WO2014089974A1 - 一种用于通信系统的频率优化方法 - Google Patents

一种用于通信系统的频率优化方法 Download PDF

Info

Publication number
WO2014089974A1
WO2014089974A1 PCT/CN2013/080093 CN2013080093W WO2014089974A1 WO 2014089974 A1 WO2014089974 A1 WO 2014089974A1 CN 2013080093 W CN2013080093 W CN 2013080093W WO 2014089974 A1 WO2014089974 A1 WO 2014089974A1
Authority
WO
WIPO (PCT)
Prior art keywords
cell
frequency
interference
probability
optimized
Prior art date
Application number
PCT/CN2013/080093
Other languages
English (en)
French (fr)
Inventor
周文千
何毓嵩
姚友群
Original Assignee
上海大唐移动通信设备有限公司
大唐移动通信设备有限公司
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 上海大唐移动通信设备有限公司, 大唐移动通信设备有限公司 filed Critical 上海大唐移动通信设备有限公司
Publication of WO2014089974A1 publication Critical patent/WO2014089974A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W24/00Supervisory, monitoring or testing arrangements
    • H04W24/02Arrangements for optimising operational condition

Definitions

  • the present invention relates to the field of communications technologies, and in particular, to a frequency optimization method for a communication system. Background technique
  • Frequency planning is an important means to reduce radio interference, improve the GSM network carrier-to-interference ratio, improve communication quality and network.
  • the frequency optimization is one of the most basic and important tasks to solve the GSM system network interference and optimize the network.
  • GSM automatic frequency change has become the main way of frequency optimization of the existing network. It realizes automatic frequency allocation through software platform, which greatly saves the tedious work of planners and improves work efficiency. At the same time, through comprehensive collection and analysis of existing network data, It also improves the quality of frequency planning.
  • the network in order to be able to perform power control and switching control of the mobile phone in the call, the network must obtain relevant information of the mobile phone, and the information is reported by the mobile phone.
  • the mobile phone in the GSM network in the call state, the mobile phone regularly reports to the network the measurement report of the serving cell and the neighboring cell measured by the mobile phone in a 480 millisecond period, and each measurement report mainly includes the BCCH and the signal of the serving cell. Level, call quality, and TA value, etc., in addition to the BCCH, signal level, and BSIC (network color code) of the six neighboring cells with the strongest signal. Based on these measurement reports, the BSC performs power control and switching control based on the power control and switching parameters defined by the network.
  • the mobile phone does not measure all the frequency points in the GSM frequency band at the same time.
  • the neighboring area information reported by the mobile phone is limited to the cell in which the BCCH frequency point is located in the BA table defined by the serving cell, that is, the neighboring cell list.
  • the interference matrix forms an interference matrix.
  • the automatic frequency allocation tool can determine whether the same frequency or adjacent frequency can be allocated according to the interference situation between different cells.
  • the automatic frequency distribution tool uses a genetic algorithm to continuously evaluate the frequency division scheme and repeat the frequency division. Eventually the interference value in the network is minimized.
  • the GSM-based MR-optimized interference matrix technology has the following disadvantages: The GSM narrow-band system that is only suitable for frequency points is allocated one by one, and the effect is not good for the 3G broadband system with limited frequency. The network topology is not tested. Cells reported by MR but geographically close may cause the same-frequency dry genetic (gene) algorithm to be inefficient, and it may not be the optimal solution after repeated iterations.
  • TD-SCDMA network With the development of TD-SCDMA network and the increasing number of 3G users, it is necessary to increase the number of sites and expand the capacity of the cell, especially for the small average station spacing, dense site distribution, ⁇ service hotspots, and high carrier frequency configuration. In the area, the interference of the network is often high. At the same time, due to frequent frequency adjustment and engineering construction, the frequent frequency adjustment and engineering construction cause interference in the frequency of the scrambling code. Continuous frequency optimization is needed to reduce the interference of the control channel and the traffic channel, and improve the network quality. And user perception. In view of this situation, the TD-SCDMA network requires periodic frequency optimization of the entire network to improve the performance index when the network load is increased. However, the traditional frequency optimization requires large-scale frequency sweep optimization test.
  • test range needs to traverse each cell, which consumes a lot of manpower and material resources, and can only cover the overlap of road cells. Therefore, the frequency optimization result is more biased towards road coverage. User distribution and mobility cannot be considered.
  • the problem to be solved by the present invention is: Traditional frequency optimization requires large-scale frequency sweep optimization test, and the test range needs to traverse each cell, which consumes a lot of manpower and material resources, and can only cover the overlap of road cells, so the frequency optimization result is more biased. In the case of road coverage, the actual user distribution and mobility cannot be considered.
  • the present invention provides a frequency optimization method for a communication system, the method comprising:
  • the steps S4 and S5 are performed according to the to-be-optimized cells in the order of the total interference probability of the primary carrier frequency.
  • steps S4 and S5 further include the following steps:
  • the total interference probability of the carrier frequency point is the sum of the products of all neighboring area IDs and their main carrier interference probabilities.
  • the method further includes:
  • step S5 If the allocation fails in step S5, the cell primary frequency point is directly deleted from the cell list to be optimized. If the allocation is successful, the primary and secondary carrier frequency information of the cell is updated according to the frequency group, and the reassigned from the to-be-optimized list is deleted. a cell of a frequency group;
  • the small interval secondary carrier interference probability XijO/ (Xij0+Xijl+Xij2)
  • XijO indicates the number of sampling points in the high-interference neighboring area
  • Xijl indicates the number of sampling points in the neighboring area
  • Xij2 indicates the number of sampling points in the low-interference neighboring area.
  • the high interference neighboring zone, the general interference neighboring zone, and the low interference neighboring zone are determined according to the level difference between the cell and its neighboring cell.
  • step S2 further includes:
  • step S23 further includes:
  • step S3 further includes:
  • a cell whose sum of the product of the neighboring cell ID and its main carrier interference probability is not zero is added to the to-be-optimized table.
  • the frequency grouping in the step S4 satisfies the following conditions:
  • the frequency points of each frequency group are different.
  • step S4 further includes:
  • the secondary carrier coefficient has a value range of [0, 1].
  • the frequency group is allocated to the cell in the step S5, the following conditions are also met: the primary frequency points of the cells in the same base station are different;
  • step S7 further includes:
  • the sum of the total interference probabilities of all the cell main carriers is calculated, and if it is lower than the sum of the total interference probabilities before the optimization, the end is ended; if not less than the sum of the total interferences before the optimization, then the selection is made.
  • the 10% cell with the highest total interference probability of the primary carrier is added to the to-be-optimized list, and steps S5-S7 are repeatedly performed for iteration.
  • the primary frequency point of the cell meets different primary frequencies of the same station cell, and the different frequency of the cell is played.
  • FIG. 1 is a flow chart showing a frequency optimization method in Embodiment 1 of the present invention
  • Figure 2 is a diagram showing the distance and relative angle between cells in the present invention.
  • Fig. 3 is a flow chart showing the frequency optimization method in the second embodiment of the present invention.
  • DETAILED DESCRIPTION OF THE INVENTION The present invention provides a frequency optimization method for a communication system. As shown in FIG. 1, the method includes obtaining measurement report data of a primary serving cell and its neighboring cells.
  • the frequency point with the smallest interference is used as the primary frequency point of the cell to be optimized, and the frequency group corresponding to the primary frequency point is allocated to the cell.
  • the steps S4 and S5 are performed on the to-be-optimized cells in the optimized list in descending order of the total interference probability of the primary carrier frequency.
  • steps S4 and S5 further include the following steps:
  • the total interference probability of the carrier frequency point is the sum of the products of all neighboring area IDs and their main carrier interference probabilities.
  • the method further includes:
  • step S5 If the allocation fails in step S5, the cell primary frequency point is directly deleted from the cell list to be optimized. If the allocation is successful, the primary and secondary carrier frequency information of the cell is updated according to the frequency group, and the reassigned from the to-be-optimized list is deleted. a cell of a frequency group;
  • the small interval secondary carrier interference probability XijO/ (Xij0+Xijl+Xij2)
  • XijO indicates the number of sampling points in the high-interference neighboring area
  • Xijl indicates the number of sampling points in the neighboring area
  • Xij2 indicates the number of sampling points in the low-interference neighboring area.
  • the high interference neighboring zone, the general interference neighboring zone, and the low interference neighboring zone are determined according to the level difference between the cell and its neighboring cell.
  • step S2 further includes:
  • step S23 further includes:
  • Pi (i, j ) Ki, j * Zi * (XijO + Xijl ) I (Xij0 + Xijl + Xij2).
  • step S3 further includes:
  • a cell whose sum of all neighboring cell IDs and its main carrier interference probability is not zero is added to the to-be-optimized list.
  • the frequency grouping in the step S4 satisfies the following conditions:
  • the frequency points of each frequency group are different.
  • step S4 further includes:
  • the secondary carrier coefficient has a value range of [0, 1].
  • the frequency group is allocated to the cell in the step S5, the following conditions are also met: the primary frequency points of the cells in the same base station are different;
  • step S7 further includes:
  • the sum of the total interference probabilities of all the cell main carriers is calculated, and if it is lower than the sum of the total interference probabilities before the optimization, the process ends; if not less than the sum of the total interferences before the optimization, then the selection is made.
  • the 10% cell with the highest total interference probability of the primary carrier is added to the to-be-optimized list, and steps S5-S7 are repeatedly performed for iteration.
  • the primary frequency point of the cell meets different primary frequencies of the same station cell, and the different frequency of the cell is played.
  • Embodiment 1 The frequency optimization method described in Embodiment 1 is applied to a TD-SCDMA network system.
  • the specific interference matrix algorithm is as follows: The server acquires a network measurement report of each cell, where the measurement report includes a serving cell PCCP (RSCP value and a measured neighbor cell PCCPCH RSCP value.
  • RSCP value serving cell PCCP
  • PCCPCH serving cell PCCP
  • the level distribution between the main service cell and the neighboring cell is counted from the massive test report, and the distribution of the cell and the interval MR is generated, and the capability of interference between any cells in the network is calculated according to the interference matrix.
  • the same neighbor frequency interference degree of each frequency point in the cell is performed and selected, and then the interference degree of all areas in the network is calculated.
  • the specific algorithm is as follows:
  • the interference level can be divided into three sections: high interference, general interference and low interference:
  • High interference neighboring area main service cell - neighboring area ⁇ a dB;
  • the high interference threshold ⁇ the default threshold is -3;
  • the general interference threshold ⁇ the default threshold is 10
  • the two parameters can be set, the effective range is [-12.12], the unit precision is ldB, and ⁇ -determination To be greater than ⁇ ;
  • the MR distribution table by threshold value is:
  • High-interference neighboring area using points/Neighboring total sampling points XijO/ (Xij0+Xijl+Xij2)
  • the neighboring area j traffic weight coefficient Ki of the cell i, j the number of sampling points of the neighboring area j in the cell i / the number of sampling points of the cell i.
  • the cell traffic weight that is, the total sampling point of the cell accounts for the proportion of the whole network, and establishes the network-level interference probability:
  • the probability of interference in the network is determined, and the interference probability is calculated as follows:
  • the frequency allocation scheme is based on frequency grouping, and the default grouping scheme is as follows:
  • the primary frequency of each packet cannot be the same
  • the main frequency points of each cell under the same base station are different;
  • the frequency of the cell is different
  • the probability of interference at the primary frequency point allocated by the cell is the smallest
  • the frequency groups available for the whole network are Fl, F2, F3, F4, and F5.
  • the interference of each neighboring cell to cell C is as follows: :
  • the primary carrier frequency of the neighboring cell uses the primary carrier interference probability Pi (i, j ), and the secondary carrier frequency uses the secondary carrier interference probability Pt (i, j ); as shown in the above table, the primary carrier of the cell 1 is F2, and the secondary carrier is F1 and F3.
  • the frequency of the secondary carrier is multiplied by the secondary carrier interference coefficient Y, and the default value is 0.5, and the value range is [0, 1];
  • the cell C frequency allocation is the primary carrier F5 and the secondary carrier is F3, F4, and the total interference of the cell C is the sum of all the interference probabilities of the primary frequency point F5. 4, antenna relative and distance limit algorithm
  • the distance and relative angle of the cell should also be considered.
  • the calculation method is shown in Figure 2. According to the distance and relative angle, the frequency optimization group has the following restrictions:
  • Generating an interference matrix generating an interference probability array of primary and secondary carriers between adjacent pairs according to an interference matrix algorithm; generating a list of cells to be optimized, and adding all i cells with a probability of not having a collision probability of 0 (Pi (i, j )) List
  • the frequency point total interference table generation method the sum of the total interferences of all the cell main carriers is calculated
  • step d is a frequency point total interference table generation method, and the difference from step d is that d only calculates the interference of the cell main carrier, and is used for evaluating the interference of the current network, and step f is to group the cells from all frequencies.
  • the frequency with the least interference is selected, so step f is to calculate the interference when all the packets are used as the primary frequency point; the frequency with the least interference degree is selected as the primary frequency of the cell, and the different primary frequencies of the same station are satisfied, and the different cells are different.
  • Frequency according to the antenna relative and distance limitation algorithm; if the allocation fails, the cell frequency point is not repaired, and is directly deleted from the cell list to be optimized;
  • the cell list to be optimized is empty, the sum of total interference of all cells is calculated again, and compared with the total interference of all cells before optimization, if it is lower than before optimization, it ends.
  • step e If the total interference is not reduced, continue to select the 10% cell with the highest total interference of the primary carrier after the cell optimization is added to the to-be-optimized list, and perform step e.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明提供一种用于通信系统的频率优化方法,其包括:根据所述网络测量报告数据确定网络中小区间干扰概率;根据小区间所述主载波干扰概率生成待优化小区列表;根据待优化小区列表、现网小区频点数据,计算所有小区在主载波频点下的干扰概率之和;计算频率分组中的各频点作为待优化小区的频点时的干扰概率之和,以从所述频率分组中选出干扰概率的频点;将所述干扰最小的频点作为小区主频点,为小区分配与所述主频点对应的频率分组以进行频率优化,所述频率优化按照主载波频点总干扰概率由大到小的顺序进行。该频率优化方法减少了频率优化需要的人力物力测试成本,而且优化结果更符合实际网络用户的移动性。

Description

一种用于通信系统的频率优化方法 技术领域
本发明涉及通信技术领域, 尤其涉及一种用于通信系统的频率优化方法。 背景技术
频率规划是降低无线干扰、 改善 GSM网络载干比、 提升通信质量和网络.— 能的重要手段; 频率优化工作是解决 GSM系统网络干扰、 优化网络最基本和最 重要的工作之一。
目前 GSM自动改频已成为现网频率优化的主要方式, 其通过软件平台实现 频率的自动分配, 大大节省了规划人员繁琐的工作, 提高了工作效率; 同时通过 对现网数据的全面采集分析, 也提高了频率规划的质量。
GSM 基于 MR频率优化干扰矩阵技术原理:
根据 GSM规范, 为了能对通话中的手机进行功控和切换控制, 网络必须得 到手机的相关信息, 这些信息是由手机汇报的。 对于 GSM网络中的手机来说, 在通话状态下,手机以 480毫秒的周期定期向网络汇报它所测量到的服务小区和 邻小区的测量报告, 每个测量报告主要包括服务小区的 BCCH、 信号电平、通话 质量和 TA值等,另外还包括信号最强的 6个邻小区的 BCCH、信号电平和 BSIC (网络色码) 等。 而 BSC则根据这些测量报告, 根据网络定义的功率控制和切 换的参数进行功率控制和切换控制等。 但手机并不同时测量 GSM频段上的所有 频点, 根据网络参数的设置, 手机所上报的邻区信息仅局限于 BCCH频点位于 该服务小区所定义的 BA表即邻区列表的小区。
由于手机上行发射的测量报告包括了网络内所有用户在所有时段通话时在 其所在位置的各个小区间信号强度情况,通过收集和分析这些测量报告, 我们就 能够得到网络内所有小区之间的信号干扰情况而形成干扰矩阵。通过这个干扰矩 阵,自动频率分配工具就能够根据不同小区间的干扰情况决定是否能够分配同频 或者邻频。 自动频率分配工具运用基因算法, 不断对分频方案进行评估、 再分频 的重复运算。 最终使网络中的干扰值降为最低。 GSM 基于 MR频率优化干扰矩阵技术存在如下缺点: 逐个频点分配只适 于频点较多的 GSM窄带系统, 对频点有限的 3G宽带系统, 效果不佳; 未考 网络拓扑情况, 对未上报 MR报告但地理位置相近的小区, 可能造成同频干 遗传 (基因) 算法效率不高, 反复迭代后也不一定是最优解。
随着 TD-SCDMA网络建设的开展和 3G用户日趋增多, 需要根据站点分 的不断增加和小区配置的扩容, 尤其是对于平均站间距小, 站点分布密集, ^ 业务热点区域, 小区载频配置高的区域, 往往网络的干扰会比较高, 同时由于日 常优化频繁地频率调整和工程建设导致频点扰码混乱出现干扰, 需要持续的进 行频率优化, 来降低控制信道和业务信道干扰, 提高网络质量和用户感知。 鉴于 此情况, 对 TD-SCDMA网络需要全网的周期性的进行频率优化, 以提升网络负 荷提高时的性能指标。但传统的频率优化需要大规模的扫频优化测试, 测试范围 需要遍历每个小区, 耗费大量人力物力, 且只能覆盖道路小区交叠情况, 所以频 率优化结果更偏向于道路覆盖情况, 对实际用户分布及移动性情况无法考虑。 发明内容 (—) 技术问题
本发明要解决的问题是: 传统的频率优化需要大规模的扫频优化测试, 测试 范围需要遍历每个小区, 耗费大量人力物力, 且只能覆盖道路小区交叠情况, 所 以频率优化结果更偏向于道路覆盖情况, 对实际用户分布及移动性情况无法考 虑。
(二) 技术方案
本发明提供一种用于通信系统的频率优化方法, 该方法包括:
51. 获取主服务小区以及其邻小区的测量报告数据;
52. 根据所述测量报告数据确定网络中小区间干扰概率, 所述干扰概率包括 小区间主载波干扰概率和小区间辅载波干扰概率;
S3. 根据小区间所述主载波干扰概率生成待优化小区列表;
S4. 计算频率分组中的各频点作为待优化小区的频点时的干扰概率之和, 以 从所述频率分组中选出干扰概率之和最小的频点; 55. 将所述干扰最小的频点作为待优化小区的主频点, 为小区分配与所述 频点对应的频率分组。
可选的, 所述步骤 S4、 S5按照主载波频点总干扰概率由大到小的顺序刘 优化列表中的各待优化小区进行。
可选的, 所述步骤 S4、 S5之间还包括有步骤:
所述载波频点总干扰概率为所有邻区 ID与其主载波干扰概率乘积的和。 可选的, 该方法还包括:
56. 如果步骤 S5分配失败,则直接从待优化小区列表删除所述小区主频点, 如果分配成功, 按照频率分组更新小区的主、 辅载波频率信息, 并从待优化列表 中删除已重新分配频率组的小区;
57. 判断待优化小区列表是否为空, 如果不为空, 重复执行步骤 S5-S7, 直 到待优化列表为空或者达到预定的迭代次数。
可选的, 小区间辅载波干扰概率 =XijO/ (Xij0+Xijl+Xij2),
小区间主载波干扰概率= (XijO+ Xijl ) I (Xij0+Xijl+Xij2),
其中 XijO表示高干扰邻区采样点数, Xijl表示一般干扰邻区采样点数, Xij2 表示低干扰邻区采样点数。
可选的, 根据小区与其邻区的电平差, 确定高干扰邻区、 一般干扰邻区和低 干扰邻区。
可选的, 所述步骤 S2进一步包括:
S23. 根据所述小区间干扰概率建立小区间干扰矩阵。
可选的, 所述步骤 S23进一步包括:
5231. 计算小区 i的邻区 j话务量权重系数 Ki,j= 小区 i的邻区 j的采样点数 I 小区 i总采样点数;
5232. 计算网络级干扰概率:
Zi= TOTALi/( TOTAL1+ + TOTALi+ +TOTALn), 其中 TOTALi为 小区 i下所有采样点总量;
5233. 根据所述系数 Ki,j 和网络级干扰概率计算网络中辅载波干扰概率 Pt (i, j ) 和网络中主载波干扰概率 Pi (i, j ):
Pt (i, j ) =Ki,j*Zi*XijO/ (Xij0+Xijl+Xij2)
Pi (i, j) =Ki,j*Zi* (XijO+Xijl ) I (Xij0+Xijl+Xij2)。 可选的, 所步骤 S3进一步包括:
将所有邻区 ID 与其主载波干扰概率乘积的和不为零的小区加入待优化 表。
可选的, 所述步骤 S4中所述频率分组满足如下条件:
每个频率分组内没有相同的频点;
每个频率分组的主频点不相同。
可选的, 所述步骤 S4进一步包括:
将小区的各邻区在所述频点的辅载波干扰概率乘上一辅载波系数,再与各邻 区在所述频点的主载波干扰概率进行相加以计算小区在各频点的干扰概率之和。
可选的, 所述辅载波系数的取值范围为 [0,1]。
可选的, 所述步骤 S5中为小区分配频率分组时还满足如下条件: 同一基站下的各小区主频点不同;
对打小区不同主频。
可选的, 所述步骤 S7进一步包括:
当待优化小区列表为空时,计算所有小区主载波的总干扰概率之和, 如果低 于优化前的总干扰概率之和, 则结束; 如果不低于优化前的总干扰之和, 则选择 小区优化后的主载波总干扰概率最高的 10%小区加入待优化列表,重复执行步骤 S5-S7进行迭代。
可选的, 小区主频点满足同站小区不同主频, 对打小区不同频。
(三) 技术效果
本发明通过基于测量值计算小区间干扰情况及生成小区干扰矩阵,根据干扰 概率进行小区频率优化,减少了频率优化需要的人力物力测试成本, 而且优化结 果更符合实际网络用户的移动性。 附图说明 图 1表示本发明实施例 1中的频率优化方法流程图;
图 2表示本发明中的小区间的距离和相对角度示意图;
图 3表示本发明实施例 2中的频率优化方法流程图。 具体实施方式 本发明提供一种用于通信系统的频率优化方法, 如图 1所示, 该方法包 ί si. 获取主服务小区以及其邻小区的测量报告数据,;
52. 根据所述测量报告数据确定网络中小区间干扰概率, 所述干扰概率包 小区间主载波干扰概率和小区间辅载波干扰概率;
53. 根据小区间所述主载波干扰概率生成待优化小区列表;
54. 计算频率分组中的各频点作为待优化小区的频点时的干扰概率之和, 以 从所述频率分组中选出干扰概率之和最小的频点;
55. 将所述干扰最小的频点作为待优化小区的主频点, 为小区分配与所述主 频点对应的频率分组。
可选的, 所述步骤 S4、 S5按照主载波频点总干扰概率由大到小的顺序对待 优化列表中的各待优化小区进行。
可选的, 所述步骤 S4、 S5之间还包括有步骤:
所述载波频点总干扰概率为所有邻区 ID与其主载波干扰概率乘积的和。 可选的, 该方法还包括:
56. 如果步骤 S5分配失败,则直接从待优化小区列表删除所述小区主频点, 如果分配成功, 按照频率分组更新小区的主、 辅载波频率信息, 并从待优化列表 中删除已重新分配频率组的小区;
57. 判断待优化小区列表是否为空, 如果不为空, 重复执行步骤 S5-S7, 直 到待优化列表为空或者达到预定的迭代次数。
可选的, 小区间辅载波干扰概率 =XijO/ (Xij0+Xijl+Xij2),
小区间主载波干扰概率= (XijO+ Xijl ) I (Xij0+Xijl+Xij2),
其中 XijO表示高干扰邻区采样点数, Xijl表示一般干扰邻区采样点数, Xij2 表示低干扰邻区采样点数。
可选的, 根据小区与其邻区的电平差, 确定高干扰邻区、一般干扰邻区和低 干扰邻区。
可选的, 所述步骤 S2进一步包括:
S23. 根据所述小区间干扰概率建立小区间干扰矩阵。
可选的, 所述步骤 S23进一步包括:
S231. 计算小区 i的邻区 j话务量权重系数 Ki,j= 小区 i的邻区 j的采样点数 I 小区 i总采样点数;
S232. 计算网络级干扰概率:
Zi= TOTALi/( TOTAL1+ + TOTALi+ +TOTALn), 其中 TOTALi 小区 i下所有采样点总量;
S233. 根据所述系数 Ki,j 和网络级干扰概率计算网络中辅载波干扰概率
(i, j ) 和网络中主载波干扰概率 Pi (i, j ):
Pt (i, j ) =Ki,j*Zi*XijO/ (Xij0+Xijl+Xij2)
Pi (i, j ) =Ki,j*Zi* (XijO+Xijl ) I (Xij0+Xijl+Xij2)。
可选的, 所步骤 S3进一步包括:
将所有邻区 ID 与其主载波干扰概率乘积的和不为零的小区加入待优化列 表。
可选的, 所述步骤 S4中所述频率分组满足如下条件:
每个频率分组内没有相同的频点;
每个频率分组的主频点不相同。
可选的, 所述步骤 S4进一步包括:
将小区的各邻区在所述频点的辅载波干扰概率乘上一辅载波系数,再与各邻 区在所述频点的主载波干扰概率进行相加以计算小区在各频点的干扰概率之和。
可选的, 所述辅载波系数的取值范围为 [0,1]。
可选的, 所述步骤 S5中为小区分配频率分组时还满足如下条件: 同一基站下的各小区主频点不同;
对打小区不同主频。
可选的, 所述步骤 S7进一步包括:
当待优化小区列表为空时, 计算所有小区主载波的总干扰概率之和, 如果低 于优化前的总干扰概率之和, 则结束; 如果不低于优化前的总干扰之和, 则选择 小区优化后的主载波总干扰概率最高的 10%小区加入待优化列表,重复执行步骤 S5-S7进行迭代。
可选的, 小区主频点满足同站小区不同主频, 对打小区不同频。
实施例 2
将实施例 1所述的频率优化方法应用于 TD-SCDMA网络系统。
一、 具体地干扰矩阵算法如下: 服务器获取各小区的网络测量报告, 所述测量报告包括服务小区 PCCP( RSCP值和测量到的邻小区 PCCPCH RSCP值。
从海量测试报告中统计出主服小区与邻区之间电平分布情况,生成小区与 区间 MR的分布情况,进行根据干扰矩阵计算出网络中任意小区间存在干扰的 能性。进行及选出小区中每一个频点的同邻频干扰度, 进而计算出网络中所有 区的干扰度情况。 具体算法如下:
1、 统计所有邻区对的 MR分布比例:
根据主服小区与邻区电平差, 干扰程度可化分为高干扰、一般干扰和低干扰 三个区间:
高干扰邻区: 主服小区 -邻区 < a dB;
一般干扰邻区: a dB 主服小区 -邻区 < β dB
低干扰邻区: 主服小区 -邻区 ^ P dB
其中,高干扰门限值 α,默认阈值为 -3;—般干扰门限值 β,默认阈值为 10, 两参数可设, 有效范围为 [-12.12],单位精度为 ldB, 且 β—定要大于 α ;
以默认值为例, 按门限值统计的 MR分布表为:
Figure imgf000009_0001
2、 建立小区间的干扰概率关系
小区间辅载波干扰概率: 高干扰邻区采用点数 /邻区总采样点数 =XijO/ (Xij0+Xijl+Xij2)
小区间主载波干扰概率= (高干扰 +—般干扰邻区采用点数) /邻区总采样点 数= (XijO+Xijl ) I (Xij0+Xijl+Xij2)
对小区干扰概率乘上相应的邻区话务量权重系数 Ki,j, 即某小区下邻区采样 点数量占比, 建立小区级的邻区干扰概率, 其中 小区 i的邻区 j话务量权重系数 Ki,j= 小区 i下邻区 j的采样点数 / 小区 i 采样点数。
3、 计算网络级小区干扰概率
通过统计小区话务量权重, 即该小区的总采样点占全网比重, 建立网络级 扰概率:
Zi= TOTALi/( TOTAL1+ + TOTALi+ +TOTALn)
其中 TOTALi为小区 S_Celli下所有采样点总量
4、 建立小区间干扰矩阵:
根据网络中的 MR的小区的情况, 当网络结构一定的情况下, 网络中干扰的 概率就会确定, 干扰概率计算如下:
辅载波干扰概率: Pt (i, j ) =Ki,j*Zi*XijO/ (Xij0+Xijl+Xij2)
主载波干扰概率: Pi (i, j ) =Ki,j*Zi* (XijO+Xijl ) I (Xij0+Xijl+Xij2) 二、 具体的频率优化算法如下:
1、 频率分组
频率分配方案基于频率分组, 默认分组方案如下:
Figure imgf000010_0001
分组四 FA7 FA4 FA5 FA8 FA6 FA 分组五 FA8 FA5 FA6 FA9 FA4 FA 分组六 FA9 FA6 FA4 FA7 FA5 FA 频率分组规则为:
每个分组内不可有相同的频点;
每个分组的主频点不能相同;
2、 频率优化原则
同一基站下的各小区主频点不同;
如果考虑小区相对性, 对打小区不同主频频;
小区分配到的主频点干扰概率最小;
3、 频点总干扰表生成方法
假设小区 C的主载波为 F2,其邻区 ID为 1到 9共 9个小区,全网可用的频 点组为 Fl、 F2、 F3、 F4、 F5, 各邻小区对小区 C的干扰如下表:
Figure imgf000011_0001
其中, 邻区的主载波频点使用主载波干扰概率 Pi (i, j ), 辅载波频点使用 辅载波干扰概率 Pt (i, j ); 如上表, 小区 1主载波为 F2, 辅载波为 F1和 F3。
在计算各个频点干扰度时, 作为辅载波的频点需乘上辅载波干扰系数 Y, 默 认值为 0.5, 取值范围为 [0,1];
计算小区 C在各个频点上的干扰概率总和,如 F1频点干扰概率和为 (Pt (c,l) +Pt (c,2) +Pt (c,7) )* y +Pi (c,4) +Pi (c,6) +Pi(c,9), 选取干扰概率最小的频点作 为小区 C的主频点, 并分配主频点对应的分组。
假设小区 C频率分配为主载波为 F5, 辅载波为 F3、 F4, 那小区 C的总干扰 就是主频点 F5所有干扰概率之和。 4、 天线相对及距离限制算法
在分配频率组时, 还要考虑小区的距离和相对角度, 计算方式如图 2所 根据距离和相对角度, 频率优化分组有以下限制:
两小区等效距离 D小于等于 500M, 则不可为同主频;
两小区相对角度5 = ( α - ( 180- β )) 在 30度以内, 则不可为同主频; 三、 具体的频率优化流程如下, 如图 3所示:
导入主服务小区邻小区测量报告数据;
生成干扰矩阵,按干扰矩阵算法,生成邻区对间的主、副载波干扰概率阵列; 生成待优化小区列表, 将所有∑j (Pi (i, j ))干扰概率不为 0的 i小区加入 列表;
根据现网小区频点数据,根据频点总干扰表生成方法, 计算所有小区主载波 的总干扰之和;
选择∑j (Pi (i, j )) 最大的小区;
生成小区频点干扰表, 按频点总干扰表生成方法, 与步骤 d的区别为, d只 计算小区主载波的干扰, 用于评估现网干扰情况, 步骤 f是要为小区从所有频率 分组中选出干扰最小的频点, 所以步骤 f要计算所有分组作为主频点时的干扰; 选择干扰度最小的频点作为小区主频点, 且满足同站小区不同主频, 对打小 区不同频, 按天线相对及距离限制算法; 如果分配失败, 则不修小区频点, 直接 从待优化小区列表中删除;
按照频率分组更新小区的主、辅载波频率信息, 下次干扰表计算将基于新的 小区频点数据;
从待优化小区表中剔除已重新分配频率组的小区;
判断待优化小区列表是否为空, 如果不为空重复执行步骤 e。
待优化小区表为空, 则再次计算所有小区总干扰之和, 并与优化前所有小区 总干扰对比, 如果低于优化前, 则结束。
如总干扰没有降低,则继续选择小区优化后的主载波总干扰最高的 10%小区 加入待优化列表, 执行步骤 e
如迭代次数到达 N, 则强制结束, 取 N次迭代中总干扰最小的结果, 作为 翻频结果。 以上实施方式仅用于说明本发明, 而并非对本发明的限制, 如本发明还可 于 GSM网络系统等其它网络系统, 有关技术领域的普通技术人员, 在不脱离 发明的精神和范围的情况下,还可以做出各种变化和变型, 因此所有等同的技 方案也属于本发明的范畴, 本发明的专利保护范围应由权利要求限定。

Claims

权 利 要 求
1、 一种用于通信系统的频率优化方法, 其特征在于, 该方法包括:
51. 获取主服务小区以及其邻小区的测量报告数据;
52. 根据所述测量报告数据确定网络中小区间干扰概率, 所述干扰概率包 小区间主载波干扰概率和小区间辅载波干扰概率;
53. 根据小区间所述主载波干扰概率生成待优化小区列表;
54. 计算频率分组中的各频点作为待优化小区的频点时的干扰概率之和, 以 从所述频率分组中选出干扰概率之和最小的频点;
55. 将所述干扰最小的频点作为待优化小区的主频点, 为小区分配与所述主 频点对应的频率分组。
2、 如权利要求 1所述的方法, 其特征在于: 所述步骤 S4、 S5按照主载波频 点总干扰概率由大到小的顺序对待优化列表中的各待优化小区进行。 3、 如权利要求 2所述的方法, 其特征还在于: 所述步骤 S4、 S5之间还包括 有步骤:
所述载波频点总干扰概率为所有邻区 ID与其主载波干扰概率乘积的和。
4、 如权利要求 1或 2所述的方法, 其特征在于, 该方法还包括:
S6. 如果步骤 S5分配失败,则直接从待优化小区列表删除所述小区主频点, 如果分配成功, 按照频率分组更新小区的主、 辅载波频率信息, 并从待优化列表 中删除已重新分配频率组的小区;
S7. 判断待优化小区列表是否为空, 如果不为空, 重复执行步骤 S5-S7, 直 到待优化列表为空或者达到预定的迭代次数。
5、 如权利要求 1所述的方法, 其特征在于:
小区间辅载波干扰概率 =XijO/ ( Xij0+Xijl+Xij2 ), 小区间主载波干扰概率= (XijO+ Xijl ) I (Xij0+Xijl+Xij2), 其中 XijO表示高干扰邻区采样点数, Xijl表示一般干扰邻区采样点数, X 表示低干扰邻区采样点数。 6、 如权利要求 5所述的方法, 其特征在于:
根据小区与其邻区的电平差,确定高干扰邻区、一般干扰邻区和低干扰邻 E
7、 如权利要求 1所述的方法, 其特征在于: 所述步骤 S2进一步包括: S23. 根据所述小区间干扰概率建立小区间干扰矩阵。
8、 如权利要求 7所述的方法, 其特征在于: 所述步骤 S23进一步包括:
5231. 计算小区 i的邻区 j话务量权重系数 Ki,j= 小区 i的邻区 j的采样点数 I 小区 i总采样点数;
5232. 计算网络级干扰概率:
Zi= TOTALi/( TOTAL1+ + TOTALi+ +TOTALn), 其中 TOTALi为 小区 i下所有采样点总量;
5233. 根据所述系数 Ki,j 和网络级干扰概率计算网络中辅载波干扰概率 Pt (i, j ) 和网络中主载波干扰概率 Pi (i, j ):
Pt (i, j ) =Ki,j*Zi*XijO/ (Xij0+Xijl+Xij2)
Pi (i, j ) =Ki,j*Zi* (XijO+Xijl ) I (Xij0+Xijl+Xij2)。
9、 如权利要求 1所述的方法, 其特征在于, 所步骤 S3进一步包括: 将所有邻区 ID 与其主载波干扰概率乘积的和不为零的小区加入待优化列 表。
10、 如权利要求 1所述的方法, 其特征在于, 所述步骤 S4中所述频率分组 满足如下条件: 每个频率分组内没有相同的频点;
每个频率分组的主频点不相同。
11、 如权利要求 1所述的方法, 其特征在于, 所述步骤 S4进一步包括: 将小区的各邻区在所述频点的辅载波干扰概率乘上一辅载波系数,再与各 区在所述频点的主载波干扰概率进行相加以计算小区在各频点的干扰概率之
12、 如权利要求 11所述的方法, 其特征在于, 所述辅载波系数的取值范围 为 [0,1]。
13、 如权利要求 1所述的方法, 其特征在于, 所述步骤 S5中为小区分配频 率分组时还满足如下条件:
同一基站下的各小区主频点不同;
对打小区不同主频。
14、 如权利要求 4所述的方法, 其特征在于, 所述步骤 S7进一步包括: 当待优化小区列表为空时,计算所有小区主载波的总干扰概率之和, 如果低 于优化前的总干扰概率之和, 则结束; 如果不低于优化前的总干扰之和, 则选择 小区优化后的主载波总干扰概率最高的 10%小区加入待优化列表,重复执行步骤 S5-S7进行迭代。
15、 如权利要求 1所述的方法, 其特征在于:
小区主频点满足同站小区不同主频, 对打小区不同频。
PCT/CN2013/080093 2012-12-14 2013-07-25 一种用于通信系统的频率优化方法 WO2014089974A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201210545557.4 2012-12-14
CN201210545557.4A CN102970696B (zh) 2012-12-14 2012-12-14 一种用于通信系统的频率优化方法

Publications (1)

Publication Number Publication Date
WO2014089974A1 true WO2014089974A1 (zh) 2014-06-19

Family

ID=47800494

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/080093 WO2014089974A1 (zh) 2012-12-14 2013-07-25 一种用于通信系统的频率优化方法

Country Status (2)

Country Link
CN (1) CN102970696B (zh)
WO (1) WO2014089974A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112351435A (zh) * 2019-08-08 2021-02-09 中国电信股份有限公司 基于网络定位的异网覆盖测评方法和装置

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102970696B (zh) * 2012-12-14 2016-01-27 上海大唐移动通信设备有限公司 一种用于通信系统的频率优化方法
CN104105115B (zh) * 2013-04-15 2018-02-23 中国移动通信集团浙江有限公司 一种频率优化方法及系统
CN103747458A (zh) * 2014-02-13 2014-04-23 方海鹰 移动通信频率修正优化方法及系统
CN104918269B (zh) * 2014-03-10 2019-10-11 中国移动通信集团广东有限公司 一种构建干扰矩阵的方法和装置
CN103916229B (zh) * 2014-04-29 2017-12-08 电信科学技术研究院 一种干扰抑制方法及装置
CN105451247B (zh) * 2014-08-29 2018-11-16 中国移动通信集团浙江有限公司 一种评估网络结构的方法和装置
CN105992369B (zh) * 2015-02-06 2019-05-21 电信科学技术研究院 一种频率优化方法及装置
CN105120478B (zh) * 2015-07-21 2019-04-16 杭州领优信息科技有限公司 一种频率优化方法及系统
CN106686668B (zh) * 2015-11-10 2019-12-13 中国移动通信集团江苏有限公司 一种语音回落小区的选择方法及装置
CN106804040A (zh) * 2015-11-26 2017-06-06 中兴通讯股份有限公司 一种动态分配频点的方法和装置
CN106649941B (zh) * 2016-09-28 2019-11-22 北京理工大学 一种用于获得视线失调角的抽样改进方法
CN106817703B (zh) * 2016-12-16 2019-11-19 京信通信系统(中国)有限公司 一种频点优化方法及装置
CN111543075B (zh) * 2018-06-26 2021-11-30 华为技术有限公司 一种频段选择方法及装置
CN109495913B (zh) * 2018-12-29 2022-02-15 中国联合网络通信集团有限公司 干扰评估方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101577918A (zh) * 2009-05-04 2009-11-11 中国移动通信集团浙江有限公司 一种基于手机测量报告的干扰矩阵建立方法
CN102149095A (zh) * 2010-02-08 2011-08-10 中国移动通信集团福建有限公司 一种移动通信系统的频点设置方法和装置
WO2012167498A1 (zh) * 2011-08-03 2012-12-13 华为技术有限公司 基于模拟退火算法的频率优化方法和装置
CN102970696A (zh) * 2012-12-14 2013-03-13 上海大唐移动通信设备有限公司 一种用于通信系统的频率优化方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101577918A (zh) * 2009-05-04 2009-11-11 中国移动通信集团浙江有限公司 一种基于手机测量报告的干扰矩阵建立方法
CN102149095A (zh) * 2010-02-08 2011-08-10 中国移动通信集团福建有限公司 一种移动通信系统的频点设置方法和装置
WO2012167498A1 (zh) * 2011-08-03 2012-12-13 华为技术有限公司 基于模拟退火算法的频率优化方法和装置
CN102970696A (zh) * 2012-12-14 2013-03-13 上海大唐移动通信设备有限公司 一种用于通信系统的频率优化方法

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112351435A (zh) * 2019-08-08 2021-02-09 中国电信股份有限公司 基于网络定位的异网覆盖测评方法和装置
CN112351435B (zh) * 2019-08-08 2022-09-23 中国电信股份有限公司 基于网络定位的异网覆盖测评方法和装置

Also Published As

Publication number Publication date
CN102970696A (zh) 2013-03-13
CN102970696B (zh) 2016-01-27

Similar Documents

Publication Publication Date Title
WO2014089974A1 (zh) 一种用于通信系统的频率优化方法
US20170238201A1 (en) Downlink interference detection using transmission matrices
CN103210593B (zh) 小区间干扰协调自组织网络的方法和装置
CN110505650B (zh) 随机异构分层网容量智能评估方法及装置
WO2016090770A1 (zh) 一种频点测量消息的发送方法及装置
So et al. Enhancing WLAN capacity by strategic placement of tetherless relay points
CN103339885A (zh) 用于无线网络的利用成本函数的分布式信道选择方法
CN104349480A (zh) 无线网络信道分配方法、装置及系统
CN101695173A (zh) 一种越区评估方法及装置
CN105578443A (zh) 邻区维护方法及装置
CN103974264A (zh) 一种频点优选方法
WO2016058448A1 (zh) 频带的分配方法及装置
CN103002495A (zh) 无线网络结构的评估方法及装置
CN108271181B (zh) 一种无线网络的pci优化方法及装置
CN101635988A (zh) 资源分配方法及其设备
KR102444229B1 (ko) 빅데이터 기반 대규모 무선랜 관리 시스템 및 방법
CN105636059B (zh) 一种分配频谱的方法和设备
CN109391946B (zh) 一种基站簇规划的方法及装置
CN113825155A (zh) 5g网络的互操作参数配置方法及装置
Mkheimer et al. Evaluation and optimisation of GSM network in Jenien City, Palestine
WO2016107262A1 (zh) 小小区频带资源的分配方法及装置
WO2016065759A1 (zh) 邻区列表优化方法及装置
CN101917724B (zh) 获取广播控制信道的联合干扰矩阵的方法及其系统
WO2017211178A1 (zh) 划分超级小区的方法及装置
CN108243434B (zh) 掉话位置的定位方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13863232

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13863232

Country of ref document: EP

Kind code of ref document: A1