

Mixed service resource distributing method for OFDM system Download PDF

Info

Publication number
CN1885748A
CN1885748A CN 200610088133 CN200610088133A CN1885748A CN 1885748 A CN1885748 A CN 1885748A CN 200610088133 CN200610088133 CN 200610088133 CN 200610088133 A CN200610088133 A CN 200610088133A CN 1885748 A CN1885748 A CN 1885748A
Authority
CN
China
Prior art keywords
user
packet
subcarrier
time slot
queue
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN 200610088133
Other languages
Chinese (zh)
Other versions
CN100512055C (en
Inventor
侯华
周武旸
朱近康
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
University of Science and Technology of China USTC
Original Assignee
University of Science and Technology of China USTC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by University of Science and Technology of China USTC filed Critical University of Science and Technology of China USTC
Priority to CNB2006100881334A priority Critical patent/CN100512055C/en
Publication of CN1885748A publication Critical patent/CN1885748A/en
Application granted granted Critical
Publication of CN100512055C publication Critical patent/CN100512055C/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Mobile Radio Communication Systems (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The hybrid business resource allocation method in FDM system comprises: presenting a service quality satisfaction level criterion based on optimal physical and data linkage layers; with this criterion, according to active user number, server quality request and received service utility function value in every time slot, deciding user PRI, and allocating adaptively sub-carrier to user with high PRI. This invention can meet user service request and ensures fairness.

Description

A kind of mixed service resource distributing method of ofdm system
Technical field:
The invention belongs to OFDM (OFDM) mobile communication technology field, particularly the mixed service resource distributing method in the ofdm communication system.
Background technology:
Traditional OFDM resource allocation research concentrates on Bit Allocation in Discrete and two aspects of power division, " international electronics communicate by letter the selected topic magazine " (IEEE J.on Select Areas.Commun for example with The Institution of Electrical Engineers, Volume 17, No 10,1999, pp 1747-1758) Multi User Adaptive subcarrier, bit and the power distribution algorithm mentioned in.Because these algorithms consider that in assigning process the professional stochastic behaviour that arrives, Subscriber Queue state, user are to service quality (Quality of Service, requirement QoS) etc., thereby be difficult to provide satisfied service for the user.
" international conference of Communications Society of U.S. electric electronic engineering NAB National Association 0f Broadcasters tissue " (IEEE Communications SocietyIEEE International Conference, Volume 1,2004, the method of the mixing resource allocation of mentioning pp58-62), owing to use index dispatching criterion and direct ratio equity criterion to calculate the utility function of real time business and non-real-time service respectively, therefore do not provide the dispatching algorithm that not only is applicable to real time business but also is applicable to non-real-time service; In addition, this method hypothesis that can not be dropped because of time delay the data traffic packets also is irrational.
Summary of the invention:
The present invention proposes a kind of resource allocation methods of ofdm system, considered that the professional stochastic behaviour that reaches, requirement that professional equity treats time delay, Subscriber Queue state, user are to the restriction to Subscriber Queue length of the requirement of service quality and system, and not only be applicable to real time business but also be applicable to non-real-time service, not only the requirement of user can be satisfied preferably, and the fairness of service quality satisfaction between the user can be guaranteed service quality.
The mixed service resource distributing method of ofdm system of the present invention comprises scheduler module and adaptive subcarrier and bit associating distribution module; The base station provides the formation of finite length for each user, queue length is the peak value arrival rate of grouping and the product of sustainable maximum wait time delay, the base station is also set up for each user's service queue and is safeguarded that label, label write down corresponding user identifier, packet identifier, the subcarrier identifier that the user is assigned to, the number of packet that accumulative total arrives in the Subscriber Queue, the number of packet that user's accumulative total is lost, the number of packet of the mistake that user's accumulative total receives, number of packet in active user's formation, the wait time delay of head of the queue grouping, the sustainable maximum wait time delay of dividing into groups, sustainable maximum packet loss divides into groups, divide into groups sustainable maximum Packet Error Ratio and the grouping the sustainable largest packet proportion of goods damageds; The base station is input to each user independently in the formation with the packet of business, and the SO service order that divides into groups in the formation is First Come First Served; The base station end carries out channel estimating by the user of number of packet non-zero in the broadcast channel notification queue; The user reports the erroneous packet quantity that receives in channel estimation results and the last resource allocation process to the base station by the feedback channel of zero defect; The every time slot of the resource allocation algorithm of base station side is carried out once, only gives a user, the equal subsystem total emission power of each subcarrier with a subcarrier allocation in each assigning process; In the initial moment of each time slot, scheduler module is user's ordering, and the user identifier sequence after will sort and the grouping in the Subscriber Queue offers adaptive subcarrier and bit is united distribution module; Adaptive subcarrier and bit associating distribution module earlier according to the channel condition information of user feedback with user's subcarrier according to the gain sort descending, obtain each user's subcarrier gains descending series, user identifier sequence that provides according to scheduler module and the grouping information in the Subscriber Queue are the packet allocation subcarrier in the Subscriber Queue again, and the number of packet that each user can be launched feeds back to scheduler module;
It is characterized in that:
Described scheduler module produces user's utility function descending series, comprising: initialization unit initial moment of each time slot according on once the resource allocation feedback that finishes back subcarrier and bit associating distribution module abandon of the grouping of wait time delay more than or equal to this sustainable maximum wait time delay of dividing into groups, the number of packet of the mistake that receives in the last resource allocation process according to user feedback is added up the number of packet of the mistake that each user accumulative total receives, the subcarrier identifier that the user is assigned to carries out zero clearing, for new packet identifier is set up in the grouping that arrives in the last time slot, add up the number of packet that accumulative total arrives in each Subscriber Queue, add up the number of packet that each user's accumulative total is lost, the remaining number of packet of statistics Subscriber Queue, add up the remaining total number packets of this time slot Subscriber Queue, write down the wait time delay of each user's head of the queue grouping and add up the total number of users that is in state of activation in this time slot sytem; After the initialization, the satisfaction factor calculating unit is according to formula S k [ i ] = PL R k [ i ] / PLR k max Calculate the satisfaction factor of each user k, PLR in the formula k MaxThe sustainable largest packet proportion of goods damageds of expression user k, PLR k[i] expression user k is at the packet loss rate of the initial moment statistics of i time slot, PLR k[i] can use equation PL R k [ i ] = PLN k [ i ] + PEN k [ i ] PAN k [ i ] Estimate, wherein PLN k[i], PEN k[i] and PAN k[i] represents the user k packet count that accumulative total is lost before i time slot, the packet count of the mistake that receives and the total number packets of arrival respectively; The satisfaction factor that the utility function computing unit calculates according to previous step is according to formula U k CL [ i ] = u k R k cur [ i ] R k aver [ i ] exp ( S k [ i ] - S [ i ] ‾ 1 + S [ i ] ‾ ) Calculate the utility function of the user k of number of packet non-zero in each formation, R in the formula k CurThe maximum data transmission rate that can support at i time slot of [i] expression user k prediction, R k AverThe average data transfer rate of [i] expression user k before i time slot used formula R k aver [ i + 1 ] = ( 1 - 1 t c ) R k aver [ i ] + 1 t c R k cur [ i ] Upgrade, wherein t cThe length of expression sliding time window is not given user k, then R if data are launched at i time slot in the base station k Aver[i+1] uses R k Cur[i]=0 is upgraded; u kUse u k = - log ( PD R k max ) / D k max Calculate, wherein PDR k MaxAnd D k MaxRepresent sustainable maximum packet loss of user k and packet delay respectively, S [ i ] ‾ = 1 K Σ k = 1 K S k [ i ] , Be in the total number of users of state of activation when wherein K is illustrated in i in the slot sytem; After calculating end, sequencing unit obtains user's utility function descending series with user's utility function sort descending, enters adaptive subcarrier and bit associating assigning process then; Adaptive subcarrier and bit associating distribution module comprises: the subcarrier sequencing unit is according to the channel condition information of user feedback, and the user's of number of packet non-zero subcarrier according to the gain sort descending, is obtained each user's subcarrier gains descending series; The head of the queue user of subcarrier and power associating allocation units from user's utility function descending series begins to be followed successively by the idle sub-carrier that this user's subcarrier gains of packet allocation in the Subscriber Queue is passed sequence, write down the identifier of the subcarrier that each user is assigned to, calculate record also and distribute to the bit number of the emission of these subcarriers, do not divide into groups to be launched or system's idle sub-carrier quantity is zero up to this user; If system also has idle sub-carrier and also has the packet awaits emission, continue as the idle sub-carrier in this user's subcarrier gains descending series of packet allocation in the next Subscriber Queue of the low one-level of user's utility function descending series medium priority, need be launched up to grouping not, or system's idle sub-carrier number is zero; Distribute after the end, the statistics feedback unit writes down the result of this resource allocation, and the number of packet that each user can launch is fed back to the initialization unit of scheduler module.
The operation principle of described scheduler module is as follows:
Because different business has different requirements to time delay, packet loss and Packet Error Ratio, so their the sustainable largest packet proportion of goods damageds also may be different.Consider from the angle of fairness, for dissimilar business, the service quality of using professional the poorest sustainable service quality that this business has been received is made the factor that normalization obtains afterwards, embodied professional satisfaction to the service that received, the present invention is referred to as the service quality satisfaction factor.And will be called service quality satisfaction fairness criterion based on the fairness criterion of this factor.
The service quality satisfaction factor can be expressed as:
S k [ i ] = PLR k [ i ] / PL R k max - - - ( 1 )
Wherein, PLR k[i] expression user k is in the packet loss rate of the initial moment statistics of i time slot, and it calculates shown in equation (2); PLR k MaxThe sustainable largest packet proportion of goods damageds of expression user k, it calculates as shown in Equation (3);
PLR k [ i ] = PLN k [ i ] + PEN k [ i ] PAN k [ i ] - - - ( 2 )
Wherein, PLN k[i], PEN k[i] and PAN k[i] represents the user k packet count that accumulative total is lost before i time slot, the packet count of the mistake that receives and the total number packets of arrival respectively.
PLR k max = 1 - ( 1 - PER k max ) ( 1 - PDR k max ) - - - ( 3 )
Wherein, PER k MaxAnd PDR k MaxBe respectively sustainable Packet Error Ratio and because the excessive packet loss that causes of grouping time-delay that brings by wireless channel of user k.
The service quality satisfaction factor not only is applicable to real time business but also be applicable to non-real-time service.It has embodied the satisfaction of the service that the user provides system on the one hand, has embodied the fairness of service quality between the user on the other hand.The user that the service quality satisfaction factor is more little is satisfied more to the service that system provides; Otherwise the user is dissatisfied more to the service that system provides.The difference of the service quality satisfaction factor between a plurality of users is more little, has reflected that system is fair more to the service quality that different user provides; Otherwise system is unfair more to the service quality that provides between the different user.Be in the system at center with user QoS, waiting for that with using the user time delay or user's average throughput compare as the criterion of system's fairness, using service quality satisfaction fairness criterion more reasonable.Because, use the user to wait for the criterion of time delay as system's fairness, can be counted as and ignore erroneous transmissions that wireless channel brings fairness criterion the user performance influence; Use the criterion of user's average throughput as system's fairness, can be counted as the hypothesis user grouping does not have the fairness criterion of waiting for delay constraint and Subscriber Queue endless.They all are the special cases of service quality fairness criterion.
Based on service quality satisfaction fairness criterion, resource allocation methods of the present invention is defined as user's utility function
U k CL [ i ] = Δ u k R k cur [ i ] R k aver [ i ] exp ( S k [ i ] - S [ i ] ‾ 1 + S [ i ] ‾ ) - - - ( 4 )
Wherein, R k CurThe maximum data transmission rate that can support at i time slot of [i] expression user k prediction; u kBe weight coefficient, it calculates as shown in Equation (5); S[i] be in state of activation in the expression system the user before i time slot to the average of system's satisfaction, it calculates as shown in Equation (6); R k AverThe average data transfer rate of [i] expression user k before i time slot, it calculates as shown in Equation (7);
u k = - log ( PDR k max ) / D k max - - - ( 5 )
Wherein, D k MaxThe sustainable largest packet time delay of expression user k;
S [ i ] ‾ = 1 K Σ k = 1 K S k [ i ] - - - ( 6 )
Be in the total number of users of state of activation when wherein, K is illustrated in i in the slot sytem;
R k aver [ i + 1 ] = ( 1 - 1 t c ) R k aver [ i ] + 1 t c R k cur [ i ] - - - ( 7 )
Wherein, t cThe length of expression sliding time window; If launching data at i time slot, the base station do not give user k, then
R k cur [ i ] = 0 .
t cSize reflected that the user is not to receiving the degree of restraining oneself of data for a long time; Window is long more, and the stand-by period that the user can restrain oneself is long more; Otherwise the time that can restrain oneself is short more.
Dispatching algorithm is only calculated user's utility function in the zero hour of each time slot, and user's data can arrive in any time in the time slot, so the grouping that arrives in this time slot only can influence the utility function that the next time slot of this user calculates.Because the present invention has considered the stochastic behaviour that the data link layer user grouping arrives, therefore,, have following three kinds of reasons if some users do not receive the data of base station at i time slot:
There is not packet etc. to be launched in the Subscriber Queue;
The medium armed packet count of Subscriber Queue is non-vanishing, but this user does not have to obtain to choose the chance of subcarrier;
Data deficiencies in the Subscriber Queue is to constitute a grouping; For example, when speech business is in state of activation, data reach with fixed rate 64kbps, each grouping is made of 1280bits, every millisecond of renewal of scheduler once, so 1ms in the time speech business arrive 64bit, be not enough to constitute a grouping, the user might divide into groups and need be launched at interval in several scheduling times like this.
First user that resource allocation methods of the present invention is chosen is
k = arg max j = 1 , L , k U j CL [ i ] - - - ( 8 )
From formula (4), formula (7) and formula (8) as can be seen, if a user's wait time delay is bigger, its packet loss and packet loss rate be all than higher, thereby make this quality of services for users satisfaction factor also than higher, will improve user's utility function; If a user channel quality is relatively poor, its packet loss and Packet Error Ratio are also higher, can improve this user's utility function equally.
Like this, described scheduler module can obtain user's utility function descending series { U k[i] } K=1 K,  k '<k wherein, k ', there is U in k ∈ [1, K] K '[i]>U k[i].
Described adaptive subcarrier and bit associating allocation algorithm, according to remaining number of packet, bit number adaptively in the channel condition information of user feedback, user's sequencing information that dispatching algorithm provides and the Subscriber Queue for carrying on user's allocation of subcarriers and each subcarrier.Target is that under the condition of given each subcarrier transmitting power, use adaptive modulation technology optimization system throughput guarantees that simultaneously the Packet Error Ratio of each user's statistics is no more than the sustainable maximum Packet Error Ratio of this user.Its operation principle is as follows:
Adaptive subcarrier and bit associating distribution module are user's utility function descending series { U earlier k[i] } K=1 KIn head of the queue user allocation of subcarriers, when this user does not have packets remaining to be launched, continue as next user's allocation of subcarriers of the low one-level of priority, need be launched or system does not have idle sub-carrier up to grouping not.Advantage is, can improve the satisfaction of the current unsatisfied user of service that system is provided to system.
The received signal to noise ratio of receiving terminal can use following formula to describe:
f ( b i , n ) = Σ k = 1 K P total × | h i , n k | 2 N × σ n 2 × A i , n k - - - ( 9 )
Wherein, P TotalIt is the base station gross power; A i , n k = 1 Expression user k i time slot allocation to n subcarrier; Because the recipient can't detect a plurality of user's data from same subcarrier simultaneously, so for the norator carrier wave n, when A i , n k = 1 The time, have A i , n k ′ = 0 ,  k ' ≠ k k, k ' ∈ [1, K]; h I, n kExpression user k is in the amplitude of the channel gain on n subcarrier of the initial moment estimation of i time slot; σ n 2Be the noise power on each subcarrier, b I, nBe illustrated in the information bit that n subcarrier of i time slot carries, it calculates as shown in Equation (10):
b i , n = Δf log 2 ( 1 + f ( b i , n ) Γ ) - - - ( 10 )
Wherein, for the MQAM modulation, Γ can use
Figure A20061008813300087
Approximate.
Compared with prior art, because the present invention has considered the restriction to Subscriber Queue length of the professional stochastic behaviour that reaches, the requirement that professional equity treats time delay, professional requirement to packet loss, professional requirement, Subscriber Queue state and system to Packet Error Ratio in resource allocation process, stride a layer criterion by use, a kind of service quality satisfaction fairness criterion has been proposed, this criterion not only is applicable to real time business but also be applicable to non-real-time service, is a kind of layer fairness criterion of striding based on physical layer and data link layer joint optimization.According to analysis to this criterion, the resource allocation methods that the present invention proposes, use is suitable for the utility function sign user's of mixed service priority, be the more unsatisfied user of the service that system is provided and give higher priority, and adaptive subcarrier and bit associating distribution module is preferably the high user's allocation of subcarriers of grade.So the present invention not only can satisfy the requirement of user to service quality preferably, and can guarantee the fairness of service quality satisfaction between the user.
Description of drawings:
Fig. 1 is the theory diagram of the resource allocation methods of ofdm system of the present invention.
Fig. 2 is that number of users is 18 o'clock Streaming Media users' a packet loss rate emulation statistic histogram;
Fig. 3 is that number of users is 22 o'clock Streaming Media users' a packet loss rate emulation statistic histogram.
Embodiment:
Embodiment below in conjunction with description of drawings this method.
Embodiment 1:
The present embodiment utilized bandwidth is that 2.048MHz, data subcarrier number are 64 OFDM transmission system.Channel adopts COST207 six footpath models, and maximum multipath time delay is 10us, and six footpaths are spacedly distributed in 0~10us, and the power spectral density in each footpath satisfies common Jake model.Maximum Doppler frequency offset is 300Hz.The long 1ms of time slot, the grouping size is fixed as 1280bits, sliding window length t cValue be 1000 time slots.Adopt the MQAM adaptive modulation system, M={0,2,4,8,16,32}.Voice user's number in the fixed system and data user's number are respectively 5 and 3.
The service source model adopts ON-OFF two condition markov Markov process.Separate and the obeys index distribution of active period and quiet period.Transition probability from the active period to the quiet period is γ=1-exp (T Slot/ t 1), the transition probability from the quiet period to the active period is u=1-exp (T Slot/ t 2), T wherein SlotFor time slot long, t 1And t 2Be respectively the average of active period and quiet period.The business model parameter of using is as shown in table 1 below:
A mistake! Do not find Reference source.Table 1 business model parameter
System parameters Numerical value System parameters Numerical value
The speech average burst length 1.0s The average length of silence of speech 1.35s
The maximum PDR that speech allows 10 -2 The maximum PER that speech allows 10 -2
The maximum delay that speech allows 50ms The voice service Mean Speed 64Kbps
The Streaming Media average burst length 5.0s The average length of silence of Streaming Media 4.05s
The maximum PDR that Streaming Media allows 10 -3 The maximum PER that Streaming Media allows 10 -3
The maximum delay that Streaming Media allows 600ms The streaming media service Mean Speed 640Kbps
The data average burst length 10.0s The average length of silence of data 8.0s
The maximum PDR that data allow 10 -3 The maximum PER that data allow 10 -3
The maximum delay that data allow 2s The data service Mean Speed 1.28Mbps
Accompanying drawing 1 has provided the theory diagram of the resource allocation methods of ofdm system of the present invention: the base station end, in the initial moment of each time slot, initialization unit 1 is upgraded the label that respectively inserts Subscriber Queue, comprise that the number of packet signal 17 that can launch according to each user who adds up feedback unit 16 feedbacks after the last resource allocation end abandons the grouping of wait time delay more than or equal to this sustainable maximum wait time delay of dividing into groups, the number of packet of the mistake that receives in the last resource allocation process according to user feedback is added up the number of packet of the mistake that each user accumulative total receives, the subcarrier identifier that the user is assigned to carries out zero clearing, for new packet identifier is set up in the grouping that arrives in the last time slot, add up the number of packet that accumulative total arrives in each Subscriber Queue, add up the number of packet that each user's accumulative total is lost, the remaining number of packet of statistics Subscriber Queue, add up the remaining total number packets of this time slot Subscriber Queue, write down the wait time delay of each user's head of the queue grouping and add up the total number of users that is in state of activation in this time slot sytem, statistics is sent initialization end signal 2 after finishing; The initialization end signal 2 that user's subcarrier sequencing unit 8 is sent according to initialization module begins user's subcarrier to obtain each user's subcarrier gains descending series according to the gain sort descending; Each user's the satisfaction factor is calculated in initialization end signal 2 beginnings that satisfaction factor calculating unit 3 is sent according to initialization module by formula (1), equation (2) and formula (3); Each user's that utility function computing unit 4 provides according to the satisfaction factor calculating unit the satisfaction factor by formula (4), formula (5), formula (6) and formula (7) is calculated the utility function of the user of number of packet non-zero in the formation; After calculating end, utility function sequencing unit 5 obtains user's utility function descending series { G with user's utility function sort descending k} K=1 K, and with { G k} K=1 KBe that signal 6 is sent into subcarrier and bit associating allocation units 7; Subcarrier and bit associating distribution module 7 subcarrier gains descending series according to each user of storage in user's utility function descending series signal 6 of sending into and the user's subcarrier order module 8, from the head of the queue user, be followed successively by the idle sub-carrier in this user's subcarrier gains descending series of packet allocation in the Subscriber Queue, write down the identifier of the subcarrier that each user is assigned to, also preserve according to the information bit that formula (9) and formula (10) dispensed carry for these subcarriers, the total number packets that the minute book time slot can be launched, do not divide into groups to be launched or system's idle sub-carrier number is zero up to this user, and the number of subcarriers that will distribute is signal 10 is sent into number of subcarriers decision unit 11, and the total number packets that this time slot can be launched is that signal 9 is sent into system's emission number of packet decision unit 14; Number of subcarriers judging module 11 judges whether the decision value of receiving equals the system subcarrier sum, if unequal, triggering signal 12 is sent into system's emission number of packet decision unit 14; If equate, generation subcarrier and bit are united distribution end signal 13 and are sent into statistics feedback unit 16; Triggering signal 12 triggering systems emission number of packet judging module 14 judges whether the decision value of receiving 9 equals the remaining total number packets of this time slot of system, if it is unequal, produce triggering signal 15 and send into subcarrier and bit associating distribution module 7, subcarrier and bit associating distribution module 7 continues as the idle sub-carrier in this user's subcarrier gains descending series of packet allocation in the next Subscriber Queue of the low one-level of user's utility function descending series medium priority according to this signal, write down the identifier of the subcarrier that each user is assigned to, give the information bit that these subcarriers carry and preserve according to formula (9) and formula (10) dispensed, all users in system do not divide into groups to be launched or system's idle sub-carrier number is zero just to send subcarrier and bit is united distribution end signal 13; Subcarrier and bit are united the information bit that carries that distributes end signal 13 to trigger the identifier of the subcarrier that statistics feedback units 16 each user of record are assigned to and distribute to each subcarrier, and are that signal 17 feeds back to initialization module 1 with the number of packet that each user can launch.
When 10 Streaming Media users were arranged in the system, the algorithm that the present invention proposes not only can maintain user's packet loss rate a lower level, and can guarantee the fairness of service quality satisfaction between the user; When the Streaming Media number of users was increased to 14, because user's the service satisfaction factor is exactly the exponential weighting factor of utility function, the variation of the user grouping proportion of goods damageds rose and fell very little, and still is maintained at 5 * 10 -4~7 * 10 -4So lower level has preferable performance.The number of users that provides respectively from Fig. 2 and Fig. 3 be 18 and 22 o'clock Streaming Media users' packet loss rate emulation statistic histogram shown following feature:
1. when different user is counted, can guarantee the fairness of service quality satisfaction between the user.
2. the user grouping proportion of goods damageds rise and fall very little with the variation of number of users.
3. when number of users increased, the user grouping proportion of goods damageds still maintained reduced levels.
The present invention is starting point to improve the user to the satisfaction of system, provided a kind of layer resource allocation algorithm of striding that is used for ofdm system down link mixed service, considered the restriction of the professional stochastic behaviour that reaches, the requirement that professional equity is treated time delay, professional requirement, professional requirement, Subscriber Queue state and system to Packet Error Ratio Subscriber Queue length to packet loss.Stride a layer criterion by use, the present invention proposes service quality satisfaction fairness criterion, considered the combined effect of the service quality that data link layer and physical layer receive the user, and not only be applicable to real time business but also be applicable to non-real-time service; On this basis, the resource allocation algorithm that the present invention proposes uses utility function to characterize user's priority, be the more unsatisfied user of the service that system is provided and give higher priority, and adaptive subcarrier and bit associating distribution module is preferably the high user's allocation of subcarriers of grade.So the present invention not only can satisfy the requirement of user to service quality preferably, and can guarantee the fairness of service quality satisfaction between the user.

Claims (1)

1, a kind of mixed service resource distributing method of ofdm system comprises scheduler module and adaptive subcarrier and bit associating distribution module; The base station provides the formation of finite length for each user, queue length is the peak value arrival rate of grouping and the product of sustainable maximum wait time delay, the base station is also set up for each user's service queue and is safeguarded that label, label write down corresponding user identifier, packet identifier, the subcarrier identifier that the user is assigned to, the number of packet that accumulative total arrives in the Subscriber Queue, the number of packet that user's accumulative total is lost, the number of packet of the mistake that user's accumulative total receives, number of packet in active user's formation, the wait time delay of head of the queue grouping, the sustainable maximum wait time delay of dividing into groups, sustainable maximum packet loss divides into groups, divide into groups sustainable maximum Packet Error Ratio and the grouping the sustainable largest packet proportion of goods damageds; The base station is input to each user independently in the formation with the packet of business, and the SO service order that divides into groups in the formation is First Come First Served; The base station end carries out channel estimating by the user of number of packet non-zero in the broadcast channel notification queue; The user reports the erroneous packet quantity that receives in channel estimation results and the last resource allocation process to the base station by the feedback channel of zero defect; The every time slot of the resource allocation algorithm of base station side is carried out once, only gives a user, the equal subsystem total emission power of each subcarrier with a subcarrier allocation in each assigning process; In the initial moment of each time slot, scheduler module is user's ordering, and the user identifier sequence after will sort and the grouping in the Subscriber Queue offers adaptive subcarrier and bit is united distribution module; Adaptive subcarrier and bit associating distribution module earlier according to the channel condition information of user feedback with user's subcarrier according to the gain sort descending, obtain each user's subcarrier gains descending series, user identifier sequence that provides according to scheduler module and the grouping information in the Subscriber Queue are the packet allocation subcarrier in the Subscriber Queue again, and the number of packet that each user can be launched feeds back to scheduler module;
It is characterized in that:
Described scheduler module produces user's utility function descending series, comprising: initialization unit initial moment of each time slot according on once the resource allocation feedback that finishes back subcarrier and bit associating distribution module abandon of the grouping of wait time delay more than or equal to this sustainable maximum wait time delay of dividing into groups, the number of packet of the mistake that receives in the last resource allocation process according to user feedback is added up the number of packet of the mistake that each user accumulative total receives, the subcarrier identifier that the user is assigned to carries out zero clearing, for new packet identifier is set up in the grouping that arrives in the last time slot, add up the number of packet that accumulative total arrives in each Subscriber Queue, add up the number of packet that each user's accumulative total is lost, the remaining number of packet of statistics Subscriber Queue, add up the remaining total number packets of this time slot Subscriber Queue, write down the wait time delay of each user's head of the queue grouping and add up the total number of users that is in state of activation in this time slot sytem; After the initialization, the satisfaction factor calculating unit is according to formula S k [ i ] = PLR k [ i ] / PLR k max Calculate the satisfaction factor of each user k, PLR in the formula k MaxThe sustainable largest packet proportion of goods damageds of expression user k, PLR k[i] expression user k is at the packet loss rate of the initial moment statistics of i time slot, PLR k[i] can use equation PLR k [ i ] = PLN k [ i ] + PEN k [ i ] PAN k [ i ] Estimate, wherein PLN k[i], PEN k[i] and PAN k[i] represents the user k packet count that accumulative total is lost before i time slot, the packet count of the mistake that receives and the total number packets of arrival respectively; The satisfaction factor that the utility function computing unit calculates according to previous step is according to formula U k CL [ i ] = u k R k cur [ i ] R k aver [ i ] exp ( S k [ i ] - S [ i ] ‾ 1 + S [ i ] ‾ ) Calculate the utility function of the user k of number of packet non-zero in each formation, R in the formula k CurThe maximum data transmission rate that can support at i time slot of [i] expression user k prediction, R k AverThe average data transfer rate of [i] expression user k before i time slot used formula R k aver [ i + 1 ] = ( 1 - 1 t c ) R k aver [ i ] + 1 t c R k cur [ i ] Upgrade, wherein t cThe length of expression sliding time window is not given user k, then R if data are launched at i time slot in the base station k Aver[i+1] uses R k cur [ i ] = 0 Upgrade; u kUse u k = - log ( PDR k max ) / D k max Calculate, wherein PDR k MaxAnd D k MaxRepresent sustainable maximum packet loss of user k and packet delay respectively, S [ i ] ‾ = 1 K Σ k = 1 K S k [ i ] , Be in the total number of users of state of activation when wherein K is illustrated in i in the slot sytem; After calculating end, sequencing unit obtains user's utility function descending series with user's utility function sort descending, enters adaptive subcarrier and bit associating assigning process then; Adaptive subcarrier and bit associating distribution module comprises: the subcarrier sequencing unit is according to the channel condition information of user feedback, and the user's of number of packet non-zero subcarrier according to the gain sort descending, is obtained each user's subcarrier gains descending series; The head of the queue user of subcarrier and power associating allocation units from user's utility function descending series begins to be followed successively by the idle sub-carrier that this user's subcarrier gains of packet allocation in the Subscriber Queue is passed sequence, write down the identifier of the subcarrier that each user is assigned to, calculate record also and distribute to the bit number of the emission of these subcarriers, do not divide into groups to be launched or system's idle sub-carrier quantity is zero up to this user; If system also has idle sub-carrier and also has the packet awaits emission, continue as the idle sub-carrier in this user's subcarrier gains descending series of packet allocation in the next Subscriber Queue of the low one-level of user's utility function descending series medium priority, need be launched up to grouping not, or system's idle sub-carrier number is zero; Distribute after the end, the statistics feedback unit writes down the result of this resource allocation, and the number of packet that each user can launch is fed back to the initialization unit of scheduler module.
CNB2006100881334A 2006-06-28 2006-06-28 Mixed service resource distributing method for OFDM system Expired - Fee Related CN100512055C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2006100881334A CN100512055C (en) 2006-06-28 2006-06-28 Mixed service resource distributing method for OFDM system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2006100881334A CN100512055C (en) 2006-06-28 2006-06-28 Mixed service resource distributing method for OFDM system

Publications (2)

Publication Number Publication Date
CN1885748A true CN1885748A (en) 2006-12-27
CN100512055C CN100512055C (en) 2009-07-08

Family

ID=37583748

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2006100881334A Expired - Fee Related CN100512055C (en) 2006-06-28 2006-06-28 Mixed service resource distributing method for OFDM system

Country Status (1)

Country Link
CN (1) CN100512055C (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100469187C (en) * 2007-01-26 2009-03-11 中国科学技术大学 A satisfaction-based multi-user scheduling method in the multi-antenna system
WO2009019608A3 (en) * 2007-03-09 2009-06-04 Univ Hong Kong Science & Techn Delay-sensitive cross layer scheduler for multi-user wireless communication systems
CN100589474C (en) * 2007-04-29 2010-02-10 中国科学技术大学 Cross-layer scheduling method in OFDM multi-antenna system
CN101917768A (en) * 2010-08-25 2010-12-15 中国科学技术大学 User fairness resource allocation method for orthogonal frequency division multiple access relay system
CN101242204B (en) * 2007-02-08 2012-01-25 联想(北京)有限公司 A soft frequency multiplexing sub-carrier optimization method and device
CN101370247B (en) * 2007-08-14 2012-02-22 大唐移动通信设备有限公司 HS-SCCH channel allocation method and apparatus
WO2012159357A1 (en) * 2011-07-29 2012-11-29 华为技术有限公司 Resource allocation method and base station in orthogonal frequency division multiplexing system
CN106161325A (en) * 2015-05-17 2016-11-23 想象技术有限公司 Selectivity subcarrier processes
WO2017177854A1 (en) * 2016-04-15 2017-10-19 索尼公司 Apparatus and method for hybrid multiple access wireless communication system

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101053610B1 (en) * 2004-06-25 2011-08-03 엘지전자 주식회사 Radio resource allocation method of OPDM / OPDM system
CN1719816A (en) * 2004-07-09 2006-01-11 北京三星通信技术研究有限公司 Resource allocation methods in the ofdm system
CN102572820B (en) * 2004-11-02 2015-11-11 苹果公司 The method used together with OFDM and base station thereof and wireless terminal

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100469187C (en) * 2007-01-26 2009-03-11 中国科学技术大学 A satisfaction-based multi-user scheduling method in the multi-antenna system
CN101242204B (en) * 2007-02-08 2012-01-25 联想(北京)有限公司 A soft frequency multiplexing sub-carrier optimization method and device
WO2009019608A3 (en) * 2007-03-09 2009-06-04 Univ Hong Kong Science & Techn Delay-sensitive cross layer scheduler for multi-user wireless communication systems
CN100589474C (en) * 2007-04-29 2010-02-10 中国科学技术大学 Cross-layer scheduling method in OFDM multi-antenna system
CN101370247B (en) * 2007-08-14 2012-02-22 大唐移动通信设备有限公司 HS-SCCH channel allocation method and apparatus
CN101917768B (en) * 2010-08-25 2013-03-13 中国科学技术大学 User fairness resource allocation method for orthogonal frequency division multiple access relay system
CN101917768A (en) * 2010-08-25 2010-12-15 中国科学技术大学 User fairness resource allocation method for orthogonal frequency division multiple access relay system
WO2012159357A1 (en) * 2011-07-29 2012-11-29 华为技术有限公司 Resource allocation method and base station in orthogonal frequency division multiplexing system
US9225569B2 (en) 2011-07-29 2015-12-29 Huawei Technologies Co., Ltd. Resource allocation method and base station in orthogonal frequency division multiplexing system
CN106161325A (en) * 2015-05-17 2016-11-23 想象技术有限公司 Selectivity subcarrier processes
CN106161325B (en) * 2015-05-17 2021-04-16 北欧半导体公司 Method and apparatus for selective subcarrier processing
WO2017177854A1 (en) * 2016-04-15 2017-10-19 索尼公司 Apparatus and method for hybrid multiple access wireless communication system
US11096069B2 (en) 2016-04-15 2021-08-17 Sony Corporation Apparatus and method for hybrid multiple access wireless communication system

Also Published As

Publication number Publication date
CN100512055C (en) 2009-07-08

Similar Documents

Publication Publication Date Title
CN1885748A (en) Mixed service resource distributing method for OFDM system
CN100581074C (en) Real-time service resource distribution method of orthogonal frequency division multisystem
Ali et al. A performance study of uplink scheduling algorithms in point-to-multipoint WiMAX networks
CN102164108B (en) Multi-carrier communication with time division multiplexing and carrier-selective loading
CN1909437A (en) Adaptive feedback method of cellular communication
CN1258894C (en) Resource scheduling method utilized in operations sensitive to downward error in OFDM
CN1498472A (en) System and method for real time adaptive capacity scheduling
CN1802798A (en) Peak-to-average power ratio management for multi-carrier modulation in wireless communication systems
CN1713561A (en) Resource allocation method in a multicarrier communication system
US7280473B2 (en) Data streaming method and apparatus using adaptive transmission scheduling
CN103460658A (en) Lte scheduling
CN1540882A (en) Wireless communicatoon device of using multiple transceiver antenna system and method thereof
CN101132608A (en) Device and method for indicating signal channel resource allocation information
CN112367523B (en) Resource management method in SVC multicast based on NOMA in heterogeneous wireless network
CN1815933A (en) OFDMA system frequency time 2-D wire-less resource scheduling model and scheduling method
CN101933384A (en) Methods and apparatus for intra-user quality of service uplink scheduling
CN101043492B (en) Method and apparatus for distributing OFDM physical channel resources
CN101489298A (en) Discrete speed cross-layer power distribution method based on mobile multicast system
CN101212410B (en) Downlink resource scheduling method, system, and device
CN1980206A (en) Orthogonal frequency-division multiplexing system scheduling method
CN101478517A (en) OFDM system dynamic bit loading method based on ordering
CN101958828B (en) Service multiplex processing method and device
CN109392163B (en) Random frequency division multiple access system multi-carrier allocation method based on collision probability
Kallel et al. A flexible numerology configuration for efficient resource allocation in 3GPP V2X 5G new radio
CN1968529B (en) Cross-layer scheduling method for OFDM system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20090708

Termination date: 20150628

EXPY Termination of patent right or utility model