WO2009030112A1

WO2009030112A1 - Method of load balancing of the multi-hop wireless network based on the relays

Info

Publication number: WO2009030112A1
Application number: PCT/CN2008/001545
Authority: WO
Inventors: Xin Guo; Wenchao Ma
Original assignee: Lenovo (Beijing) Limited
Priority date: 2007-08-30
Filing date: 2008-08-28
Publication date: 2009-03-12
Also published as: CN101378578B; CN101378578A

Abstract

A method of load balancing of the multi-hop wireless network based on the relays, including the steps: determining the regulation groups and the regulation relay stations; dividing the load of the regulation relay stations; leaving part of the divided load in the initial group in which the regulation relay stations staying and making part of the divided load entering into the other groups. The method of load balancing in the present invention can be used to the complicated multi-hop wireless network based on the relays, increasing the system resource utility efficiency and dynamically regulating the load balancing strategy as the network load changes.

Description

Load balancing method for relay-based multi-hop wireless network

Technical field

The present invention relates to a load balancing method for a wireless network, and more particularly to a load balancing method based on a relay multi-hop broadband wireless network. Background technique

In recent years, broadband wireless communication technologies and networks have developed rapidly. Due to spectrum planning reasons, the carrier frequencies of emerging wireless communication systems are relatively high. Although the higher frequency band can solve the frequency allocation problem, the higher the frequency band, the worse the diffraction ability of the electromagnetic wave and the worse the penetration ability. This situation poses a challenge to the coverage of the wireless signal. The current wireless access network architecture is mainly a single-hop network composed of a base station (BS, Base Sta ti on) and a mobile subscriber (MSS, Mob i le Subscr i ber S ta t ion), each Each MSS accesses the network through a wireless link connected to the base station. Such a single-hop network is difficult to meet the requirements of a high-frequency wireless communication system, because high-frequency signals are easily affected by the shadow fading effect caused by terrain due to poor transmission and diffraction capabilities. In addition, the frequency band is more High, the faster the electromagnetic wave attenuation under the same transmission power condition, the smaller the coverage area of the system base station is. Therefore, the wireless communication network proposes the concept of a multi-hop wireless network. The multi-hop wireless network uses a relay station (RS, Re 1 ay ta t ion) to forward signals between the BS and the MSS, thereby forming a path of the BS-RS-MSS. Wherein, the RS may be one level or multiple stages, and a node of a multi-hop wireless network can reach the adjacent node by using a smaller transmission power, so the same power level is used compared with the single-hop network. The coverage area of the network following the network has greatly increased. Moreover, the RS location planning is flexible, which can effectively avoid the influence of terrain on the network layout.

However, the data in the relay-based multi-hop wireless network needs to be relayed one or more times between the BS and the MSS, and the signal transmission time overhead or frequency occupation angle will waste the system resources, thereby Reduce system capacity. To this end, some new technologies attempt to use resources or other effective methods to reuse resources. The resources referred to here may be the time in a time division multiple access communication system or the frequency in a frequency division multiple access communication system. Bandwidth, which can also be a pseudo-random code in a code division multiple access communication system, or a combination of these resources,

! From the perspective of information theory, the allocation methods of different types of resources are similar. Taking a time division multiple access communication system as an example, an effective resource multiplexing method is shown in FIG. 1 , which is a schematic diagram of a relay network downlink subframe, wherein a data transmission link between the base station and each relay station Called as a relay link, denoted as BS→RS1, BS→RS2, BS→RS3, and BS RS4; the data transmission link between the base station, the relay station, and the user is called an access link, and the base station and all directly communicating with it The BS→MSS link between users is treated as a whole and is scheduled in units of BS. Similarly, the RS→MSS link between any relay station and all users directly communicating with it is also regarded as a whole. The scheduling is performed in units of RSs. Therefore, the time when the BS is serving the MSS in the figure is marked as BS on the time frame, and the time when each RS is serving its MSS is marked as the RS on the time frame, for example, RS1, RS2. RS3, RS4, etc. The resource multiplexing scheduling method does not limit the order of the relay link and the access link in the downlink subframe, and the uplink subframe can be obtained in the same manner. Part (a) of Figure 1 shows the frame structure without resource multiplexing, and part (b) of Figure 1 shows the frame structure after resource multiplexing scheduling. The core idea is that there will be no interference. The RS is divided into the same packet, and all the RSs in the packet are simultaneously scheduled. The RSs in the same packet are shown to have the same time start point for the MSS service in the coverage area in the time frame; the RSs that interfere with each other are different. Packets, different packets need to be scheduled in time, and displayed on the time frame as the scheduling time of each packet is staggered.

Each RS in Figure 1 occupies a different length in the time frame due to the different amount of service load within its coverage. For example, in a certain period of time, the MSS requires a large amount of resources in the coverage of an RS, that is, the RS has a large load, and the time of the RS for the MSS service in the coverage is on the time frame. The length occupied is longer, and vice versa.

As can be seen from FIG. 1, in the resource multiplexing method, if the difference in RS load in the same packet is larger, that is, the difference in length occupied by each RS service time on the time frame is large, the resource utilization ratio is lower. On the other hand, if the RSs in the same packet are balanced, the resource utilization of the method is high, and thus the system capacity can be effectively increased. The load of each RS cannot be balanced due to the difference between the MSS and its services. Therefore, the current resource reuse scheduling method based on the relay multi-hop wireless network is difficult to achieve better results. Summary of the invention

In view of this, an object of the present invention is to provide a load balancing method for a multi-hop wireless network based on a relay, which can further subdivide and reorganize the load of each RS according to the load status of each RS, thereby realizing each The balance of RS in the group.

In order to achieve the above object, the technical solution adopted by the present invention is a load balancing method for a relay-based multi-hop wireless network, the method comprising the following steps: A. determining an adjustment packet and adjusting a relay station; B, adjusting the relay station The load is divided; (:, after the division, the part is left in the original group where the adjustment relay station is located, and part of the reorganization is entered into other groups.

In step A, the adjustment packet may be an original packet whose load imbalance is greater than a predetermined balance threshold. The adjustment relay station can be a re-station relay station.

In step B, the division may be: controlling the power of the relay station, and dividing the load of the relay station into a load formed by a mobile user close to the relay station and a load formed by a mobile user remote from the relay station.

In step C, other packets may be new packets or other original packets.

Step C may further comprise: continuing to perform step A to step on the new packet

C.

The packet load imbalance can be measured by the quotient of the standard variance and the mean of the load of each relay within the packet.

The heavy-duty relay station may be a relay station that bears an average load predetermined threshold value of a relay station that exceeds the packet in which the relay station is located.

The method may further include repeating the step A to the step (for a period of a specified period of time).

The load balancing method for the relay-based multi-hop wireless network provided by the present invention divides and reorganizes the load of each RS according to the load condition and the grouping situation of each RS, and has the following advantages: The load balancing method does not It is necessary to know the network topology or interference in advance, and to determine the load balancing result by using the BS and RS load and interference conditions dynamically measured during network operation. The load balancing method can be applied to complex relay-based multi-hop. The wireless network system can improve the resource utilization of the relay-based multi-hop wireless network and increase the system capacity. DRAWINGS

1 is a schematic structural diagram of a downlink subframe before and after using a relay-based multi-hop wireless network packet resource multiplexing scheduling method in the prior art;

2 is a flow chart of a relay-based multi-hop wireless network load balancing method according to the present invention; FIG. 3 is a topological structural diagram of an embodiment of a relay-based multi-hop wireless network load balancing method according to the present invention; ;

Figure 3 (b) is a downlink subframe structure diagram of the radio network packet resource multiplexing scheduling shown in Figure 3 (a);

4 is a schematic diagram of a division result of a relay-based multi-hop wireless network load balancing method for adjusting a relay station in the present invention;

FIG. 5( a ) is a schematic diagram of a method for dividing a load of an adjusted relay station by power adjustment in a relay-based multi-hop wireless network load balancing method according to the present invention;

Figure 5 (b) is a schematic diagram showing the result of the division of the load after the method shown in Figure 5 (a); Figure 6 is a structural diagram of the downlink subframe after re-grouping in the first embodiment of the present invention; The downlink subframe structure diagram after re-grouping in the second embodiment. detailed description

The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

The core idea of the present invention is: dividing the load of a heavy-duty relay station in a packet with a large load imbalance, and the divided load is partially left in the original packet where the relay station is adjusted, and partially reorganized into other packets. Eliminate the imbalance within the group.

As shown in FIG. 2, the relay-based multi-hop wireless network load balancing method of the present invention includes the following steps:

Step 101: Determine the adjustment group and adjust the RS.

The manner in which packet resources are multiplexed for each RS within a coverage of a BS is as described in the background. As shown in FIG. 3( a ), in an embodiment of the present invention, the BS includes six RSs, which are respectively identified as RSI-RS6. In the figure, the lengths of the RSI-RS6 time segments on the time frame respectively represent that each RS is The length of time that the covered MSS provides services, and also reflects the load size of each RS. According to the bandwidth requirement and the interference relationship, the BS and the RSI-RS6 are divided into three groups for resource multiplexing scheduling, and the three groups are: {BS}, {RSK RS3, RS5} and {RS2, RS4, RS6}, the downlink subframe after packet multiplexing scheduling is shown in Figure 3 (b). In order to distinguish from the subsequent steps, these three groups are called the original group.

After each original packet and the RS included therein are determined, the adjustment packet is determined in each original packet, and the adjustment packet may be determined to optionally use the original packet as an adjustment packet, or all the original packets may be selected as an adjustment packet, or only the specific requirements may be selected. The original packet is used as an adjustment packet. As a preferred embodiment of the present invention, the original packet whose load imbalance is greater than a predetermined threshold is selected as the adjustment packet. First, the respective load conditions are obtained by the original intra-packet RS, and the load is the resource requirement in each RS coverage, which is determined by the quality of service requirements of each connection of the MSS in the RS coverage. Then, the load unbalance degree in each original group is obtained, and the load unbalance degree can be measured in various ways, for example, the load difference between the maximum load RS and the minimum load RS, the standard variance of each RS load in the original group, and the like. In this embodiment, the quotient of the standard deviation and the mean of each RS load in the original packet is selected to identify the load imbalance in each original packet. For example, in this embodiment, the load imbalances of the original packets (RS1, RS3, RS5} and {RS2, RS4, RS6} are both greater than a predetermined balance threshold Ba lanceDegree _lh , and the two original packets are selected as the adjustment packets.

After determining the adjustment packet, the adjustment RS needs to be further determined therein. There may be multiple ways to adjust the RS selection, for example, selecting the RS with the largest load in the adjustment packet as the adjustment RS, or selecting the RS whose load is greater than the load average of the adjustment packet. Adjust RS and so on. In this embodiment, the reload RS is selected as the adjustment RS. The overloaded RS means: setting the reload threshold to &> 0, and setting BW _{ase to} represent the average value of the RSs in an adjustment group, if the load of an RS in the adjustment packet is BW > ( 1+ a) BW _eve , then mark the RS as a reloaded RS. For example, in the present embodiment, the RSI in the packet {RS1, RS3, RS5} is selected as the adjustment RS, and the RS4 in the packet {RS2, RS4, RS6} is used as the adjustment RS.

Step 102: Divide the load of the adjustment RS.

The load of the RS is further divided to achieve a more efficient balancing of the load. There are various strategies for dividing, such as dividing according to different carrier frequencies, and dividing the load under the same RS according to different channel qualities.

In an embodiment of the present invention, as shown in FIG. 4, according to the result of step 101, the loads of RS1 and RS4 are further divided into corresponding time frames of RS1-1, RS1-2, and RS4-. 1 and the load part of RS4-2. There are many strategies for dividing, for example, the corresponding plants of RS1 and RS4-2 correspond to the negative subtraction of RS1 and RS4 respectively. Adjust the average load of the packet, or the load corresponding to RS1-1 and RS1-2 and RS4-1 and RS4-1, respectively, is 1 /2 of the load of RS1 and RS4.

In the second embodiment of the present invention, the distance from the geographic location of the relay station is selected as the basis for the load division. At this time, the power of the RS should be controlled to divide the load of the RS into the load and the distance formed by the MSS close to the RS. Negative planting formed by RS's MSS. As shown in FIG. 5( a ), the power of the RS1 and the RS4 is adjusted to reduce the power of RS1 and RS4 in a time period, and the time period corresponding to the occupied time frame shown in FIG. 5( b ) is RSI- The load of C and RS4 - C provides services to recover the power of RS1 and RS4 in another time period, in order to remove the load other than RSI-C and RS4-C, corresponding to the time frame shown in Figure 5 (b) Serve the load of RSI-W and RS4-W.

Step 103: After the division, the part is left in the original group where the adjustment relay station is located, and the part is reorganized into other groups.

For the load divided in step 102, in order to maintain the consistency of the original packet, the original packet mode is not changed. Therefore, the partially divided load remains in the original packet where the relay station is located, and the original packet is received by the other RSs. The partially divided load is reorganized into other packets.

As shown in Figure 6, the divided RS1-1 and RS4-1 remain in the original packet, and serve with RS3, RS5, and RS2, RS6, respectively, and the divided RS1-2 and RS4-1 form a new one. Packets, if RS 1 and RS 4 are mutually interfered with each other when communicating with their respective MSSs, they cannot be assigned to the same new packet, but should form a new packet with the divided bearers that do not have an interference relationship.

As shown in FIG. 7, corresponding to the second embodiment of the present invention, when RS1 provides a service for a load corresponding to a time period of RS1-C, since RS1 is actually serving a MSS that is close to it, it will not Serious interference is caused to the services of other RSs in the vicinity, so the time for serving the load of the corresponding time period for RSI-C can be classified into the packets {RS2, RS4, RS6), and the service is provided together with the packet. When RS4 can reduce power, it first provides services for RS4-W corresponding time segments. Conversely, when RS4 provides services for the load with the corresponding time period of RS4-C, it can also be classified into the packet (RS1, RS3 RS5}, and provide services together with the packet. At this time, RS1 can reduce power, which is the corresponding time period. Serve the load of RSI-W. Of course, RS I-W and RS4-W can also be selected to be classified into new packets, and are not scheduled simultaneously with the original packet.

After the end of step 1 01-1 03, if a new packet is generated, the load of the new packet needs to be further balanced, and steps 1 01-103 are repeated for the new packet.

In addition, since the network load condition changes at any time, in order to enable the dynamic adjustment process, the relay-based multi-hop wireless network load balancing method of the present invention may repeatedly perform the above steps 101 to 103 in a period of a specified period of time. And the subsequent balancing steps for the new grouping.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention. Those skilled in the art should be able to associate different wireless communication networks as the application background of the present invention, select different original packets and relay stations as adjustment packets and adjust relay stations, and use other parameters to represent the resource imbalance in the original packets. Different ways of dividing and reorganizing the load of the relay station should belong to the protection scope of the present invention.

Claims

Claim

A load balancing method for a relay-based multi-hop wireless network, characterized in that the method comprises the following steps:

A. Determine the adjustment group and adjust the relay station;

B. Divide the load of the adjustment relay station;

C. After the division, the part is left in the original group where the adjustment relay station is located, and part of the reorganization is entered into other groups.

2. The load balancing method of the relay-based multi-hop wireless network according to claim 1, wherein in step A, the adjustment packet is an original packet whose load imbalance is greater than a predetermined balance threshold.

3. The load balancing method of a relay-based multi-hop wireless network according to claim 1, wherein in step A, the adjustment relay station is a heavy-duty relay station.

The load balancing method of the relay-based multi-hop wireless network according to claim 1, wherein in the step B, the dividing is: controlling the power of the relay station, and dividing the load of the relay station into the proximity of the relay station. The load formed by the mobile user and the mobile user away from the relay station.

The load balancing method of the relay-based multi-hop wireless network according to claim 1, wherein in step C, the other packets are new packets or other original packets.

The method for balancing a load-based multi-hop wireless network according to claim 5, wherein the step C further comprises: performing step A to step (:) for the new packet.

7. The method according to claim 2, wherein the packet imbalance is measured by a quotient of a standard variance and a mean of load of each relay station in the packet. .

8. The load balancing method of a relay-based multi-hop wireless network according to claim 3, wherein the heavy-duty relay station is a relay station whose load exceeds an average load predetermined threshold of a relay station of a packet in which the relay station is located.

9. The load balancer of a relay-based multi-hop wireless network according to the claims The method is characterized in that the method further comprises: repeating the step A to the step (with a period of a specified period of time).