CN101917768B

CN101917768B - User fairness resource allocation method for orthogonal frequency division multiple access relay system

Info

Publication number: CN101917768B
Application number: CN2010102666095A
Authority: CN
Inventors: 刘畅; 秦晓卫; 张四海; 周武旸
Original assignee: University of Science and Technology of China USTC
Current assignee: University of Science and Technology of China USTC
Priority date: 2010-08-25
Filing date: 2010-08-25
Publication date: 2013-03-13
Anticipated expiration: 2030-08-25
Also published as: CN101917768A

Abstract

The invention discloses a user fair resource allocation method for an OFDMA relay system, which is characterized in that relay selection, subcarrier allocation and power allocation are jointly optimized to select the most suitable medium for each subcarrier and distribute it to the optimal relay-user link. At the same time, the base station and each relay perform adaptive power allocation on each subcarrier. By introducing a power adjustment factor, after several iterations, the total power is guaranteed to converge. to optimize system throughput. When selecting the optimal link, the goal is to ensure fairness among users, and when power allocation is performed, the average rate of users is used as one of the variables investigated. The method of the invention can ensure approximately strict fairness among users while optimizing the system throughput.

Description

A user fair resource allocation method for OFDA relay system

技术领域 technical field

本发明属于正交频分多址接入(OFDMA)移动通信技术领域，特别涉及OFDMA中继通信系统中保证用户公平性的资源分配方法。 The invention belongs to the technical field of Orthogonal Frequency Division Multiple Access (OFDMA) mobile communication, and particularly relates to a resource allocation method for ensuring user fairness in an OFDMA relay communication system. the

背景技术 Background technique

OFDMA技术能够根据不同用户信道衰落的独立性，充分利用多用户分集，提高频谱效率。另一方面，采用中继技术可以增大无线网络的覆盖范围，提高系统容量和可靠性。因此，OFDMA技术和中继技术是未来无线移动通信系统的核心技术。如何在OFDMA中继系统中进行合理的资源分配是一个越来越重要的研究课题。 OFDMA technology can make full use of multi-user diversity and improve spectrum efficiency according to the independence of channel fading of different users. On the other hand, the use of relay technology can increase the coverage of wireless networks and improve system capacity and reliability. Therefore, OFDMA technology and relay technology are the core technologies of future wireless mobile communication systems. How to allocate resources reasonably in OFDMA relay system is an increasingly important research topic. the

《国际电子与电气工程师协会通信选题杂志》(IEEE J.on Select Areas.Commun，Volume 17，No 10，1999，pp 1747-1758)中提到了一种OFDMA系统中的多用户自适应子载波、比特和功率分配算法。不过该算法应用场景为无中继节点的传统蜂窝小区，而在OFDMA中继系统中，对不同的用户要选择合适的中继节点传输数据，并且需要同时对基站和中继节点进行子载波和功率分配，因此该算法无法应用到OFDMA中继系统中。 "International Institute of Electronics and Electrical Engineers Communication Journal" (IEEE J.on Select Areas.Commun, Volume 17, No 10, 1999, pp 1747-1758) mentioned a multi-user adaptive subcarrier in OFDMA system , bit and power allocation algorithm. However, the application scenario of this algorithm is a traditional cell without a relay node. In an OFDMA relay system, it is necessary to select an appropriate relay node to transmit data for different users, and it is necessary to perform subcarrier summation on both the base station and the relay node at the same time. Power allocation, so this algorithm cannot be applied to OFDMA relay systems. the

在《国际电子与电气工程师协会通信选题杂志》(IEEE J.on Select Areas.Commun，Volume 25，No 2，2007，pp 328-339)中还提到了一种针对OFDMA中继系统的自适应子载波和功率分配算法，不过该算法的目标是最大化系统总效用，没有考虑用户间的公平性问题。而保证用户间的公平性、更好地为多用户服务，是未来无线通信系统的核心目标之一，因此保证用户间公平性的资源分配对未来无线通信系统具有重要意义。 In the "International Institute of Electronics and Electrical Engineers Communication Topic Selection Journal" (IEEE J.on Select Areas.Commun, Volume 25, No 2, 2007, pp 328-339) also mentioned an adaptive method for OFDMA relay system Subcarrier and power allocation algorithm, but the goal of this algorithm is to maximize the total utility of the system, without considering the issue of fairness among users. Ensuring fairness among users and better serving multiple users is one of the core goals of future wireless communication systems. Therefore, resource allocation that ensures fairness among users is of great significance to future wireless communication systems. the

发明内容 Contents of the invention

本发明的目的是提出一种正交频分多址接入中继系统的用户公平资源分配方法，该方法在优化系统吞吐量的同时，能够保证用户间具有近似严格的公平性。 The object of the present invention is to propose a user fair resource allocation method of an OFDMA relay system, which can ensure approximately strict fairness among users while optimizing system throughput. the

本发明正交频分多址接入中继系统的用户公平资源分配方法，基站根据输入的系统参数和信道参数，经以下处理步骤，由单片机输出得到的资源分配结果，其特征在于基站端依次执行以下步骤： The user fair resource allocation method of the OFDMA access relay system of the present invention, the base station, according to the input system parameters and channel parameters, through the following processing steps, the resource allocation result output by the single-chip microcomputer, is characterized in that the base station end sequentially Perform the following steps:

第一步：设时隙编号为t，对于系统中每一个用户m＝1，...，M，以

表示在时隙t计算得到的用户m的平均速率，初始化在时隙t＝0时用户m的平均速率

为非负值，M为系统中的总用户数； The first step: set the time slot number as t, for each user in the system m=1,...,M, with

Indicates the average rate of user m calculated at time slot t, and initializes the average rate of user m at time slot t=0

is a non-negative value, and M is the total number of users in the system;

第二步：设迭代轮次为i，对于每一个中继节点k＝1，...，K，以μ_k(i)表示中继节点k在迭代轮次i的功率调整因子，η(i)表示基站在迭代轮次i的功率调整因子，初始化中继节点k在迭代轮次i＝0的功率调整因子μ_k(0)以及基站在迭代轮次i＝0的功率调整因子η(0)为非负数，K为系统中的中继节点数； Second step: set the iteration round as i, for each relay node k=1,..., K, express the power adjustment factor of relay node k in iteration round i with μ _k (i), η( i) represents the power adjustment factor of the base station in iteration round i, initializes the power adjustment factor μ _k (0) of relay node k in iteration round i=0 and the power adjustment factor η( 0) is a non-negative number, and K is the number of relay nodes in the system;

基站s根据中继节点反馈的信道状态信息更新已存储的信息：对于所有的中继节点k＝1，...，K，和用户m＝1，...，M，更新基站到s中继节点k在子载波n上由小尺度衰落产生的信道增益

更新中继节点k到用户m在子载波n上由小尺度衰落产生的信道增益

更新基站s和中继节点k之间的路径损耗l_sk，更新中继节点k和用户m之间的路径损耗l_km； The base station s updates the stored information according to the channel state information fed back by the relay node: for all relay nodes k=1,...,K, and users m=1,...,M, update the base station to s Channel gain due to small-scale fading at subcarrier n following node k

Update channel gain from relay node k to user m on subcarrier n caused by small-scale fading

Update the path loss l _sk between the base station s and the relay node k, update the path loss l _km between the relay node k and the user m;

第三步：记系统中子载波总数为N，对于每一个子载波n＝1，...，N，根据当前各用户的平均速率

基站功率调整因子η(i)、中继功率调整因子μ_k(i)，和基站s到中继节点k在子载波n上由小尺度衰落产生的信道增益

中继节点k到用户m在子载波n上由小尺度衰落产生的信道增益基站s和中继节点k之间的路径损耗l_sk，中继节点k和用户m之间的路径损耗，分别计算当前在该载波上的功率分配结果以及该载波在链路上的分配结果； The third step: record the total number of subcarriers in the system as N, for each subcarrier n=1,...,N, according to the average rate of each user at present

Base station power adjustment factor η(i), relay power adjustment factor μ _k (i), and channel gain from base station s to relay node k on subcarrier n caused by small-scale fading

Channel gain from relay node k to user m on subcarrier n caused by small-scale fading The path loss l _sk between the base station s and the relay node k, the path loss between the relay node k and the user m, respectively calculate the current power allocation result on the carrier and the allocation result of the carrier on the link;

由下面所列的中继节点k到用户m在子载波n上的发射功率公式(1)得出功率分配结果： The power allocation result is obtained from the transmit power formula (1) of relay node k to user m on subcarrier n listed below:

${p p}_{km km}^{n no} = = {[[\frac{W W / / 22 N N}{{Φ Φ}_{km km}^{n no} {\overset{&OverBar; &OverBar;}{R R}}_{m m} ((t t)) ln ln 22} - - \frac{11}{{g g}_{km km}^{n no}}]]}^{+ +} - - - - - - ((11))$

式(1)中W为系统总带宽，被分为N个子载波，运算符[x]⁺＝max{x，0}，即表示取数值x和0中的较大值，前一个中间变量

和后一个中间变量

分别为 In formula (1), W is the total bandwidth of the system, which is divided into N subcarriers. The operator [x] ⁺ =max{x, 0} means to take the larger value of x and 0. The previous intermediate variable

and the latter intermediate variable

respectively

${Φ Φ}_{km km}^{n no} = = {μ μ}_{k k} ((i i)) = = \frac{η η ((i i)) {l l}_{km km} {| | {h h}_{km km}^{n no} | |}^{22}}{{l l}_{sk sk} {| | {h h}_{sk sk}^{n no} | |}^{22}} - - - - - - ((22))$

${g g}_{km km}^{n no} = = \frac{{l l}_{km km} {| | {h h}_{km km}^{n no} | |}^{22}}{Γ Γ {WN WN}_{00} / / N N} - - - - - - ((33))$

式(3)中N₀为系统中的单边带噪声功率谱密度，信噪比间隙Γ是目标误比特率BER的函数，Γ＝-ln(5BER)/1.5； In formula (3), N ₀ is the SSB noise power spectral density in the system, and the signal-to-noise ratio gap Γ is a function of the target bit error rate BER, Γ=-ln(5BER)/1.5;

根据中继节点k到用户m在子载波n上的发射功率公式(1)求得的中继节点k到用户m在子载波n上的发射功率的值，计算中继节点k到用户m在子载波n上的传输速率 The transmit power from relay node k to user m on subcarrier n calculated according to the transmit power formula (1) from relay node k to user m on subcarrier n The value of , calculate the transmission rate from relay node k to user m on subcarrier n

${r r}_{km km}^{n no} = = \frac{W W}{22 N N} lo lo {g g}_{22} ((11 + + \frac{{p p}_{km km}^{n no} {l l}_{km km} {| | {h h}_{km km}^{n no} | |}^{22}}{ΓW ΓW {N N}_{00} / / N N})) - - - - - - ((44))$

根据中继节点k到用户m在子载波n上的传输速率公式(4)求得的传输速率

进行子载波分配：对于某个子载波n，首先选取

最大的(k，m)组合，即 The transmission rate calculated according to the transmission rate formula (4) from relay node k to user m on subcarrier n

Perform subcarrier allocation: For a certain subcarrier n, first select

The largest (k,m) combination, i.e.

$(({k k}^{* *},, {m m}^{* *})) = = arg arg \underset{k k,, m m}{max max} \frac{{r r}_{km km}^{n no}}{{\overset{&OverBar; &OverBar;}{R R}}_{m m} ((t t))} - - - - - - ((55))$

然后将子载波n分配给选取的(k^*，m^*)，即分配给中继节点k^*到用户m^*的链路；定义表示载波分配的指示变量

表示将载波n分配给中继节点k到用户m的链路，否则，指示变量

对于中继节点k^*到用户m^*的链路，指示变量

而对于中继节点k^*到用户m^*以外的其他链路，即对于中继节点k≠k^*到用户m≠m^*的链路，指示变量

用公式表示为： Then assign the subcarrier n to the selected (k ^* , m ^* ), that is, assign to the link from the relay node k ^* to the user m ^* ; define the indicator variable representing the carrier assignment

Indicates that carrier n is allocated to the link from relay node k to user m, otherwise, the indicator variable

For a link from relay node k ^* to user m ^* , the indicator variable

And for other links from relay node k ^* to user m ^* , that is, for the link from relay node k≠k ^* to user m≠m ^* , the indicator variable

Expressed as:

${ρ ρ}_{{k k}^{* *} {m m}^{* *}}^{n no} = = 11,, {ρ ρ}_{km km}^{n no} = = 00,, &ForAll; &ForAll; k k &NotEqual; &NotEqual; {k k}^{* *},, m m &NotEqual; &NotEqual; {m m}^{* *} - - - - - - ((66))$

则基站到中继节点k在子载波n上的发射功率由下式算出： Then the transmit power from base station to relay node k on subcarrier n is calculated by the following formula:

$p_{sk}^{n} = \frac{l_{km} {| h_{km}^{n} |}^{2}}{l_{sk} {| h_{sk}^{n} |}^{2}} p_{km}^{n},$ 当 $ρ_{km}^{n} = 1$ 时 (7) $p_{sk}^{no} = \frac{l_{km} {| h_{km}^{no} |}^{2}}{l_{sk} {| h_{sk}^{no} |}^{2}} p_{km}^{no},$ when $ρ_{km}^{no} = 1$ hours (7)

第四步：更新迭代轮次i＝i+1，并采用下面的式子更新中继和基站功率调整因子 Step 4: Update iteration round i=i+1, and use the following formula to update relay and base station power adjustment factors

${μ μ}_{k k} ((i i + + 11)) = = {[[{μ μ}_{k k} ((i i)) - - {α α}_{11} ((i i)) (({P P}_{k k}^{max max} - - {Σ Σ}_{m m = = 11}^{M m} {Σ Σ}_{n no = = 11}^{N N} {p p}_{km km}^{n no}))]]}^{+ +},, k k = = 11,, . . . . . .,, K K - - - - - - ((88))$

$η η ((i i + + 11)) = = {[[η η ((i i)) - - {α α}_{22} ((i i)) (({P P}_{BS BS}^{max max} - - {Σ Σ}_{m m = = 11}^{M m} {Σ Σ}_{k k = = 11}^{K K} {Σ Σ}_{n no = = 11}^{N N} {p p}_{sk sk}^{n no}))]]}^{+ +} - - - - - - ((99))$

其中表示中继节点k的最大功率约束，

表示基站的最大功率约束，α₁(i)和α₂(i)为迭代步长； in Denotes the maximum power constraint of relay node k,

Indicates the maximum power constraint of the base station, α ₁ (i) and α ₂ (i) are the iteration step size;

第五步：根据第四步中的中继功率调整因子更新公式(8)和基站功率调整因子更新公式(9)的计算结果，判断发射功率是否收敛，如果不收敛，则返回第三步，如果收敛，进入第六步； The fifth step: according to the calculation results of the relay power adjustment factor update formula (8) and the base station power adjustment factor update formula (9) in the fourth step, it is judged whether the transmission power converges, if not converged, then return to the third step, If convergent, enter the sixth step;

第六步：对于用户m＝1，...，M，根据前面第三步到第五步迭代计算得到的载波和功率分配结果，按瞬时速率计算公式 Step 6: For users m=1,...,M, according to the carrier and power allocation results obtained through iterative calculations from the third to fifth steps above, calculate the formula according to the instantaneous rate

$R_{m} (t) = Σ_{k = 1}^{K} Σ_{n = 1}^{N} ρ_{km}^{n} r_{km}^{n},$ m＝1，...，M (10) $R_{m} (t) = Σ_{k = 1}^{K} Σ_{no = 1}^{N} ρ_{km}^{no} r_{km}^{no},$ m=1,...,M (10)

计算当前瞬时速率R_m(t)，并按平均速率

计算公式 Calculate the current instantaneous rate R _m (t), and calculate the average rate

Calculation formula

${\overset{&OverBar; &OverBar;}{R R}}_{m m} ((t t + + 11)) = = ((11 - - \frac{11}{T T})) {\overset{&OverBar; &OverBar;}{R R}}_{m m} ((t t)) + + \frac{11}{T T} {R R}_{m m} ((t t)) - - - - - - ((1111))$

更新平均速率

式中T为滑动窗口长度； update average rate

where T is the length of the sliding window;

第七步：更新时隙t＝t+1，返回第二步计算下一时隙的资源分配方案。 Step 7: update time slot t=t+1, return to step 2 to calculate the resource allocation scheme for the next time slot. the

与现有资源分配技术相比，本发明的资源分配针对的是未来通信中具有普遍意义的多用户、多中继系统，并且是在保证下行通信用户间公平性情况下进行的资源分配。本发明将资源分配过程中涉及到的中继选择、子载波分配和功率分配问题进行联合优化，并且基于分解理论和凸优化理论，给出了最优的资源分配方法。由于本发明为每个子载波选择最合适的中继，并将其分配给最优的中继-用户链路，同时基站和各中继在各个子载波上进行自适应功率分配，通过引入功率调整因子，经过若干次迭代，可以保证总功率收敛，因此可以优化系统的吞吐量。同时，本发明在进行中继选择和载波分配过程中在选择最优链路时是以保证用户间公平性为目标，同时进行功率分配的时候也以用户的平均速率作为其中一个考察变量，因此可以保证用户间资源分配近似严格的公平性。综上，本发明提出的资源分配方法不仅能够优化系统吞吐量，同时能够保证用户间具有近似严格的公平性。 Compared with the existing resource allocation technology, the resource allocation of the present invention is aimed at multi-user and multi-relay systems with universal significance in future communication, and is resource allocation under the condition of ensuring the fairness among downlink communication users. The invention jointly optimizes the relay selection, subcarrier allocation and power allocation problems involved in the resource allocation process, and provides the optimal resource allocation method based on the decomposition theory and the convex optimization theory. Since the present invention selects the most suitable relay for each subcarrier and assigns it to the optimal relay-user link, at the same time the base station and each relay perform adaptive power allocation on each subcarrier, by introducing power adjustment Factor, after several iterations, the total power can be guaranteed to converge, so the throughput of the system can be optimized. At the same time, the present invention aims to ensure fairness among users when selecting the optimal link in the process of relay selection and carrier allocation, and also uses the average rate of users as one of the investigation variables when performing power allocation, so Approximate strict fairness of resource allocation among users can be guaranteed. In summary, the resource allocation method proposed by the present invention can not only optimize system throughput, but also ensure approximately strict fairness among users. the

附图说明 Description of drawings

图1为本发明正交频分多址接入中继通信系统的资源分配方法的原理框图； Fig. 1 is the functional block diagram of the resource allocation method of OFDMA relay communication system of the present invention;

图2为本发明正交频分多址接入中继通信系统的资源分配方法的硬件实现系统图； Fig. 2 is the hardware realization system diagram of the resource allocation method of OFDMA access relay communication system of the present invention;

图3为两种算法下系统吞吐量对比图； Figure 3 is a comparison chart of system throughput under the two algorithms;

图4为两种算法下用户公平性对比图。 Figure 4 is a comparison of user fairness under the two algorithms. the

具体实施方式 Detailed ways

以下结合附图说明本发明的实施方案。 Embodiments of the present invention are described below in conjunction with the accompanying drawings. the

实施例1： Example 1:

本实施例采用下行OFDAM中继蜂窝网络，小区半径为1km，内环半径为0.6km，各中继节点均匀的分布在内环上，用户的位置随机产生，并且均匀地分布在内外环之间。具体仿真参数设置如表1所示。仿真统计1000次资源分配的平均结果，从系统吞吐量和用户公平性两方面来考察本专利提出的资源分配方法的性能。 This embodiment adopts the downlink OFDAM relay cellular network, the radius of the cell is 1km, and the radius of the inner ring is 0.6km. The relay nodes are evenly distributed on the inner ring, and the positions of users are randomly generated and evenly distributed between the inner and outer rings. . The specific simulation parameter settings are shown in Table 1. The average result of 1000 times of resource allocation is calculated by simulation, and the performance of the resource allocation method proposed in this patent is examined from two aspects of system throughput and user fairness. the

表1.参数设置 Table 1. Parameter settings

本实施例在基站端依次执行以下具体操作步骤： In this embodiment, the following specific operation steps are sequentially performed at the base station:

is a non-negative value, and M is the total number of users in the system;

基站根据中继节点反馈的信道状态信息更新已存储的信息：对于所有的中继节点k＝1，...，K，和用户m＝1，...，M，更新基站到中继节点k在子载波n上由小尺度衰落产生的信道增益

更新基站和中继节点k之间的路径损耗l_sk，更新中继节点k和用户m之间的路径损耗l_km； The base station updates the stored information according to the channel state information fed back by the relay node: for all relay nodes k=1,...,K, and users m=1,...,M, update the base station to the relay node k channel gain due to small-scale fading on subcarrier n

Update the path loss l _sk between the base station and relay node k, and update the path loss l _km between relay node k and user m;

基站功率调整因子η(i)，中继功率调整因子μ_k(i)，和基站到中继节点k在子载波n上由小尺度衰落产生的信道增益

中继节点k到用户m在子载波n上由小尺度衰落产生的信道增益基站和中继节点k之间的路径损耗l_sk，中继节点k和用户m之间的路径损耗，分别计算当前在该载波上的功率分配结果以及该载波在链路上的分配结果； The third step: record the total number of subcarriers in the system as N, for each subcarrier n=1,...,N, according to the average rate of each user at present

Base station power adjustment factor η(i), relay power adjustment factor μ _k (i), and channel gain from base station to relay node k on subcarrier n caused by small-scale fading

Channel gain from relay node k to user m on subcarrier n caused by small-scale fading The path loss l _sk between the base station and the relay node k, the path loss between the relay node k and the user m, respectively calculate the current power allocation result on the carrier and the allocation result of the carrier on the link;

由下式得出功率分配结果：中继节点k到用户m在子载波n上的发射功率 The power allocation result is obtained by the following formula: the transmit power from relay node k to user m on subcarrier n

和后一个中间变量分别为 In formula (1), W is the total bandwidth of the system, which is divided into N subcarriers. The operator [x] ⁺ =max{x, 0} means to take the larger value of x and 0. The previous intermediate variable

and the latter intermediate variable respectively

${Φ Φ}_{km km}^{n no} = = {μ μ}_{k k} ((i i)) + + \frac{η η ((i i)) {l l}_{km km} {| | {h h}_{km km}^{n no} | |}^{22}}{{l l}_{sk sk} {| | {h h}_{sk sk}^{n no} | |}^{22}} - - - - - - ((22))$

${g g}_{km km}^{n no} = = \frac{{l l}_{km km} {| | {h h}_{km km}^{n no} | |}^{22}}{ΓW ΓW {N N}_{00} / / N N} - - - - - - ((33))$

根据中继节点k到用户m在子载波n上的发射功率公式(1)求得的中继节点k到用户m在子载波n上的发射功率

的值，计算中继节点k到用户m在子载波n上的传输速率 The transmit power from relay node k to user m on subcarrier n calculated according to the transmit power formula (1) from relay node k to user m on subcarrier n

The value of , calculate the transmission rate from relay node k to user m on subcarrier n

进行子载波分配：对于某个子载波n，首先选取

Perform subcarrier allocation: For a certain subcarrier n, first select

The largest (k,m) combination, i.e.

对于中继节点k^*到用户m^*的链路，指示变量

用公式表示如下： Then assign the subcarrier n to the selected (k ^* , m ^* ), that is, assign to the link from the relay node k ^* to the user m ^* ; define the indicator variable representing the carrier assignment

For a link from relay node k ^* to user m ^* , the indicator variable

The formula is as follows:

其中表示中继节点k的最大功率约束，

第五步：根据第四步中的中继功率调整因子更新公式(8)和基站功率调整因子更新公式(9)的计算结果，判断发射功率是否收敛，如果不收敛，则返回第三步，如果收敛，进入第六步； The fifth step: according to the calculation results of the relay power adjustment factor update formula (8) and the base station power adjustment factor update formula (9) in the fourth step, it is judged whether the transmission power converges, if not converged, then return to the third step, If it converges, go to the sixth step;

计算当前瞬时速率R_m(t)，并按平均速率

Calculation formula

更新平均速率

式中T为滑动窗口长度； update average rate

where T is the length of the sliding window;

附图1为本发明正交频分多址接入中继通信系统的资源分配方法的原理框图：在基站端，初始化步骤1在起始时刻初始化时隙编号t＝0，初始化各用户的平均速率

为非负值；信道更新步骤2初始化迭代轮次i＝0，初始化中继功率调整因子μ_k(0)，k＝1，...，K和基站功率调整因子η(0)为非负数，更新信道状态信息；资源分配步骤3对每个子载波，分别按照中继节点的发射功率计算公式(1)、前一个中间变量

的计算公式(2)、后一个中间变量的计算公式(3)计算中继功率分配结果，之后按照传输速率公式(4)计算链路速率，并根据链路选取公式(5)、载波分配公式(6)求得该载波分配结果，最后由基站发射功率计算公式(7)得到基站功率分配结果；迭代步骤4更新迭代轮次i＝i+1，并按照中继功率调整因子更新公式(8)、基站功率调整因子更新公式(9)分别更新中继和基站功率调整因子μ_k(i)和η(i)；收敛判别步骤5判断发射功率是否收敛，如果不收敛，则返回资源分配步骤3，如果收敛，进入速率更新步骤6；速率更新步骤6根据资源分配步骤3到收敛判别步骤5迭代计算得到的载波和功率分配结果，按照瞬时速率计算公式(10)计算当前瞬时速率，并按照平均速率更新公式(11)更新平均速率；时隙更新步骤7更新时隙t＝t+1，返回信道更新步骤2计算下一时隙的资源分配方案。 Accompanying drawing 1 is the functional block diagram of the resource allocation method of OFDMA relay communication system of the present invention: at base station end, initialization step 1 initializes time slot number t=0 at initial moment, initializes the average of each user rate

Is a non-negative value; channel update step 2 initialization iteration round i=0, initialization relay power adjustment factor μ _k (0), k=1,..., K and base station power adjustment factor η (0) are non-negative numbers , update the channel state information; resource allocation step 3 for each subcarrier, according to the transmit power calculation formula (1) of the relay node, the previous intermediate variable

Calculation formula (2), the latter intermediate variable The calculation formula (3) calculates the relay power allocation result, and then calculates the link rate according to the transmission rate formula (4), and obtains the carrier allocation result according to the link selection formula (5) and the carrier allocation formula (6), and finally The base station power allocation result is obtained by base station transmission power calculation formula (7); iterative step 4 updates iteration round i=i+1, and updates formula (8) according to relay power adjustment factor and base station power adjustment factor update formula (9) Respectively update relay and base station power adjustment factors μ _k (i) and η (i); Convergence discrimination step 5 judges whether the transmission power is convergent, if not convergent, then returns to resource allocation step 3, if convergent, enters rate update step 6; The rate update step 6 calculates the current instantaneous rate according to the instantaneous rate calculation formula (10) according to the carrier and power allocation results iteratively calculated from the resource allocation step 3 to the convergence judgment step 5, and updates the average rate according to the average rate update formula (11); The time slot update step 7 updates the time slot t=t+1, and returns to the channel update step 2 to calculate the resource allocation scheme for the next time slot.

图2为本发明正交频分多址接入中继通信系统的资源分配方法的硬件实现系统示意图，该实现系统包括MSC-51单片机A、外设接口RS-232C、可擦除可编程只读存储器EPROM和随机存取存储器RAM。将可擦除可编程只读存储器EPROM和随机存取存储器RAM连接到单片机A上，单片机A与外设接口RS-232C相互连接。实现系统的输入参数包括各系统参数B：有总带宽W、子载波数目N、信噪比间隙Γ、噪声功率谱密度N₀、中继最大功率约束

基站最大功率约束

中继节点的数目K和用户数目M；还包括对于所有中继节点k＝1，...，K和用户m＝1，...，M，在各子载波n＝1，...，N上的信道参数C：有基站到中继节点k在子载波n上由小尺度衰落产生的信道增益

中继节点k到用户m在子载波n上由小尺度衰落产生的信道增益

基站和中继节点k之间的路径损耗l_sk以及中继节点k和用户m之间的路径损耗l_km。实现系统的输出为对于所有中继节点k＝1，...，K和用户m＝1，...，M，在各子载波n＝1，...，N上的资源分配结果D：有指示载波n是否分配给中继节点k到用户m链路的载波分配指示变量

中继节点k到用户m在子载波n上的发射功率

以及基站到中继节点k在子载波n上的发射功率

将系统参数B和信道参数C通过外设接口RS-232C输入到单片机A，经过运算处理得到资源分配结果D，结果由单片机A通过外设接口RS-232C输出。 Fig. 2 is the hardware implementation system diagram of the resource allocation method of OFDMA access relay communication system of the present invention, and this implementation system comprises MSC-51 single-chip microcomputer A, peripheral interface RS-232C, erasable programmable only Read memory EPROM and random access memory RAM. Connect the erasable programmable read-only memory EPROM and the random access memory RAM to the single-chip microcomputer A, and the single-chip microcomputer A and the peripheral interface RS-232C are connected to each other. The input parameters to realize the system include various system parameters B: the total bandwidth W, the number of subcarriers N, the signal-to-noise ratio gap Γ, the noise power spectral density N ₀ , and the maximum relay power constraint

Base station maximum power constraint

The number of relay nodes K and the number of users M; also include for all relay nodes k=1,...,K and users m=1,...,M, in each subcarrier n=1,... , channel parameter C on N: channel gain from base station to relay node k on subcarrier n caused by small-scale fading

Channel gain from relay node k to user m on subcarrier n caused by small-scale fading

The path loss l _sk between the base station and relay node k and the path loss l _km between relay node k and user m. The output of the realization system is the resource allocation result D on each subcarrier n=1,...,N for all relay nodes k=1,...,K and users m=1,...,M : There is a carrier allocation indicator variable indicating whether carrier n is allocated to the link from relay node k to user m

Transmit power from relay node k to user m on subcarrier n

and the transmit power from base station to relay node k on subcarrier n

Input the system parameter B and channel parameter C to the single-chip computer A through the peripheral interface RS-232C, and obtain the resource allocation result D after calculation and processing, and the result is output by the single-chip computer A through the peripheral interface RS-232C.

将本发明的资源分配方法的性能与常用的一种采用启发式算法的方法进行比较。在对比的方法中，每个用户选择与其距离最近的中继节点传输数据，采用贪婪算法进行子载波分配，即将每个子载波分配给在其上信道条件最好的用户，并且基站和各个中继在它们传输的各载波上平均分配功率。下面从系统吞吐量和用户公平性两方面来考察方法的性能。 The performance of the resource allocation method of the present invention is compared with a commonly used method using a heuristic algorithm. In the comparative method, each user selects the relay node with the closest distance to transmit data, and the greedy algorithm is used for subcarrier allocation, that is, each subcarrier is allocated to the user with the best channel condition on it, and the base station and each relay node Power is evenly distributed across the carriers on which they transmit. In the following, we examine the performance of the method from two aspects of system throughput and user fairness. the

附图3给出了两种算法下随用户数变化系统吞吐量的变化曲线。上面的曲线a为采用本发明的资源分配方法得到的曲线，下面的曲线b为采用对比方法得到的曲线。从图中可以看出，在不同用户数情况下，采用本发明的资源分配方法得到的系统吞吐量性能都超过对比方法。这是因为本发明的方法为每个载波选择合适的中继，并且进行自适应分配功率，提高了系统吞吐量。 Figure 3 shows the variation curves of the system throughput with the number of users under the two algorithms. The upper curve a is a curve obtained by using the resource allocation method of the present invention, and the lower curve b is a curve obtained by using a comparison method. It can be seen from the figure that in the case of different numbers of users, the system throughput performance obtained by adopting the resource allocation method of the present invention exceeds that of the comparison method. This is because the method of the present invention selects a suitable relay for each carrier, and performs adaptive power allocation, thereby improving system throughput. the

采用公平性因子来考察算法的公平性，定义公平性因子FI如下 The fairness factor is used to examine the fairness of the algorithm, and the fairness factor FI is defined as follows

$FI FI = = {(({Σ Σ}_{m m = = 11}^{M m} {\overset{&OverBar; &OverBar;}{S S}}_{m m}))}^{22} / / ((M m {Σ Σ}_{m m = = 11}^{M m} {(({\overset{&OverBar; &OverBar;}{S S}}_{m m}))}^{22})) - - - - - - ((1212))$

其中

为用户m经过1000次资源分配的平均速率。当所有用户平均速率相等时，公平性因子的值为1。越接近1，说明用户间公平性越好。 in

is the average rate of 1000 resource allocations for user m. When the average rate of all users is equal, the value of the fairness factor is 1. The closer to 1, the better the fairness among users.

附图4给出了两种算法在不同用户数下公平性因子变化曲线，其中上面的曲线c为采用本发明的资源分配方法得到的曲线，下面的曲线d为采用对比方法得到的曲线。可见，在不同用户数情况下，采用本发明的资源分配方法得到的用户公平性因子均非常接近1，说明当系统中具有不同数目用户时，本发明的方法均能够保证各用户的平均速率近似相等，具有近似严格的公平性。而从采用对比方法得到的公平性因子曲线中可以看出，随着用户数增加，公平性迅速下降。这是因为对比方法采用贪婪算法分配载波，当用户数增多时，一些处于小区边缘的用户由于信道状态差而一直无法获得载波资源。 Accompanying drawing 4 has given the change curve of the fairness factor of the two algorithms under different numbers of users, wherein the upper curve c is the curve obtained by using the resource allocation method of the present invention, and the lower curve d is the curve obtained by using the comparison method. It can be seen that in the case of different numbers of users, the user fairness factors obtained by using the resource allocation method of the present invention are all very close to 1, indicating that when there are different numbers of users in the system, the method of the present invention can ensure that the average rate of each user is approximately equal, with approximately strict fairness. From the fairness factor curve obtained by using the comparison method, it can be seen that as the number of users increases, the fairness decreases rapidly. This is because the comparison method uses a greedy algorithm to allocate carriers. When the number of users increases, some users at the edge of the cell have been unable to obtain carrier resources due to poor channel conditions. the

Claims

1. the user fairness resource allocation method of an orthogonal frequency division multiple access relay system, the base station is according to system parameters and the channel parameter of input, through following treatment step, by the resource allocation result that single-chip microcomputer output obtains, it is characterized in that the base station end carries out following steps successively:

The first step: establishing time-gap number is t, for each user m=1 in the system ..., M, with Be illustrated in the Mean Speed of the user m that time slot t calculates, the Mean Speed of initialization user m when time slot t=0

Be nonnegative value, M is the total number of users in the system;

Second step: establishing the iteration round is i, for each via node k=1 ..., K is with μ _k(i) expression via node k is in the power adjusting factor of iteration round i, and η (i) expression base station is in the power adjusting factor of iteration round i, and initialization via node k is at the power adjusting factor μ of iteration round i=0 _k(0) and the base station be nonnegative number at the power adjusting factor η (0) of iteration round i=0, K is the via node number in the system;

Base station s upgrades canned data according to the channel condition information of via node feedback: for all via node k=1 ..., K, and user m=1 ..., M upgrades the channel gain that base station s is produced by multipath fading on subcarrier n to via node k

Upgrade the channel gain that via node k is produced by multipath fading on subcarrier n to user m Upgrade the path loss l between base station s and the via node k _Sk, upgrade the path loss l between via node k and the user m _Km

The 3rd step: note system sub-carriers adds up to N, for each subcarrier n=1 ..., N is according to current each user's Mean Speed

Base station power is adjusted factor η (i), relaying power adjusting factor μ _kAnd the channel gain that on subcarrier n, produced by multipath fading to via node k of base station s (i),

The channel gain that via node k is produced by multipath fading on subcarrier n to user m

Path loss l between base station s and the via node k _Sk, the path loss between via node k and the user m is calculated respectively current power distribution result and the allocation result of this carrier wave on link on this carrier wave;

Draw power distribution result by following listed via node k to the transmitting power formula (1) of user m on subcarrier n:

W is overall system bandwidth in the formula (1), is divided into N subcarrier, operator [x] ⁺=max{x, 0} namely represents the higher value in peek value x and 0, previous intermediate variable

With a rear intermediate variable

Be respectively

N in the formula (3) ₀Be the single-side belt noise power spectral density in the system, signal to noise ratio gap Γ is the function of target bit BER, Γ=-ln (5BER)/1.5;

The via node k that tries to achieve at the transmitting power formula (1) on the subcarrier n to user m according to via node k is to the transmitting power of user m on subcarrier n Value, calculate via node k to the transmission rate of user m on subcarrier n

The transmission rate of trying to achieve to the transmission rate formula (4) of user m on subcarrier n according to via node k

Carrying out subcarrier distributes: for certain subcarrier n, at first choose

Maximum (k, m) combination, namely

Then subcarrier n is distributed to the (k that chooses ^*, m ^*), namely distribute to via node k ^*To user m ^*Link; The indicator variable of definition expression allocation of carriers

Expression is distributed to via node k to the link of user m with carrier wave n, otherwise, indicator variable

For via node k ^*To user m ^*Link, indicator variable

And for via node k ^*To user m ^*Other links in addition are namely for via node k ≠ k ^*To user m ≠ m ^*Link, indicator variable

Be formulated as:

Then the base station is calculated by following formula to the transmitting power of via node k on subcarrier n:

When

The time (7)

The 4th step: upgrade iteration round i=i+1, and the formula below adopting is upgraded relaying and base station power is adjusted the factor

Wherein

The maximum power constraint of expression via node k, The maximum power constraint of expression base station, α ₁(i) and α ₂(i) be iteration step length;

The 5th step: according to new formula (8) and the base station power adjustment factor result of calculation of new formula (9) more more of the relaying power adjusting factor in the 4th step, judge whether transmitting power restrains, if do not restrain, then returned for the 3rd step, if convergence entered for the 6th step;

The 6th step: for user m=1 ..., M goes on foot carrier wave and the power distribution result that the 5th step iterative computation obtains according to front the 3rd, by the momentary rate computing formula

m＝1，...，M (10)

Calculate current momentary rate R _m(t), and by Mean Speed

Computing formula

Upgrade Mean Speed

T is sliding window length in the formula;

The 7th step: upgrade time slot t=t+1, return the Resource Allocation Formula that second step calculates next time slot.