文档库 最新最全的文档下载
当前位置:文档库 › Detection for MIMO Systems

Detection for MIMO Systems

Detection for MIMO Systems
Detection for MIMO Systems

Detection for MIMO Systems

IIntroduction

To date, there are several categories of detection schemes in the reverse link, Such as: Iterative linear filtering schemes and Random step methods.

Iterative linear filtering schemes work by resolving the detection of the Signal vector by iterative linear filtering, and at each iteration by means of new propagated information from the previous estimate. The propagated information can be either hard, i.e., consist of decisions on the signal vectors, or soft, i.e., contain some probabilistic measures of the transmitted symbols. The methods typically employ matrix inversions repeatedly during the iterations, which, if the inversions occur frequently, may be computationally heavy when transmit antenna number is large. Luckily, the matrix inversion lemma can be used to remove some of the complexity stemming from matrix inversions.

As an example of a soft information-based method, we describe the conditional MMSE with soft interference cancellation (MMSE-SIC) [1] scheme in this paper. Random step methodsarematrix-inversion-free, except possibly for the initialization stage, where the MMSE solution is usually used. A basic matrix inversion-free search method starts with the initial vector, and evaluates the MSE for vectors in its neighborhood. The neighboring vector with smallest MSE is chosen, and the process restarts, and continues like this for several iterations.

As an example of a Random step method, TABU SEARCH (TS) scheme keeps a list of recently traversed signaling vectors, with maximum number of entries,which are temporarily forbidden moves, as a means for moving away to new areas of the search space.We will describeit latter in this paper.

I Channel Model

Consider a generic MIMO channel with signal dimension of K and observation dimension of N as follows:

x Hs n =+ (1)

Where N K H C ?∈the uncorrelated channel matrix with 12[h ,h ,,h ],

K H =

12[,,,]T K s s s s = is the 1K ?transmitted signal vector consists of K symbols,

12[,,,]T N x x x x = is the 1N ? received signal vector, 12[,,,]T N n n n n = is

the 1N ? noise vector.

The signal-to-noise ratio (SNA) is defined as 22()/()s n γσσ=. Assuming a

block-fading channel, i.e., the channel is static over a frame of T symbol blocks, our objection is to detect the transmitted symbols s for a given observation vector x assuming that the CSI: H is perfectly known at the receiver.

II MMSE-SIC

2.1 Conditional MMSE-SIC

We first apply the MMSE-SIC receiver to detect the transmitted symbols for MIMO systems.MMSE-SIC algorithm is initialized with a linear MMSE estimate. Then for each user, aninterference-canceled signal is constructed by removing inter-user interference.Since the estimated symbols at each iteration are not perfect, there will still be interference from other users. This interference is modeled as Gaussian [2], [3] and the residual interference plus noise power is estimated. Using this estimate, an MMSE filter conditioned on filtered output from the previous iteration is computed for each user. The bias is removed and a soft MMSE estimate of each symbol given the filtered output is propagated to the next iteration. Next, MMSE-SIC detection process is explained in details:

Step1:denote ,k l w as the MMSE weight vector of user k at the lth iteration, and assume that the SNR γ is perfectly known at the receiver. In the zeroth iteration, the conventional MMSE receiver is applied for user k , which is given by

1

,01H H k k k j j N k j k w h h h h I h γ-≠??=++????∑ (2) The output of the MMSE receiver after bias removing is given by

,0,0,0,0H k k k k k w x s s e α∨=

=+ (3) Where ,0,0H k k k w h α= is the signal amplitude after MMSE receiver if no bias removing

is carried out, and ,0,0,0,0(1)/()[]H H k k k j j k j k e w h s w n α≠=+∑ contains the residual

interference and noise after bias removing. The variance of ,0k e is expressed as

22,0,0

,02,0[H ]w H H k k k s n N k k k w H I σσσα∨+= (4)

Where k H denotes the matrix with the kth column being deleted. Thus, the output SINR is given by

2

1,02,0[H (1/)I ]H H s k k k k N k k h H h σβγσ-∨==+ (5)

Step2: having acquired the output of the MMSE receiver, the log-likelihood ratios (LLRs) of the modulation candidates are then derived. The soft estimate of user k ’s symbol at the zeroth iteration can be generated as

{}11,00

,0,0,0

10,00(|s )|(s |)(|s )M M k i k i i k k k k i k i M i k k i i a p s a s E s s a p a s p s a ∨

--∨∨=∨-========∑∑∑ (6)

Where 01,,M αα- are the constellation points, and ()/P x y denotes the probability of event x occurs under condition y .When the signal dimension becomes large, the multiple access interference (MAI) after MMSE filtering and bias removing becomes asymptotic Gaussian [2],[3]with mean zero and variance 2,0k σ∨

, thus

2,0,02,0,01(|s )exp()2k i k k i k k s a p s a σ∨∨∨∨-==

- (7)

,1k l S +

X

Fig.1. Block diagram of MMSE-SIC receiver for detecting user k at the ()1l th +iteration. At this point, we can assume that we have obtained the soft estimates ,k l s

’s for all

users at the lth iteration. The block diagram of the MMSE-SIC receiver is shown in Fig. 1 where soft cancellation of the MAI is carried out to produce the output for

detecting user k for subsequent iterations. This output is given by

,,,1(s )n j l j l k l j k k j j j k j k y x h s h s h s +≠≠=-=+-+∑∑

(8)

Given the above output, subsequent iterations shall be carried out with a new optimal MMSE weight vector for every user k .The optimal MMSE weight vector ,1k l w + is obtained to minimize the mean-square-error (MSE) between ,1,1H k l k l w y ++and k s is given by

1,1,1[h ]H H k l k k k k l k N k w h H D H I h γ-+=++

(9)

where ,1,1,1,,(d ,,d ,d ,d )k l l k l k l K l D diag -+= (10) with

{}222,,,,,2211||j l j l j l j l j l

j j s s d E s s s E s s s σσ∨∨??????=-=-?????????? (11) In (11), ,j l d defines the normalized residual interference power for user j , which will vary for every symbol interval and every iteration, thus matrix ,k l D is termed instantaneous interference power matrix. Following (2) and (5), the MMSE output and output SINR for ()1l th + iteration can be calculated accordingly.

Step3:soft interference cancellation can be carried out iteratively. The above receiver is referred to as the conditional MMSE-SIC (C-MMSE-SIC)receiver [1].

2.2 U-MMSE-SIC

Let us replace the instantaneous interference power matrix in (10) with a scaled identity matrix ,k l D in (10) with a scaled identity matrix

2,,1(1)I k l k l k D δ-=- (12)

Where

{}

222,,,211|(k 1)j l j l k l j j k s E s s s δσ∨≠??=--??-??∑ (13) Here,,k l D is an approximation of ,k l D and 2,k l δcan be considered as the average

reliability for all users except user k , at iteration l . Thus, we can then design the weight vector as

1

,,11H H k l k l k k k k N k w h h H D H I h γ-+??=++???

? (14) Note from (11) and (13), we have ,,(D )Tr()k l k l Tr D =. Thus, in (12), we in fact use the average residual interference power to replace the instantaneous residual interference powers for all interferers. The above receiver is termed the unconditional MMSE-SIC(U-MMSE-SIC) receiver.Note this receiver is differentfrom the one from

[4] where the averaging is calculatednot only over all users, but also over time. By doing so, the receiver may introduce longer processing delay. Finally, the definition of in (13) will enable us to link the U-MMSE-SIC receiver with the BI-GDFE receiver, which requires further research.

Using the above weight vector to obtain the output ,1,1,1H k l k l k l s w y ∨+++=, the corresponding variance 2

,1k l σ∨+of noise-plus-interference after bias removing can be generated as 22,1,,1,12,1[H ]w H H k l k k l k s k l k l k l w D H σσα∨++++=

(15)

At each iteration, the soft symbols ,k l s and the residual interference matrix ,k l D will vary. Thus,similar to C-MMSESIC,the U-MMSE-SIC receiver requires the calculation of the MMSE weights for each user, each symbol interval, and each iteration.Matrix inversions need to be computed for every user, and every iteration. We canemploy the matrix inversion lemma in order to reduce the number of matrix inversions.

III TABU SEARCH

3.1 OriginalTS

We apply the Tabu Search (TS) receiver to detect the transmitted symbols for MIMO

systems.

According to ML criterion, the objection of Tabu Search (TS)detection is to find a vector with a smallest metric, which is defined as:

()2

i i f x Hx y =- (16) Each transmitted vector maps a certain point in the hyperspace and has a metric as the definition (16). We can give the definition of the neighbors for a certain transmitted vector m x , as:

(){}|m m x N x x x ∈-=? (17)

Where ? is the minimum distance between two constellation points in a plane. That is to say, the neighbor of m x only has one different element from it.

As we all know, TS can be considered to some extent as a local search strategy with a tabu list, which records the visited points. Then the search can start from a random initial point and is extended to its neighbors.

For the sake of illustration, we take a simple system with BPSK and 4t N =for example to explain TS process.

Step1:0(1,1,1,1)x -+-+is chosen as the initial solution. According to the definition of neighborhood (23), the four yellow points (1(1,1,1,1)n ++--2(1,1,1,1)n ----3(1,1,1,1)n -++-4(1,1,1,1)n -+-+) are 0'x s neighbors, because their Hamming distance to 0x is 1.Their metrics are calculated respectively according to (22). Among them, 3n is chosen as the new candidate solution due to its minimum metric and renamed as 1x , while 0x is recorded in the tabu list.

Step2: Now, {}1123(x ),,N n n n =is 1'x s neighbor set. 0x is prohibited so that it isn’t regarded as 1'x s neighbor. Assuming ()()()()()()3132&f n f n f n f n ≤≤,3n becomes thenew candidate and 1x is added to the tabu list. It is noted that ()3f n isn’t always smaller than ()1f n

Step3:This process will continue to find the new candidatesolution. When the tabu list with a length P is full, the new prohibited solution will push the first point out of the list, like the operation in queue. At the same time, the pushed-out point is free and added to the constellation set again. Finally, the algorithm will be terminated when the maximum number of iterations M is reached. The final solution is selected from the

candidate solution in the tabu list according to their metric.

As for initialization, intuitively, the complexity of TS will increase with the increase of P and M. Owning to the limited P and M, the search won’t access all the points in the search space sometimes. As a result, it isn’t guaranteed that the transmitted vector is detected without error, even if there is no noise in the channel. A method of avoiding the problem caused by the limited P and M is to choose the solution of traditional detection (such as ZF/MMSE detection or ZF/MMSE-OSIC detection [5]) as the initial solution of TS. Then, the whole flow of TS based on traditional detection can be depicted in Fig.2.

Fig.2 Flow of TS based on original detection

3.2 Improvement on TS

In this part, Improvement on TS detection is proposed to obtain a tradeoff between the TS detecti on’s performance and complexity.

In TS process, a variable s m is used to record the historical smallest metric. Namely, only if the metric of the new candidate solution is smaller than s m ,s m is renewed as the smaller metric. In Fig.3, we give an example of 's m s changing curve with the iteration number. It has a decreasing tendency with the increase of iteration number roughly. But because the metric of new candidate solution is not always smaller than the former solution,s m may be changeless in a period. So we can command that the search should be stopped in advance if the value of s m keep changeless after a certain number of iterations. The cutout parameter r is defined as the ratio of this certain iteration number to M. In Fig.3, cutout parameter r is chosen as 0.4, which means that the search will be cutoff if s m keeps invariable in the continuous 8 iterations, even if the number of iteration doesn’t reach M=20. This method can cut down the algorithm’s complexity greatly with little loss in performance.

Fig.3 m s in each iteration and the definition of cutout parameter r

3.3 computation analysis

In all the detection algorithms based on searching, most of the computation is consumed to calculate the metric of each point defined by (22). As to ML detection, each calculation will cost ()20t N times operation of complex multiplication for each

024681012

14161820iterationnumber m

s

point and there are 2c t M N points at all. For TS detection, the complexity of solving the initial solution is constant. Sequentially, owing to the definition of neighborhood, the metric of the current solution can be obtained based on the known metric of the last solution, by only substituting the metric of the unique different element. Then the metric-calculation of a new solution only needs ()0t N times operation of complex

multiplication. Therefore, in TS process with M iterations, to calculate the metric will expend the total ()20t c N M M ?? times operation of complex multiplication. Here M

is irrelative with the channel state. So the complexity of TS detection won’t vary with SNR. The choice of M is experiential according to the different t N and c M . When the method of skipping out of the search is adopted, the multiplicative parameter after 2t N will be smaller than c M M ?very much. By adjusting the value of M and r , a tradeoff is obtained between the TS detection’s performance and complexity. The computations for ML and TS are listed as Tab.1.

Table https://www.wendangku.net/doc/ab17120069.html,parison between ML and TS

References

[1] X. Wang and H. V. Poor, “Iterative (turbo) soft interference cancellationand decoding forcoded it,” IEEE Trans. Commun., vol. 47, no. 7,pp. 1046–1061, Jul. 1999.

[2] J. Zhang, E. Chong, and D. Tse, “Output MAI distributions of linearMMSE multiuserreceivers in DS-CDMA systems,” IEEE Trans. Inf.Theory , vol. 47, no. 3, pp. 1028–1144, Mar. 2001.

[3] D. Guo, S. Verdú, and L. K. Rasmussen, “Asymptotic normality ofl inear multiuser receiver outputs,” IEEE Trans. Inf. Theory , vol. 48,no. 12, pp. 3080–3095, Dec. 2002.

[4] G. Caire, R. R. Müller, and T. Tanaka, “Iterative multiuser jointdecoding: Optimal power allocation and low-complexity implementation,”IEEE Trans. Inf. Theory , vol. 50, no. 9, pp. 1950–1973, Sep.2004.

[5] A. Paulraj, RohitNabar, Dhananjay Gore, “Introduction to space -timewireless

communications”, Cambridge University Press, 2003.

[6] H.Zhao, L.Yang, and W Wang, “Tabu Search Detection for MIMO Systems”International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC07), Jan 2007.

[7] Peng Hui Tan and Lars Rasmussen, “Multiuser detection in CDMA – acomparison of relaxations exact and heuristic search methods”, IEEETrans. On wireless communications, V ol.3,No.5, 2004,pp:1802-1809.

[8] Hui Zhao, Hang Long, Wenbo Wang, “Combined-Weight SphereDecoder in Bad Channel”,International Symposium on WirelessPervasive Computing(ISWPC), Jan 2006.

[9] A. M. Chan and G. W. Wornell, “A class of block-iterative equalizersfor intersymbolinterference channels: Fixed channel results,” IEEETrans. Commun., vol. 49, no. 11, pp. 1966–1976, Nov. 2001.

[10] A. Paulraj, R. Nabar, and D. Gore, IT Introduction to Space-TimeWireless Communications. Cambridge, U.K.: Cambridge Univ.Press, 2003.

[11] Foschini G.J., Layered Space-Time Architecture for WirelessCommunication in

a Fading Environment When Using Multi-ElementAntennas, Bell Labs Technical Journal, 1996,Vol.1,pp. 41-59

[12] T.Fukatani, R, Matsumoto and T. Uyematsu, “ Two methods for decreasing the computational comp lexity of the MIMO ML decoder”,IEICE Trans. Fundamental,Vol.E87-A, No.10, Oct.2004, pp.2571-2576

华为AAU的有源天线解决方案

华为AAU的有源天线解决方案 国内三大运营商经过十多年从2G到3G移动网络建设、升级、改造,相当数量的站点空间变得越来越紧张,新设备的增加将极大加重运营商的建站和运维成本,为满足日益增长的移动宽带应用,需要运营商能迅速扩充网络容量以满足业务增长。 当前国内LTE建设迅速起步, 需要在现有的移动站点基础上,再加入一层LTE(FDD和TDD)网络,对站点机房空间,天面位置都提出了新的要求,据华为内部调查和经验,大部分一线和二线城市45%左右的站点天面空间拥挤,无法新增独立天线。 在多频多模建网模式下,如何高效利用有限的站点资源,进一步提升网络容量,为用户提供更好的宽带业务体验,成为各运营商面临的主要挑战。 以华为AAU为代表的有源天线解决方案,正在逐步成为行业的趋势。从90年代的GSM宏基站,到2004年华为首先推出分布式基站,华为一直通过基站形态的创新适应移动网络建设,在移动宽带MBB时代,需要进一步提升网络性能、解决天面获取难题的困扰,华为正在引领行业推动有源天线标准化。 AAU称做有源天线单元,如下图1所示,在多个频段组网下,传统方式需要选择两个RRU连接到一个无源天线, 采用AAU后,2个RRU集成到天线中,形成有源天线单元AAU。 图1 RRU和天线集成于一体的AAU 对于一个已有RRU和天线的站点,如果要增加LTE业务,将需要新增加一套新的RRU、天线、以及相关的附件, 而华为的AAU解决方案可以将新频段的LTE RRU集成在AAU内部,同时集成原来的两副天线,如下图2所示。

图2 高度集成的AAU替换传统大量的零散部件 华为的AAU,有效整合运营商的天面资源,简化了天面配套要求,将射频单元与天线合为一体,减小馈线损耗,增强了覆盖效果,更加适合多频段多制式组网的需求,有效保护了运营商比机房更重要的核心资产--天面资源。采用AAU 解决方案后,整个天面变得简洁、可靠、稳定,带来的好处有: 部署方便,节省空间,AAU尺寸和单频天线相当,降低选址和物业协调难度,同时集束线缆设计,AAU与原AAU连接仅需4根馈线。在节省70%的空间下,能够获得30%到70%的容量增益。 管理效率高, AAU本身支持多种电调模式,手动、近端、远端都可以方便地对天线进行调整,远端方式通过AISG接口实现和远端网管通信免进站,免上塔,提升维护效率可以实时调整,避免业务中断。 省钱省时,实现快速建设,通过一次部署,降低了物业协调难度,能够极大地减少抱杆、土建及楼面的租金成本,减少安装工程成本,据测算,采用AAU 能够减少30%的站点建设成本。 华为的AAU可集成两个不同频段的射频单元以及天线, 集成度是业界同类产品的2倍,帮助运营商平滑演进到HSPA+、LTE以及LTE-Advanced而无需新增射频单元或天线,每个站点最大可以节省75%的模块数量,也是业界唯一能够支持2个4×4 MIMO射频模块的产品,并支持波束成型技术,可实现85%的容量提升。 AAU解决方案,不仅是对基站架构、天面安装方式的创新,而且反映AAU产品稳定性、可靠性达到了相当的水平。AAU作为高度集成的产品,安装位置又很特别,要求其内部有源模块必须达到非常高的可靠性。华为将其十多年对RRU

华为微波天线调测指导书

天线调测指导书 (仅供内部使用) 拟制:邢子彬日期:2009-03-30 审核:日期:yyyy/mm/dd 审核:日期:yyyy/mm/dd 批准:日期:yyyy/mm/dd 华为技术有限公司 版权所有侵权必究

修订记录

天线调测指导书 关键词:天线、主瓣、旁瓣、接收电平 摘要:介绍了天线主瓣与旁瓣相关知识,以及单极化天线和双极化天线的调整方法。 缩略语清单: 一、主瓣和旁瓣 在对调天线前,需掌握天线主瓣和旁瓣的相关知识。 1、主瓣和旁瓣的定义 天线辐射的电场强度在空间各点的分布是不一样的,我们可以用天线方位图来表示。通常取其水平和垂直两个切面,故有水平方向图和垂直方向图,如图1所示为垂直方向图。方向图中有许多波瓣,最大辐射方向的波瓣叫主瓣,其它波瓣叫旁瓣,旁瓣中可以影响对调天线的是第一旁瓣。 图1 主瓣和旁瓣 2、定位主瓣

微波天线的主瓣宽度很窄,通常在0.6~3.7度之间,例如:一个1.2m的天线(工作频率为23 GHz),信号电平从主瓣信号峰值衰减到零只有0.9度的方位角。所以在定位主瓣的时候,一旦检测到信号,则只需要对天线做微调即可。 在对调天线扫描过主瓣的时候,信号电平要经历一个快速变化的过程,通过比较接收到的信号峰值可以确定天线主瓣是否对准,通常情况下主瓣信号峰值比第一旁瓣的信号峰值高20~25dB。当两端天线同时收到对端的主瓣信号,如果两个信号强度差在2dB以内,属于允许范围。 如图2是天线在自由空间传播模型的正面图,旁瓣围绕在以主瓣为圆心的周围成放射状传播。 图2 天线水平方向图 3、扫描路径 在不同的俯仰角(方位角)上扫描信号时,扫描到的旁瓣信号有时被误认为主瓣信号。如图3是天线水平方向上的辐射模型,天线在三种不同仰角位置扫描到的信号电平值: 图3 三种扫描路径

频点与对应频率【更新版】汇总

频点与频率 1、CDMA800系统载频信道号与中心频率的计算 上行频宽:825MHz~835MHz 下行频宽:870MHz~880MHz 载频中心频率计算公式: 上行载频中心频率=0.03MHz×信道号n+825MHz 下行载频中心频率=0.03MHz×信道号n+870MHz 具体对应关系如下: 2、GSM900系统频点与频率的计算 具体对应关系如下:

具体对应关系如下:

4、WLAN频率使用情况 ▲ 2.4GHz信道划分 802.11b和802.11g的工作频段在 2.4GHz(2.410GHz~2.483GHz),其可用带宽为83.5MHz,划分为13个信道,每个信道带宽22MHz。具体如下:

▲5.8GHz信道划分 802.11a的工作频段在5.8GHz(5.725GHz~5.850GHz),其可用带宽为125MHz,划分为5个信道,每个信道带宽20MHz。具体如下: 5、联通WCDMA 上行频宽:1940MHz~1955MHz 下行频宽:2130MHz~2145MHz 载频中心频率计算公式: 上行载频中心频率=信道号m÷5MHz 下行载频中心频率=信道号n÷5MHz 6、移动TD-SCDMA 目前中国移动TD-SCDMA系统可使用频率资源为85MHz,具体如下: A频段(2010~2025 MHz,原B频段):共计15MHz,可供全国范围室内室外覆盖使用。F频段(1880~1900MHz,原A频段):共计20MHz,可供全国范围室内室外覆盖使用。E频段(2320~2370 MHz,原C频段):共计50MHz,可供全国范围室内覆盖使用。 TD-SCDMA采用时分双工(TDD)模式,因此无上下行信号之分。 ▲A频段 频宽:2010MHz~2025MHz 载频中心频率计算公式: 载频中心频率=信道号n÷5MHz (该频段信道有华为系和大唐系两种不同划分方式)

各种移动通信制式频率与信道号之间的换算

各种移动通信制式 频率与信道号之间的换算

一、GSM信道与频率的换算 GSM多址方式:TDMA(时分多址) GSM双工方式:FDD(频分双工) GSM占用带宽:上下行各25MHz(上下行共用以FDD方式工作) GSM上下行频率隔离:45MHz GSM信道间隔:200KHz 移动占用带宽:上下行各19 MHz 上行:890MHz ~909MHz 下行:935MHz ~954MHz (1 ~ 95)联通用带宽:上下行各6 MHz 上行:909MHz ~915MHz 下行:954MHz ~960MHz (95 ~ 124)GSM一般换算公式: 信道→频率:上行:890+CH×0.2=F 上行 (MHz) 下行:935+CH×0.2=F 下行 (MHz) 频率→信道:上行:(F 上行-890)×5= CH

下行:(F 下行 -935)×5= CH GSM工程算法: 低端信道号(即移动较低频率点信道号)的算法:可采用一般换算公式高端信道号(即联通或移动较较高频率点信道号)算法: 频率→信道:下行:(F 下行 -954)×5+95= CH 上行:(F 上行 -909)×5+95= CH 信道→频率:下行:[(CH - 95)×0.2]+954=F 下行 上行:F 下行–45= F 上行 注:GSM中95频点为保护频点,无委规定联通、移动均不能占用,因此该频点内信号较为干净如做模拟测试可考虑采用该频点。

二、CDMA信道与频率的换算 CDMA多址方式:CDMA(码分多址) CDMA双工方式:FDD(频分双工) CDMA占用带宽:上下行各10MHz(上下行共用以FDD方式工作) CDMA上下行频率隔离:45MHz CDMA信道间隔:1.23 MHz CDMA带宽:上行:825MHz ~835MHz 下行:870MHz ~880MHz (37~283) 现联通所用CDMA-IS95制式为美国高通制定,当时美国为实现AMP(模拟制式)向CDMA的平滑过渡因此定采用双制式兼容方案,即使用同时支持AMP 和CDMA的双模手机,并让AMP退出部分频率资源给CDMA使用。因此CDMA 信道编号沿用AMP制式的编号方案。(CDMA信道号指与该载波中心频点相对应的AMP信道号,因此CDMA信道频率换算可参考AMP信道编号的算法)AMP起始频点:869 MHz

通信领域频率划分总结

B a n d M o d e U p p e r F r e q L o w e r F r e q B1 WW Core 1920-1980 2110-2170 B2 US PCS 1850-1910 1930-1990 B3 Euro DCS 1710-1785 1805-1880 B4 US AWS-I 1710-1755 2110-2155 B5 US Cell 824-849 869-894 B6 Japan Cell 830-840 875-885 B7 Euro LTE 2500-2570 2620-2690 B8 Euro EGSM 880-915 925-960 B9 Japan DCS 1750-1785 1845-1880 B10 Lat Am AWS 1710-1770 2110-2170 B11 Japan LTE 1426-1453 1475-1501 B12 US.7 A-C Lower 698-716 728-746 B13 US.7 C Upper 777-787 746-757 B14 US.7 D Upper 788-798 758-768 B17 US B-C.7 Lower 704-716 734-746 B20 Euro DDR 790-820 832-862 B33 LTE TDD 1900-1920 B34 China TD-SCDMA A 2010-2025 B35 LTE TDD 1850-1910 B36 LTE TDD 1930-1990 B37 LTE TDD 1910-1930 B38 LTE TDD 2570-2620 B39 China TD-SCDMA F 1880-1920 B40 China TD-SCDMA E 2320-2370 中国移动: GSM:上行890-909MHZ;下行935-954MHZ 频点:1-95 EGSM:上行885-890MHZ;下行930-935MHZ 频点:999-1024 DCS1800:上行1710-1720MHz,下行1805~1815MHz以及上行1725-1735MHz,下行1820~1830MHz 频点:512-561以及587-636 1805-1825 1710-1730 TD-SCDMA:1880 MHz~1920MHz(F频段),2010 MHz~2025 MHz(A频段),2300 MHz~2400 MHz (E频段) 中国联通: GSM:上行909-915MHZ,下行954-960MHZ 频点:96-125 DCS1800:上行1740-1755MHz,下行1835~1850MHz 频点:662-736 WCDMA:1940MHz-1955MHz(上行)、2130MHz -2145MHz(下行),上下行各15MHz。相邻频率间隔间隔采用5MHz时,可用频率是3个。WCDMA频点计算公式:频点号=频率×5 上行中心频点号: 9612~9888 下行中心频点号:10562~10838 中国电信: CDMA:825MHz-835MHz ,870MHz-880MHz 共7个频点:37,78,119,160,201,242,283 ;其中283为基本频道,前3个EVDO频点使用,后3个CDMA2000使用;160隔离

信道号与工作频率对应一览表

(1)W-CDMA(FDD):(UE/BS,ARFCN)

IMT2000:1920~1980/2110~2170,10562~10838 PCS1900:1850~1910/1930~1990, 9662~9938&412&437&462&487&512&537&562&587&612&637&662&687 DCS1800:1710~1785/1805~1880,9037~9388 (2)TD-SCDMA China:1785~1805,1880~1900,1900~1920,2010~2025,2300~2400 3GPP:1900~1920,2010~2015 (3)HSDPA:(UE/BS) IMT2000:1920~1980/2110~2170(832~870MHz) PCS1900:1850~1910/1930~1990 DCS1800:1710~1785/1805~1880 (4)IS95A/B:(MS/BS) US/Korea:824~849/869~894 Japan:887~925/832~870 US:1850~1910/1930~1990 Korea:1750~1780/1840~1870 (5)CDMA2000(1xRTT,1xEV-DO,1xEV-DV):(MS/BS) IS95并增加 NMT450:411~483/421~493 GSM/GPRS/EDGE(UL/DL,ARFCN): GSM450:450.4~457.6MHz/460.4~467.6MHz,259~293 GSM480:478.8~486MHz/488.8~496MHz,306~340 GSM750:777~792MHz/747~762MHz,438~511 GSM850:824~849MHz/869~894MHz,128~251 E-GSM:880~915MHz/925~960MHz,975~1023&0~124 P-GSM:890~915MHz/935~960MHz,1~124 R-GSM:876~915MHz/921~960MHz,955~1023&0~124 DCS:1710~1785MHz/1805~1880MHz,512~885 PCS:1850~1910MHz/1930~1990MHz,512~810 TETRA(MS/BS): 380~390,410~420,450~460,870~915MHz/390~400,420~430,460~470,915~950MHz Bluetooth: 2400~2483.5MHz 802.11B:2.4~2.4835GHz 802.11G:2.4~2.4835GHz 802.11A/H/J: 4.9~5GHz(Japan) 5.03~5.09GHz(Japan) 5.15~5.35(UNII) 5.47~5.725GHz 5.725~5.825GHz(ISM,UNII) 802.15.3A:3.1~10.6GHz 802.16A:2~11GHz 802.16E:<5GHz 825-835MHz/870-880MHz 联通CDMA 885-915MHz/930-960MHz 联通GSM、移动GSM 1710-1725MHz/1805-1820MHz 移动DCS

华为AAU3910 产品概述-天线

AAU3910概述 文档版本01 发布日期2013-08-31

版权所有? 华为技术有限公司2013。保留一切权利。 非经本公司书面许可,任何单位和个人不得擅自摘抄、复制本文档内容的部分或全部,并不得以任何形式传播。 商标声明 和其他华为商标均为华为技术有限公司的商标。 本文档提及的其他所有商标或注册商标,由各自的所有人拥有。 注意 您购买的产品、服务或特性等应受华为公司商业合同和条款的约束,本文档中描述的全部或部分产品、服务或特性可能不在您的购买或使用范围之内。除非合同另有约定,华为公司对本文档内容不做任何明示或默示的声明或保证。 由于产品版本升级或其他原因,本文档内容会不定期进行更新。除非另有约定,本文档仅作为使用指导,本文档中的所有陈述、信息和建议不构成任何明示或暗示的担保。 华为技术有限公司 地址:深圳市龙岗区坂田华为总部办公楼邮编:518129 网址:https://www.wendangku.net/doc/ab17120069.html, 客户服务邮箱:support@https://www.wendangku.net/doc/ab17120069.html, 客户服务电话:4008302118

AAU3910概述目录 目录 1 概述 (1) 1.1 外观 (1) 1.2 物理接口 (2) 2 技术指标 (4) 2.1 频段 (4) 2.2 容量 (4) 2.3 接收灵敏度 (5) 2.4 EIRP (5) 2.5 功耗 (5) 2.6 电源 (6) 2.7 风阻 (6) 2.8 整机规格 (6) 2.9 环境指标 (6) 3 缩略语表 (8)

1 概述 多年以来,基站的架构正在发生不断的变化。相比于RFU模块,华为创新提出的RRU 解决方案,支持近天线安装,降低约3dB的馈线损耗的同时还可降低发射功率,提升网 络覆盖率。射频模块的功能向天线上移,已经成为业界的一种趋势,当前越来越多的运 营商选择全网使用RRU解决方案。 AAS有源天线系统(Active Antenna System)是继RFU、RRU之后衍生出的一种新型射 频模块形态。其主要特征在于AAS将原有的RRU单元功能和天线的功能合并,简化站 点资源;另外采用射频多通道技术对天线阵列进行控制,通过不同波束赋形的方式可以 达到改善无线信号覆盖质量以及提升网络容量的目的。 随着MBB(Mobile Broadband)的快速发展,多频多模建站渐渐成为业界主流。随着运 营商站点复杂度的增加以及站点上模块数量的增多,运营成本变得越来越高,所以简化 站点正在变成当前多频多模建网的一种趋势。为了响应这种趋势,华为在AAS技术上 率先投入,并且推出AAU系列化解决方案产品。 AAU3910是华为实现AAS技术的主打产品,相比于华为在RRU上的创新,该产品使 用有源天线技术,将射频模块的功能进一步上移到天线中。 AAU3910最大可集成2个有源频段。 AAU3910通过射频模块的上移,可节省大约0.8dB的跳线损耗,同时节能10%。 AAU3910简化当前的站点安装方式,相比传统的RRU安装方式,可节省安装时间。 AAU3910支持上行多接收,消除上行容量的瓶颈。 AAU3910作为有源天线产品,也可以外接RRU/RFU,通过替换现网天线,进一步简化 站点。 1.1 外观 AAU3910外观如图1-1所示。

2017年华为天线产品手册-中国移动

华为技术有限公司

创新共赢,助力中国移动打造LTE精品网络 随着LTE精品网络的深入建设,以及无线网络的不断演进,天馈系统已成为建设LTE精品网络的重要组成部分。 ●在业务需求日趋旺盛,潜力巨大的农村郊区广域场景,通过高增益天线解决方案,增强覆 盖效果,增加用户数量,提升用户体验;在高铁等特殊场景,通过高增益、窄波束天线专网解决方案保证覆盖效果,提升用户体验,扩大市场份额; ●针对网络容量不足,需精准覆盖的城区场景,通过多频天线解决方案实现独立电调,提升 精准覆盖;另外针对站点获取困难,急需补盲覆盖的城区场景,通过美化天线方案,降低站点获取难度,提高部署效率,提升用户体验; ●针对天面空间紧张场景,通过全频段智能天线进行天面收编,简化站点从而降低站点租金 和维护成本; ●随着4G站点规模继续增长,天线数量不断增加,通过天线智能化管理方案,可大幅提升 网络部署与管理效率,有效降低网络管理与维护成本。 华为依托20多年丰富的无线网络经验积累和对4G网络部署与发展的深刻理解,采用天线与RAN协同设计,聚焦整网性能最优,跨界创新推出了一系列无源和有源天线解决方案,助力中国移动打造持续领先的LTE精品网络。

全频段智能天线解决方案,打造用户体验更好的精品网络 全频段智能天线可对当前复杂天面进行收编,解决天面紧张问题。华为全频段智能天线, 两大系列产品:2288天线和4488天线,天线可同时支持900M、1800M、FA频段8T8R以及D频 段 8T8R,最大程度节省天面空间及TCO;同时,采用创新天线阵列架构设计,使天线尺寸更小,降低部署难度,并且保证各系统增益满足覆盖要求。 3D电调及美化天线解决方案,可独立优化、灵活调整,提升网络性能华为FA/D频段3D电调天线解决方案,支持FA与D频段同时接入,独立调节,解决天面空间 紧张问题,同时结合华为创新的EasyBeam解决方案,针对不同覆盖场景,实现天线波束的3D远程调节,包括:广播波束水平方位角连续调整、水平波瓣宽度远程调整和天线电下倾角度远程 连续可调。另外,华为美化天线解决方案降低站点获取难度,使网络部署更灵活,提高部署效率,提升用户体验。 高增益天线解决方案,实现农村郊区场景广覆盖 华为高增益天线解决方案,针对农村郊区等需要广覆盖的地区,依托现有站址资源,加速 广覆盖进程,通过阵列优化设计,提高天线增益2.5~3dB,有效提高覆盖半径。 高铁天线解决方案,实现专网覆盖,使部署与管理更高效 华为高铁天线解决方案,针对高铁场景,支持F与D频段一次部署,满足未来演进;同时采 用高增益窄波束天线对高铁沿线进行专网覆盖,减小对宏网的干扰,提升了网络部署与优化效率、提高网络性能。

信道号与频率对应关系(网络)

无线通信各制式频段划分 GSM900频率划分与频点说明: 说明:1、EGSM为GSM扩展频段。能扩展频段的公司为中国移动,目前部分省市已经将使用频段扩展到EGSM频段。 2、GSM上行频段890-915MHZ,下行935-960MHZ。 DCS1800频率划分与频点说明: 说明:1、早起移动频段的上、下行的划分带宽为10MHZ(上行1710-1720MHZ,下行1805-1815MHZ),后将频段扩展为20MHZ。 2、虽然早期在频段规划中的全频段为上行1710-1785MHZ下行1805-1880MHZ,但在3G频段划分中将1755-1780MHZ,1850-1880MHZ重新划分为3G的扩展频段。 IS95 CDMA频率划分与频点说明: 说明:下行频率与上行频率一一对应,因IS95中工作频率带宽为1.23MHZ,故采用的频点间隔为41,考虑到频带保护,规划用频点为283,242,201,160,119,78和37。

我国3G频率划分: 一、WCDMA频点号与频率对应关系: 根据工信部规定,中国联通可用的频段 是1940-1955MHZ,2130-2145MHZ,上下行各15MHZ。相邻频率间隔5MHZ时,可用频率为3个。 载波频率是由UTRA绝对无线信道号指定的。 UTRA绝对无线频率信道号 根据可用频段和绝对无线频段信道号计算公式,中国联通可用的频率号见下表: 二、TD-SCDMA频点号与频率对应关系: 目前使用的TD频段为2010~2025MHZ,总共15M,每个频点是1.6M,这样每5M就是3个频点,前3个频点留作室内分布使用,后面6个用作室外基站使用。第一个频点前面和第九个频点后面留有0.2M的保护频带,室内频点和室外频点之间也有0.2M的保护频带,这样第一个频点就是2010.2~2011.8,中心频点是2011,换算为频点号就是用中心频率乘以5,即10055,其他类推。(与WCDMA方法相同,只是频率不同) 三、CDMA2000频点号与频率对应关系:

RRU专题介绍-华为案例

目录 1.概述 (2) 1.1.分布式基站结构 (2) 1.2.RRU产品介绍 (3) 1.2.1.覆盖能力 (4) 1.2.2.组网 (4) 1.2.3.安装 (5) 1.2.4.环境适应 (5) 1.2.5.更软切换 (6) 1.2.6.产品外观 (6) 1.2.7.RRU3801C分布式基站支持的典型配置类型 (7) 1.3.基站与RRU连接方式 (3) 1.4.分布式基站解决方案 (4) 1.4.1.解决方案一 (8) 1.4.2.解决方案二 (9) 1.4.3.解决方案三 (9) 1.5.分布式基站所能带来的好处 (10) 1.5.1.解决站址选取困难的问题 (10) 1.5.2.解决低成本快速建网的问题 (10) 1.5.3.满足降低运营成本的需求 (10) 1.5.4.满足充分利用原有设备投资的需求 (10) 1.5.5.解决传统宏基站安装复杂的问题 (10) 1.5.6.提供简单的升级换代方案 (11) 1.5.7.提供多模基站产品形态的解决方案 (11) 1.5.8.满足高可靠性的要求 (11) 2.RRU应用实例 (11) 2.1.实例一:四川移动祥福苑分布式RRU使用 (11) 2.2.实例二:四川移动中海名城分布式RRU使用 (12) 3.RRU与直放站的比较 (12) 4.总结 (14)

1.概述 在现有的2G无线网络实际建设中,我们已出现一些难点,如城区选址困难、现有的2G 机房内设备拥挤、区乡的大面积覆盖投资过于巨大等,在未来的3G商业网建设中,我们就不得不考虑到以上这些2G建设中已出现的问题。 由华为公司提出的分布式基站解决方案能够为运营商提供一流的低成本快速建网解决方案。华为分布式基站由RRU(Radio Remote Unit)和BBU(Base Band Unit)组成。RRU 与BBU分别承担基站的射频处理部分和基带处理部分,各自独立安装,分开放置,通过电接口或光接口相连接,形成分布式基站形态。 RRU是室外型射频拉远模块(除与BBU对接外,还可作为宏基站的拉远模块)。RRU 可以直接安装于靠近天线位置的金属桅杆或墙面上,具有体积小、重量轻、安装简单方便的特点。 BBU是采用小型化设计的盒式设备,可安装于任何具有19英寸宽,1U高空间的标准机柜中,具有占地面积小、易于安装、功能全面、功耗低的特点,便于与现有站点共存,并且支持堆叠方式扩展容量。 RRU与BBU都基于3GPP R4/R5 FDD协议开发,能够针对运营商的不同需求、不同网络环境提供WCDMA无线接入网络的解决方案,满足城市、郊区、农村、高速公路、铁路、热点地区等的无线覆盖的要求。 1.1.分布式基站结构 基带处理与RRU连接(下行) 华为分布式NodeB系统由BBU3806与RRU3801C以及天线与天馈系统组成。

手机通信制式频率和信道换算总结

手机通信制式频率和信道换算总结 一、WCDMA 根据工信部规定,中国联通可用的频段是1940MHz-1955MHz(上行)、2130MHz -2145MHz(下行),上下行各15MHz。 WCDMA的频点称为UARFCN(UTRA Absolute Radio Frequency Channel Number,UTRA绝对频点号)。 2.1GHz频段上行频点号为9612~9888,下行频点号为10562~10838,频点除以5就可以得到频点中心对应的频率值(以MHz为单位)。每个频点间隔为200kHz,与GSM系统兼容。 当然每个频点的带宽远超过200kHz,这与CDMA的频点编号方式类似。 目前联通WCDMA系统下行第一频点号为10713(中心频率2142.6MHz),第二频点号为10688,第三频点号为10663。 上行频点号分别为9763(中心频率1952.6MHz)、9738以及9713。 二、GSM

GSM多址方式:TDMA(时分多址) GSM双工方式:FDD(频分双工) GSM占用带宽:上下行各25MHz(上下行共用以FDD方式工作)GSM上下行频率隔离:45MHz GSM信道间隔:200KHz 移动占用带宽:上下行各19 MHz 上行:890MHz ~909MHz 下行:935MHz ~954MHz (1 ~ 95) 联通用带宽:上下行各6 MHz 上行:909MHz ~915MHz 下行:954MHz ~960MHz (95 ~ 124) GSM一般换算公式:

信道→频率:上行:890+CH×0.2=F上行(MHz) 下行:935+CH×0.2=F下行(MHz) 频率→信道:上行:(F上行-890)×5= CH 下行:(F下行-935)×5= CH GSM工程算法: 低端信道号(即移动较低频率点信道号)的算法:可采用一般换算公式 高端信道号(即联通或移动较较高频率点信道号)算法: 频率→信道:下行:(F下行-954)×5+95= CH 上行:(F上行-909)×5+95= CH 信道→频率:下行:[(CH - 95)×0.2]+954=F下行 上行:F下行–45= F上行 注:GSM中95频点为保护频点,无委规定联通、移动均不能占用,因此该频点内信号较为干净如做模拟测试可考虑采用该频点。 三、DCS信道与频率的换算 DCS双工方式:FDD(频分双工)

800兆赫集群系统频道序号与频率对应表

800兆赫集群系统频道序号与频率对应表(MHz) 频道序号基台发信移动台发信频道序号基台发信移动台发信 1 851.0125 806.0125 51 852.2625 807.2625 2 851.0375 806.0375 52 852.2875 807.2875 3 851.0625 806.0625 53 852.3125 807.3125 4 851.087 5 806.0875 54 852.3375 807.3375 5 851.1125 806.1125 55 852.3625 807.3625 6 851.1375 806.1375 56 852.3875 807.3875 7 851.1625 806.1625 57 852.4125 807.4125 8 851.1875 806.1875 58 852.4375 807.4375 9 851.2125 806.2125 59 852.4625 807.4625 10 851.2375 806.2375 60 852.4875 807.4875 11 851.2625 806.2625 61 852.5125 807.5125 12 851.2875 806.2875 62 852.5375 807.5375 13 851.3125 806.3125 63 852.5625 807.5625 14 851.3375 806.3375 64 852.5875 807.5875 15 851.3625 806.3625 65 852.6125 807.6125 16 851.3875 806.3875 66 852.6375 807.6375 17 851.4125 806.4125 67 852.6625 807.6625 18 851.4375 806.4375 68 852.6875 807.6875 19 851.4625 806.4625 69 852.7125 807.7125 20 851.4875 806.4875 70 852.7375 807.7375 21 851.5125 806.5125 71 852.7625 807.7625 22 851.5375 806.5375 72 852.7875 807.7875 23 851.5625 806.5625 73 852.8125 807.8125 24 851.5875 806.5875 74 852.8375 807.8375 25 851.6125 806.6125 75 852.8625 807.8625 26 851.6375 806.6375 76 852.8875 807.8875 27 851.6625 806.6625 77 852.9125 807.9125 28 851.6875 806.6875 78 852.9375 807.9375 29 851.7125 806.7125 79 852.9625 807.9625 30 851.7375 806.7375 80 852.9875 807.9875 31 851.7625 806.7625 81 853.0125 808.0125 32 851.7875 806.7875 82 853.0375 808.0375 33 851.8125 806.8125 83 853.0625 808.0625 34 851.8375 806.8375 84 853.0875 808.0875 35 851.8625 806.8625 85 853.1125 808.1125 36 851.8875 806.8875 86 853.1375 808.1375 37 851.9125 806.9125 87 853.1625 808.1625 38 851.9375 806.9375 88 853.1875 808.1875 39 851.9625 806.9625 89 853.2125 808.2125 40 851.9875 806.9875 90 853.2375 808.2375 41 852.0125 807.0125 91 853.2625 808.2625 42 852.0375 807.0375 92 853.2875 808.2875 43 852.0625 807.0625 93 853.3125 808.3125 44 852.0875 807.0875 94 853.3375 808.3375 45 852.1125 807.1125 95 853.3625 808.3625 46 852.1375 807.1375 96 853.3875 808.3875

频点与对应频率

频点与对应频率

频点与频率 1、CDMA800系统载频信道号与中心频率的计算 上行频宽:825MHz~835MHz 下行频宽:870MHz~880MHz 载频中心频率计算公式: 上行载频中心频率=0.03MHz×信道号n+825MHz 下行载频中心频率=0.03MHz×信道号n+870MHz 具体对应关系如下: 载频号信道号n 上行 (MHz) 下行 (MHz) 1 37 826.11 871.11 2 78 827.34 872.34 3 119 828.57 873.57 4 160 829.80 874.80 5 201 831.03 876.03

6 242 832.26 877.26 7 283 833.49 878.49 2、GSM900系统频点与频率的计算 频段信道号上行下行 GS M 0≤n≤12 5 FUL=890+0.2× n (890~914.8MH z) FDL=935+0.2× n (935~959.8MH z) E-G SM 975≤n≤ 1023 FUL=890+0.2× (n-1024) (880.2~889.8M Hz) FDL=935+0.2× (n-1024) (925.2~934.8M Hz) 具体对应关系如下:GSM 信道上 行 下 行 信 道 上 行 下 行 信 道 上 行 下 行 信 道 上 行 下 行 0 89 93 5 3 2 89 6.4 94 1.4 6 4 90 2.8 94 7.8 9 6 90 9.2 95 4.2 1 8993389946909499095

0.2 5.2 3 6.6 1.6 5 3 8 7 9.4 4.4 2 89 0.4 93 5.4 3 4 89 6.8 94 1.8 6 6 90 3.2 94 8.2 9 8 90 9.6 95 4.6 3 89 0.6 93 5.6 3 5 89 7 94 2 6 7 90 3.4 94 8.4 9 9 90 9.8 95 4.8 4 89 0.8 93 5.8 3 6 89 7.2 94 2.2 6 8 90 3.6 94 8.6 1 91 95 5 5 89 1 93 6 3 7 89 7.4 94 2.4 6 9 90 3.8 94 8.8 1 1 91 0.2 95 5.2 6 89 1.2 93 6.2 3 8 89 7.6 94 2.6 7 90 4 94 9 1 2 91 0.4 95 5.4 7 89 1.4 93 6.4 3 9 89 7.8 94 2.8 7 1 90 4.2 94 9.2 1 3 91 0.6 95 5.6 信道上 行 下 行 信 道 上 行 下 行 信 道 上 行 下 行 信 道 上 行 下 行 8 89 1.6 93 6.6 4 89 8 94 3 7 2 90 4.4 94 9.4 1 4 91 0.8 95 5.8

移动通信系统频点划分和频率规划

移动通信系统频点划分 、GSM900 (上下行差45MHz ) 说明: GSM频率在890M~915M (上行),935M~960M (下行),频点为0~124,其中95为临界频点。分配给移动公司的890M~909M,分配给联通公司的为909M?915M。其中对应移动的频点为0?94,联通的频点为96?124。 E-GSM 说明: GSM 频率在880M~890M (上行),925M~935M (下行),频点为975~1024,其中1024 为临界频点。 分配给移动公司的885M~890M ,未分配给联通公司。其中对应移动的频点为1000?1023。

1、GSM1800 (上下行差95MHz ) 说明: GSM 频率在1710M~1785M (上行),1805M~1880M (下行),频点为512~886。 分配给移动公司的1710M~1720M、1725M~1735M 共20M、100 个频点(其中1730- 1735MHz/1825-1830MHz 是07年信息产业部新批),而上海、广东、北京特殊分配了 1720M~1725M (据集团公司技术部2006年2月通信资源管理信息)。广西移动全网可使用的频点范围为512?562、586?636共100个频点,分配给联通公司的为1745M~1755M。(其中一些地市1735M-1745M 已经被联通占用) 1、频道间隔 相邻两频点间隔为为200kHz,每个频点采用时分多址(TDMA)方式,分为8个时隙,既8 个信道(全速率),如GSM采用半速率话音编码后,每个频点可容纳16个半速率信道,可使系统容量扩大一倍,但其代价必然是导致语音质量的降低。 2、频道配置 绝对频点号和频道标称中心频率的关系为: GSM900MHZ 频段: f1(n)=890.2MHz+(n-1)X 0.2MHz(移动台发,基站收) fh(n)=f1(n)+45MHz(基站发,移动台收);n€ [1 , 124] GSMI800MHZ 频段为: f1(n)=1710.2MHz+(n-512)X 0.2MHz(移动台发,基站收) fh(n)=f1(n)+95MHz(基站发,移动台收);n€ [512 , 885] 其中:f1(n)为上行信道频率、fh(n)为下行信道频率,n为绝对频点号(ARFCN)。

相关文档
相关文档 最新文档