WO2013059990A1

WO2013059990A1 - Relay path selection method and system

Info

Publication number: WO2013059990A1
Application number: PCT/CN2011/081271
Authority: WO
Inventors: 邬钢
Original assignee: 中兴通讯股份有限公司
Priority date: 2011-10-25
Filing date: 2011-10-25
Publication date: 2013-05-02

Abstract

Disclosed are a relay path selection method and system, so as to solve a technical problem that in a scenario of hierarchical mesh structure having a universal meaning in a wireless communication system based on band relay, an existing optimal relay path selection method has an overlarge feedback signaling overhead and cannot realize the optimization of the entire path. The present invention, based on a maximum throughput or shortest path rule, selects an optimal path, so as to ensure that an optimal entire path selection result is achieved. A relay station (RS) participates in path selection comparison and saves a later stage CID, so as to reduce the signaling feedback overhead. A DeNB only receives the later stage CID and reserved link quality information group/path total hop-number information, and only needs to broadcast CID information of the first hop in the selected path, thereby further reducing the signaling overhead.

Description

Relay path selection method and system

The present invention relates to a wireless communication relay technology, and in particular to a method and system for selecting a relay path in an Enhanced Long Term Evolution (LTE-A) system. Background technique

Relay is an effective method to improve system throughput and expand coverage in wireless communication systems. The introduction of relay technology into cellular communication systems is considered in the Enhanced Long Term Evolution (LTE-A) system. Among them, a relay station (RS) may be a micro base station or an enhanced user equipment to provide a connection between the base station/relay and other relay/user equipment. When a user equipment (UE) accesses the cellular communication system, it can directly access the primary base station (Donor Evolved NodeB, DeNB) or indirectly access the primary base station through one or several relays. Choosing the optimal path to access the primary base station is an indispensable feature of the LTE-A system with trunking.

In order to better describe the path selection problem, the typical multi-hop relay hierarchical network model in Figure 1 is taken as an example to illustrate:

(1) A total of M hops from the DeNB to the UE;

(2) Available RSs are divided into (M - 1) groups;

(3) Total in the mth RS group! ^! ^, where m = l,..., (M - l) ;

(4) There is no link between the RSs in the group, and there is a link between the RSs in the group. That is, each RS can only connect to different levels of RS, and cannot connect to the same level RS. (Note: One RS group and multi-level RS group can have Contact, but the DeNB is only associated with some RS groups. Figure 1 is only for the sake of simplicity. Only in the multi-hop relay, the common practice is to link the quality information of each hop (for example: channel quality information). (Channel Quality Information, CQI); For LTE-A, Carrier Interference Noise Ratio (CINR) or Received Signal Strength Indicator can also be used. (Received Signal Strength Indicator, RSSI)), and the corresponding Connection Identifier (CID) are fed back to the DeNB, and the DeNB selects an optimal path according to certain criteria. IEEE C802.16j-07/027rl End-to-End Throughput Metrics for QoS Management 802.16j The maximum throughput criteria proposed in MR Systems optimizes the throughput of selected paths, namely:

M 1

T = max

R: „ where R _i>m is the data rate of the i-th path, the m-th hop, M is the maximum hop count from the DeNB to the UE, C is the set of all paths, and T represents the throughput in all path sets. The throughput of the largest one path. There is a corresponding relationship between the CQI and the R _im for the specified communication system. There is also a shortest path criterion: the sum of the number of paths traversed by the UE to the DeNB Path_Num (also called the total path hop) Number) The least.

In an actual system, if the number of RSs is large, the CQI and CID of each hop are fed back to the DeNB, which causes a large overhead of feedback signaling. The IEEE WCNC 2007 conference record proposes two methods of Ad-hoc routing and N-hop routing to reduce these feedback signaling. However, the Ad-hoc routing path in the text is selected in each. Independently in the RS group, N-hop routing only considers the combination of N hops, but N is less than the total hop count M, so the entire path cannot be optimal. US Patent Publication No. US2009285112A1, entitled Patent Management for Managing Delay Paths in a Wireless Communication Environment Considers Tree Trunking Relay Path Selection, Also Using Conventional Practices, the DeNB selects paths based on the simplest shortest path criteria, Consider the hierarchical network structure of the more general meaning shown in Figure 1, and do not consider optimizing system throughput. The Chinese publication number CN101827358A, the patent application entitled LTE-A system and its routing method, relay node and its donor base station, provides a relay routing method for the LTE-A system, focusing on different relays and the main base station. The handover process and related signaling do not consider how to find the optimal path and related signaling procedures. Summary of the invention

In view of the above, an object of the present invention is to provide a method and system for enhancing a relay path in a long term evolution system, which is used to solve a hierarchical network structure in a wireless communication system based on a relay. The feedback signaling overhead of the existing optimal relay path selection method is too large, and the technical problem of the entire path optimization cannot be achieved.

According to an aspect of the embodiments of the present invention, the present invention provides a relay path selection method based on a maximum throughput criterion, the method comprising:

A1: each RS receives a connection identification number CID fed back by each RS in the RS group of the subsequent level and a link quality information group corresponding to the CID;

The link quality information group here is a set of values composed of a plurality of paths, and may be a channel quality information CQI group (a set of values composed of a plurality of paths CQI), or a carrier interference noise ratio CINR group, or a received signal strength. Indicate the RSSI group;

The CID described in the embodiment of the present invention refers to the number of the RS path between the group and other groups.

A2: According to the maximum throughput criterion, each RS calculates each piece according to the link quality information group fed back by the RS in the RS group of the latter stage and the link quality information between the RSs currently measured and the RSs in the previous level RS group. Path throughput corresponding to the path, and selecting the path with the largest throughput as the persistence path;

A3: each RS forwards the corresponding CID and the remaining link quality information group corresponding to the newly selected remaining path (the set of link quality information values that are forwardly fed back after selection), and is in the present The CID of the corresponding subsequent level RS in the retention path is saved in the level RS;

Each step RS performs the above steps A1 to A3 in sequence, until the latest level RS group directly connected to the primary base station DeNB feeds back the CID and the surviving link quality information group to the primary base station DeNB;

A4: The DeNB calculates the path throughput corresponding to each path based on the remaining link quality information group fed back by the RS group, selects a path corresponding to the maximum throughput, and broadcasts the selection result to the subsequent RS group. , the selected RS is further based on the previously saved remaining path. The CID of the primary RS is notified to the subsequent primary RS group until the UE is notified;

Further, the method is applicable to a multi-hop hierarchical network, in which each RS in each level of the RS group has a path connection only with the RS group of the preceding and succeeding stages, or each RS in each level of the RS group There is a path connection between the preceding and following RS groups and/or two or more multi-level RS groups. According to another aspect of the embodiments of the present invention, the present invention provides a relay path selection method based on a shortest path criterion, the method comprising:

B1: the CID of each RS feedback in the RS group of each subsequent RS group and the total number of hops corresponding to the CID Path_Num;

The total number of hops of the path Path_Num refers to the total number of hops of the path that the UE reaches the RS corresponding to the CID;

B2: According to the shortest path criterion, each RS selects the smallest path among the Path_Nums of the RS feedback in the RS group of the latter stage as the persistence path;

B3: Each of the RSs forwards the corresponding CID to the previous level RS group and the newly selected remaining Path_Num corresponding to the remaining path (ie, the minimum Path_Num corresponding to the persisted path of the forward feedback), and at the current level RS Saving the CID of the corresponding subsequent level RS in the retention path;

Each of the first-level RSs sequentially performs the above steps B1 to B3 until the latest level RS group directly connected to the primary base station DeNB feeds back the CID and the persistence Path_Num to the DeNB;

B4: The DeNB selects one path corresponding to the minimum value based on the persistence Path_Num of the feedback of the latter stage, and notifies the selection result of the subsequent RS group, and the selected RS further corresponds to the subsequent level RS according to the previously saved remaining path. The CID notifies the next-level RS group until the UE is notified; further, the method is applicable to a multi-hop hierarchical network in which each RS and the pre- and post-level RS groups and/or each level of the RS group There are path connections between more than two multi-level RS groups. The present invention also provides a relay path selection method, which is applied to a multi-hop hierarchical network, and the method includes: Optimizing the relay path selection strategy based on the maximum throughput criterion or the optimal relay path selection strategy based on the shortest path criterion by configuring the module configuration;

When the configuration uses the optimal relay path selection strategy based on the maximum throughput criterion, the foregoing relay path selection method based on the maximum throughput criterion is adopted;

When the configuration uses the optimal relay path selection strategy based on the shortest path criterion, the aforementioned relay path selection method based on the shortest path criterion is employed. Accordingly, the present invention also provides a relay path selection system including a primary base station

a network element such as a DeNB, a relay station RS, a user equipment UE, and the DeNB, the RS, and the UE constitute a multi-hop relay hierarchical network;

The system also includes a configuration module for configuring the system to use an optimal relay path selection strategy based on a maximum throughput criterion or an optimal relay path selection strategy based on a shortest path criterion;

When the system is configured to use an optimal relay path selection policy based on a maximum throughput criterion, each network element function in the system is implemented by each network element in the foregoing relay path selection method based on the maximum throughput criterion. ;

When the system is configured to use the optimal relay path selection policy based on the shortest path criterion, each network element function in the system is a function implemented by each network element in the foregoing relay path selection method based on the shortest path criterion. The path selection method based on the feedback persistence CQI group/Path_Num provided by the present invention has the following beneficial effects: The entire path selection result can be ensured to be optimal; the RS participation path selection comparison saves the subsequent level CID, and reduces the signaling feedback overhead; The DeNB only receives the CID and the CQI group/Path_Num information of the subsequent stage, and only needs to broadcast the first hop CID information in the selected path, which further reduces the signaling overhead. DRAWINGS

Figure 1 is a typical multi-hop relay hierarchical network model diagram;

2 is a flowchart of a method for selecting a relay path based on a maximum throughput criterion according to the present invention; FIG. 3 is a flowchart of a method for selecting a relay path based on a shortest path criterion according to the present invention; FIG. Multi-hop relay hierarchical network model (no contact between different RS groups, 2RS/hop, 3 hops total);

FIG. 5 is a multi-hop relay hierarchical network model used in Embodiment 2 (no connection between different RS groups, 3RS/hop, 4 hops in total);

Figure 6 is a multi-hop relay hierarchical network model used in Embodiment 3 (there are links between different RS groups). detailed description

In order to make the technical solution of the present invention clearer and more complete, the technical scheme of the present invention will be described in detail below with reference to the accompanying drawings;

The RS in the embodiment of the present invention may be a simple relay having only a data forwarding function, or may be a base station or a user equipment having a relay function.

2 is a flowchart of a method for selecting a relay path based on a maximum throughput criterion according to the present invention. The method is applicable to any multi-hop hierarchical network, and each RS may be associated with only the RS group of the preceding and succeeding stages, or may be each The RS is associated with a multi-level RS group (> level 2). In this method, each RS needs to have a path selection function and a function of selecting a result for a subsequent stage broadcast route. The following describes the symbol identifiers used in this method:

CQITM. : Corresponds to the ί· ( = l, 2,..., L _m — RSs in the (m - 1)th RS group to the C^I'UJ RSs in the mth RS group CQI, where m = l,..., (M - 1) , L _m refers to the number of RSs in the mth RS group;

CIDTM. : Corresponding to the (m - 1)th RS group = l, 2,..., L _m - RS to mth The CID of the jth (= l, 2, ..., L _m ) RSs in the RS group, the CID refers to the group and other groups

The number of the RS path.

The specific steps of the relay path selection method based on the maximum throughput criterion are as follows:

Step 201: The DeNB and all activated RSs in the system send measurement signals to the subsequent stage. Step 202: The UE measures all the previous RS groups that may be connected thereto to obtain a set of CQ3⁄4. Meanwhile, each RS also measures all A group of CQs that may be connected to the previous RS group;

Step 203: The UE feeds back CQ3⁄4 and CID^ to the RS group of the previous level;

Step 204: Each UE connected to the RS ^ - ¹ feedback received CQl pCID UE, the CID stored ^, measured binding level before CQ - ¹ (Note: RS and CQ - _ ¹ / corresponds CQ¾ The CQ- ¹ and CQ3⁄4 in the CQI group are formed, and the obtained CQI group and CID- ^{1 are} fed back to the previous RS group;

Step 205: each of the groups after the RS ² receives a CQI feedback and RS CIE ^ - ^1, the binding level measured before CQ - ² retention calculated maximum CQI certain criteria group (where CQ -. ² j corresponds CQ - i ^1, and CQ - ¹ corresponds to each i j = 1, 2, ..., L M _,, so that each RS - ² has a corresponding L _M - ₂ a front- The CQI value of each previous level corresponds to the CQI value of the L _M - the first level of the CQI group, so the possible CQI group has Ζ^- χΖ^^), that is, RS^ ^{-2 is} in front of each The CQI group corresponding to the first-level CQI value is selected as the surviving CQI group, and the corresponding path is used as the persistence path, and only the remaining CQI group corresponding to the selected remaining path is fed back to the RS, and the ClO is fed back. - ² , and save the CID^ of the persistence path;

Step 206: The operation of RS (m = l, 2, ..., M_3) is the same as RS; ^.- ² , and the RS update remains.

The CQI group and the persistence path, and the previous level RS group feeds back the corresponding CID and the newly selected surviving CQI group;

Step 207: The DeNB calculates a surviving CQI group of the RS. feedback according to the maximum throughput criterion, and selects a path corresponding to the maximum throughput value as a final selected path result. Step 208: The DeNB broadcasts the selection result to the subsequent one stage, that is, the selected RS is confirmed by broadcasting the corresponding signal, to prepare the forwarding of the subsequent DeNB data, and the other unselected RSs are in the silent state for the DeNB. Similarly, the selected RS broadcasts the selection result to the next level until it reaches the UE, so that the entire path selection is completed. FIG. 3 is a flowchart of a method for selecting a relay path based on a shortest path criterion according to the present invention. In this method, each RS needs a function of having a path selection function and a result of selecting a broadcast path to a subsequent stage. The following describes the symbol identifiers used in this method:

Path_Num;: corresponds to the path hops of the (; = 1, 2, ..., L _m ) RSs in the mth RS group, where m = l,..., (M - 1) , L _m refers to the number of RSs in the mth RS group;

CIDTM. : Corresponds to the first (m-1) RS group = l, 2, ..., L _m - RS to mth

The CID of the jth (= l, 2, ..., L _m ) RSs in the RS group, the CID refers to the group and other groups

The number of the RS path.

The relay path selection method based on the shortest path criterion specifically includes the following steps: Step 301: The DeNB and all activated RSs in the system send a path selection enable signal to the next stage;

Step 302: The UE feeds back Path_Nm (current value is 0) and CI to the RS group of the previous level connected thereto;

Step 303: Each RS receives feedback from the UE.

Update Path_Num (that is, add hop 1) as Path_Num — ¹ , and feed Path_Num — ¹ and dD of the current RS group to the previous RS group;

Step 304: each RS ² after receiving a feedback Path_Num and CIE ^ - ^1, the respective Path_Num comparison, selecting the smallest path Path_Num path as retention, update Path_Num minimum retention (i.e., number of hops plus 1) Recorded as Path_Num - ² , feedback to RS" Path.Num^" ² and CID^- ² , and save the CID^- ^{1 of the} persistence path;

Step 305: RS^ (m = l, 2, ..., M_3) operates in the same manner as RS ² , and the RS update remains.

Path_Num and the persistence path, and feedback the corresponding CID and the updated Path_Num to the RS group of the previous level;

Step 306: The DeNB calculates the persistence Path_Num of the feedback according to the shortest path criterion, and selects the minimum value corresponding path as the last selected path result;

Step 307: The DeNB broadcasts the selection result to the next stage, that is, the corresponding signal to confirm the selected RS, to prepare the forwarding of the subsequent DeNB data, and at the same time, the other unselected RSs are in a silent state for the DeNB. Similarly, the selected RS broadcasts the selection result to the next level until it reaches the UE, so that the entire path selection is completed.

Compared with the maximum throughput criterion, the relay path selection method based on the shortest path criterion has the advantages of simple implementation and less feedback signaling. However, the shortest path criterion is only applicable to the direct connection between different RS groups in a multi-hop relay hierarchical network, because the path hop count has a size difference at this time; and the maximum throughput criterion is applicable to any multi-hop relay hierarchical network. Regardless of whether there is a direct connection between different RS groups, see the examples below. Example 1:

4 is a schematic diagram of a multi-hop relay hierarchical network based on Embodiment 1 of the present invention (only the front and back RS groups are associated), and the multi-hop relay hierarchical network includes 3 hops (ie, M=3) two layers (ie, L=2). ). If the shortest path criterion is adopted, the hop count of any path is the same and cannot be distinguished. Therefore, the maximum throughput criterion is adopted. Let the CQI level be: 1-16 (4 bits), randomly distributed on each CID; compare the two schemes: feedback all CQI and CID are selected by the DeNB (conventional method) and the method provided by the present invention, and the selection of the final path is accurate. Degree and signaling overhead.

In this example, there are four paths from the DeNB to the UE. The CQI distribution is shown in Table 1 below (the value in the box where the rows and columns meet is the CQI level value): Table 1 CQI distribution (2RS/hop, 3 hops total)

Assuming that all CQI and CID are fed back for DeNB selection, path 1 (shown in dashed lines in the figure) is the result of the last selection. If the method of the present invention is used, the two sets of CQI (4, 10) and (3, 12) are compared in the RS; the maximum throughput is compared, and the maximum value (4, 10) corresponding path is the retention path, RS; The updated CQI group (4, 10, 8) to the DeNB, while saving the CID (RS; ->RSf); similarly, the RS compares the two sets of CQI (4, 1) and (3, 9), taking the maximum value (3,9) The corresponding path is the persistence path, and the updated CQI group (3, 9, 15) is fed back to the DeNB, and the CID (RS^->RS^) is saved; the DeNB compares the two groups (4, 10, 8) And (3, 9, 15), take the (4, 10, 8) corresponding path as the final result, and broadcast to the next level RS, RS; after receiving the selection result, the selection result is broadcast to RSf, and RSf will select the result again. Broadcast to the UE. The signaling overhead is shown in Table 2. It can be seen that the selection result of the optimal path is the same, and the signaling overhead is reduced.

Table 2 Comparison of Signaling Costs PRS/hop, a total of 3 hops)

Example 2:

FIG. 5 is a schematic diagram of a multi-hop relay hierarchical network based on Embodiment 2 of the present invention (only the front and back RS groups are associated), and the main parameters of the multi-hop relay hierarchical network are: M=4, L=3. If using the shortest path The criterion is that the hop count of any path is the same and cannot be distinguished. Therefore, the maximum throughput criterion is adopted. Set CQI level: l-16 (4bits), randomly distributed on each CID; compare two schemes: feedback all CQI and CID are selected by DeNB (conventional method) and the method of the present invention, evaluate the selection accuracy and letter of the last path Make the cost.

There are a total of 27 paths in this embodiment. The specific search process is performed according to the detailed steps of the present invention, and details are not described herein again. The last best path found is the same as the conventional method, both (7131611), as shown by the dashed line in the figure. The signaling overhead is shown in Table 3. It can be seen that the selection result of the optimal path is the same, and the signaling overhead is greatly reduced.

Table 3 Comparison of signaling overhead (3RS/hop, 4 hops total)

In addition to the above implementations, the method provided by the present invention can be extended to any M and L values (M ≥ ³ , L ≥ ² ), and the signaling overheads of the two methods are calculated as shown in Table 4. It can be seen that with M As the value of L and L increase, the signaling overhead is greatly reduced.

Table 4 Comparison of signaling overhead (any hop count and number of relays)

Conventional method

CQI feedback r ₍ -3)( -4) _L4+3(M _ _3)L3 [ ₍ M — ₂₎₍ M _{+ 1)L2 +} ( ) ] + (M +l)L ² +2L)

CID feedback r ₍ -3)( -4) _{L4 +3(} _ _3)L3 [(M - 2)L ² + 2L)J + (M +l)L ² +2L)

Broadcast path [L( -1) + 1] M

ID Example 3:

6 is a schematic diagram of a multi-hop relay hierarchical network based on Embodiment 3 of the present invention (the RS can be associated with a multi-level RS group, not only the front-back RS group is associated), and the main parameters of the multi-hop relay hierarchical network are: M =4, h, =2, L ₂ =2 , L ₃ =3 ; Comparing the two schemes: Feedback All Path_Num and CID are selected by the DeNB (conventional method) and the method of the present invention to evaluate the selection accuracy and signaling of the last path Overhead. There are a total of 13 paths in this embodiment. The specific search process is performed according to the detailed steps of the present invention, and details are not described herein again. The last best path found is the same as the conventional method, as shown by the dashed line in the figure. The signaling overhead is shown in Table 4. The same is true for the signaling overhead.

Table 4 Comparison of signaling overhead (contact between different RS groups, 3 hops total)

Compared with the link quality information/Path_Num and CID methods of the existing feedback all paths, the method according to the present invention has the following characteristics:

(1) The relay path selection method proposed by the present invention is based on a hierarchical network structure having a universal significance, and has a wider application range;

(2) The relay path selection method proposed by the present invention selects an optimal path based on a maximum throughput or a shortest path criterion.

(3) The DeNB only receives the CID of the subsequent stage and the residual link quality information group/Path_Num information, and compares the one-hop persistent link quality information group /Path_Num, which reduces the feedback signaling overhead;

(4) RS participates in path selection comparison, pre-screens out some paths, and only feeds back the link quality information group/Path_Num and CID information to the previous level, and saves the CID of the corresponding subsequent level RS in the remaining path. (5) The DeNB only needs to broadcast the first hop CID information in the selected path, and the CID information of other hops is broadcasted by the selected RS to the next stage. Industrial applicability

The invention selects the optimal path based on the maximum throughput or the shortest path criterion, and can ensure that the whole path selection result is optimal; the relay station RS participates in the path selection comparison and saves the subsequent level CID, which reduces the signaling feedback overhead; the DeNB only receives The CID and the remaining link quality information group/path total hop information of the latter stage only need to broadcast the first hop CID information in the selected path, which further reduces the signaling overhead.

Claims

Claim

A relay path selection method, which selects an optimal relay path based on a maximum throughput criterion, the method comprising:

Each relay station RS receives a connection identification number CID fed back by each RS in the RS group of the latter stage and a link quality information group corresponding to the CID;

Each RS calculates a path throughput corresponding to each path according to the link quality information group fed back by the RS in the RS group of the latter stage and the link quality information currently measured between the RSs in the RS group of the previous level, and selects The path with the largest throughput is used as the persistence path; and each of the RSs feeds back the corresponding CID of the RS group and the surviving link quality information group corresponding to the selected path, and saves the corresponding path in the existing path in the RS. The CID of the primary RS, until the latest primary RS group directly connected to the primary base station DeNB feeds back the CID and the surviving link quality information group to the DeNB;

The DeNB calculates the path throughput corresponding to each path based on the surviving link quality information group fed back by the RS group, selects a path corresponding to the maximum throughput, and notifies the selection result to the subsequent RS group. The RS further notifies the subsequent RS group according to the CID corresponding to the subsequent RS in the previously saved remaining path until the user equipment UE is notified.

2. The method according to claim 1, the method is applicable to a multi-hop hierarchical network, in which each RS in each level of the RS group has a path connection with the RS group of the preceding and succeeding stages, or each There is a path connection between each RS in the primary RS group and the preceding RS group and/or two or more multi-level RS groups.

3. The method according to claim 1, wherein the link quality information group is a set of values consisting of multiple paths, specifically channel quality information CQI group, or carrier interference noise ratio CINR group, or received signal strength indication. RSSI group.

4. The method according to claim 1, wherein the RS is a simple relay with a data forwarding function, or a base station or user equipment having a relay function.

5. The method according to claim 1, wherein the maximum throughput criterion is specifically: From all the path sets between the DeNB and the UE, one path with the largest throughput is selected as the optimal relay path.

6. A relay path selection method, which selects an optimal relay path based on a shortest path criterion, the method comprising:

Each RS receives the CID of each RS feedback in the RS group of the latter stage and the total number of hops of the path corresponding to the RS corresponding to the CID by the UE Path_Num;

According to the shortest path criterion, each RS selects the smallest path among the Path_Nums of the RS feedback in the RS group of the latter stage as the persistence path; and each RS forwards the corresponding CID to the previous RS group and the latest selection and the self a persistence Path_Num corresponding to the persistence path, and storing the CID of the subsequent primary RS in the legacy path in the local RS, until the latest primary RS group directly connected to the primary base station DeNB feeds back the CID and the persistence Path_Num to the primary base station DeNB;

The primary base station DeNB selects only one path corresponding to the minimum value based on the persistence Path_Num of the feedback of the subsequent stage, and notifies the selection result of the next-level RS group, and the selected RS further corresponds to the subsequent-level RS according to the previously saved remaining path. The CID notifies the subsequent RS group until notification to the UE.

7. The method according to claim 6, the method is applicable to a multi-hop hierarchical network, wherein each of the RSs and the pre- and post-level RS groups and/or more than two in each hierarchical RS group There is a path connection between the RS groups.

8. The method according to claim 6, wherein the RS is a simple relay with a data forwarding function, or a base station or user equipment having a relay function.

A relay path selection method, the method being applied to a multi-hop hierarchical network, the method comprising: configuring, by using a configuration module, an optimal relay path selection strategy based on a maximum throughput criterion or an optimal relay path based on a shortest path criterion Selection strategy

When the configuration uses an optimal relay path selection policy based on the maximum throughput criterion, the relay path selection method according to claim 1 is employed;

When the configuration uses the optimal relay path selection strategy based on the shortest path criterion, The relay path selection method described in claim 6.

10. A relay path selection system, the system comprising a primary base station DeNB, a relay station RS, and a user equipment UE, wherein the DeNB, the RS, and the UE form a multi-hop relay hierarchical network:

Each relay station RS of the system is configured to receive a connection identification number CID fed back by each RS in the RS group of the subsequent stage and a link quality information group corresponding to the CID; and each RS is further used according to the subsequent level The link quality information group of the RS feedback in the RS group and the link quality information currently measured between each RS in the RS group of the previous level are used to calculate the path throughput corresponding to each path, and the path with the largest throughput is selected as the persistence path. And each RS is further used to feed back the corresponding CID of the previous-level RS group and the surviving link quality information group corresponding to the selected path, and save the corresponding-level RS in the existing path in the current RS. CID;

The primary base station DeNB is configured to receive the CID and the surviving link quality information group of the directly-connected latest-level RS group feedback, and calculate the path throughput corresponding to each path based on the surviving link quality information group fed back by the latest-level RS group. The quantity is selected, and a path corresponding to the maximum throughput is selected, and the result of the selection is notified to the latest level RS group; the RS in the selected latest level RS group is also used to correspond to the subsequent level RS in the previously saved remaining path. The CID notifies the subsequent RS group until the selection result is notified to the user equipment UE.

A relay path selection system, the system comprising a primary base station DeNB, a relay station RS, and a user equipment UE, wherein the DeNB, the RS, and the UE form a multi-hop relay hierarchical network:

Each relay station RS of the system receives the CID of each RS feedback in the RS group of the latter stage and the total number of hops Path_Num of the path that the UE reaches the CID corresponding to the RS; and each RS is also used to select the RS group of the latter stage. The smallest path in the Path_Num of the RS feedback is used as the persistence path; and each RS forwards the corresponding CID and the newly selected surviving Path_Num corresponding to the existing path to the RS group, and saves and retains in the current RS. The CID corresponding to the subsequent level RS in the path;

The primary base station DeNB is configured to receive the CID and the persistence of the directly connected RS group feedback of the latest level Path_Num, and selecting a path corresponding to the minimum value based on the persistence Path_Num of the feedback of the latest level RS group, and notifying the result of the selection to the RS group of the latest level; the RS in the selected first-level RS group is saved according to the previous The CID corresponding to the latter-level RS in the persistence path notifies the subsequent-stage RS group until the selection result is notified to the UE.

12. A relay path selection system, the system comprising a primary base station DeNB, a relay station RS, and a user equipment UE, wherein the DeNB, the RS, and the UE form a multi-hop relay hierarchical network;

The system also includes a configuration module for configuring the system to use an optimal relay path selection policy based on a maximum throughput criterion or an optimal relay path selection policy based on a shortest path criterion;

When configuring the system to use an optimal relay path selection policy based on a maximum throughput criterion, each network element in the system functions as each network element in the system according to claim 10.

When configuring the system to use an optimal relay path selection policy based on the shortest path criterion, each network element in the system functions as each network element in the system of claim 11.