CN101471756B - HARQ method in multi-hop relay downlink system with centralized scheduling - Google Patents

Abstract

The invention provides a hybrid automatic repeat request (HARQ) method for a centralized-scheduling multi-hop relay downlink system. When employing the non-self-adapting HARQ and the relevant signaling feedback mode and flow, the method enables the HARQ retransmission to be performed on the resource block that is allocated during the first-time transmission, without needing rescheduling. In order to increase the resource utilization rate, the method designs the signaling feedback to notify the base station to release the resources of the already-successful link as early as possible, thereby greatly reducing the retransmission time delay. The relay station employs the assembly line-type HARQ and the self-adapting hybrid forward mode. The relay station receives the data packet and then decodes and forwards the data packet at the same time, just like the mode of the assembly line. The relay station employs the decoding forwarding under such condition that the time-slot allows the decoding operation and the decoding operation is successful, otherwise employs the amplification forwarding or the demodulation forwarding; that is, the relay station automatically adapts to select the forwarding mode according to the actual condition, thereby greatly reducing the end-to-end time delay and preventing the influence of coding delay and providing the better service for the real-time traffics.

Description

HARQ method in multi-hop relay downlink system with centralized scheduling

technical field

The present invention is a novel HARQ (hybrid automatic retransmission) scheme in a centralized scheduling multi-hop relay downlink system. It is proposed to solve the problems of data transmission delay in multi-hop relay system and signaling transmission delay in centralized scheduling.

Background technique

In a traditional wireless communication system, data transmission is performed directly between a base station (BS) and a mobile station (MS), which is also called a single-hop system. However, with the continuous development of future wireless communication systems to higher data rates, wider frequency bands, and higher-end spectrums, it has gradually been unable to meet the needs. Therefore, the multi-hop relay system has attracted more and more attention due to its advantages of increasing coverage and improving spectrum efficiency. There are two resource management methods in the multi-hop relay system, centralized and distributed scheduling. In centralized scheduling, BS is responsible for all uplink and downlink resource allocation, while in distributed scheduling, BS and each relay station (RS) are capable of resource allocation. Because centralized scheduling has a simpler protocol structure and interference management, it is currently the preferred research hypothesis for multi-hop relay systems.

A major disadvantage of the multi-hop relay system is that the end-to-end delay is large, because the transmission between the BS and the MS has to go through one or more RSs. Especially for centralized scheduling, this problem is more serious. The bandwidth request of any link must be submitted to the BS for resource scheduling through multiple hops, and then the relevant nodes will be notified through multiple hops, which increases the HARQ retransmission delay and signal quality. order overhead. Therefore, in a multi-hop relay system, how to reduce the end-to-end delay to provide better services for real-time services, such as video on demand, multimedia video streaming, and interactive games, is a major challenge.

A HARQ method for a multi-hop relay downlink system with centralized scheduling is proposed in the prior art. Since in centralized scheduling, any bandwidth request needs to be submitted to the BS for resource scheduling, so if a transmission failure occurs in a multi-hop link, especially a link that is not directly adjacent to the BS and needs to request HARQ, a specially designed feedback is required The mechanism notifies the BS which link is faulty, so that the BS can allocate resources to the link for retransmission. Therefore, this method also specifically proposes a way to perform orthogonal coding on the uplink feedback ACK channel to indicate to the BS whether the transmission of different links fails or not, so as to request bandwidth for retransmission. The specific method is as follows:

First, the BS allocates bandwidth for all links from the BS to the user terminal (UE) for the first HARQ transmission of the data packet;

Then, the RS located on the path receives the downlink data packet and decodes it. If the decoding fails, for example, the CRC check fails, the RS will send an uplink orthogonally coded NACK ₀ feedback to the previous hop and stop the continued forwarding of the erroneous data packet, waiting for retransmission; if the decoding is successful, the RS The data packet is forwarded to the next hop, and the uplink feedback is not performed temporarily, but the feedback signaling of the next hop is waited for. When the RS receives the uplink feedback from the next hop as NACK _k , it sends another orthogonally coded NACK _k+1 to the previous hop; when the RS receives the uplink feedback from the next hop as ACK, it still forwards the ACK Give the front a jump.

Next, when the UE receives the data packet and decodes it correctly, it will feed back ACK to the previous hop RS, otherwise it will feed back NACK ₀ .

Finally, when the BS receives NACK _K feedback, it indicates that the data packet has an error at the (K+1) hop, so that the BS re-allocates resources for the (K+1) hop and all subsequent links for retransmission, and transmits the resource through the downlink signaling Notify the corresponding node. When the BS receives the ACK feedback, it indicates that the data packet has been correctly received by the UE, and the transmission of the data packet ends.

The above method has two characteristics: 1) Each retransmission requires the BS to re-allocate resources, that is, an adaptive HARQ scheme. This scheme has the advantage of high spectrum utilization, but in the multi-hop relay system with centralized scheduling, any bandwidth request must notify the BS for resource scheduling, so this scheme will increase the HARQ retransmission delay; 2) RS can only be correct Only after decoding, the data packet is forwarded to the next hop, that is, the selective forwarding protocol. The premise of this protocol is that the RS first decodes the data packet and then judges whether to forward it. Due to the decoding delay, it is impossible for the RS to forward it in the time slot immediately after receiving the data packet, thus increasing the end-to-end transmission time. delay.

Figure 1 and Figure 2 show two examples of existing methods. In the specification of the present invention, it is assumed that 3 RSs are required between the BS and the UE, which are represented by R1, R2, and R3 respectively, and t1, t2, ..., tn represent time slots. Due to the delay in decoding the data packet, each RS needs to delay at least one time slot to forward the data even if the decoding is correct. A solid circle indicates that the RS performs channel decoding.

Example 1: The data Data1 is transmitted successfully for the first time. As shown in Figure 1, the process is as follows:

R1 receives Data1 at time slot t1, and re-encodes it after correct decoding. Due to the decoding delay of R1, the data packet cannot be forwarded at t2, and the data packet can only be forwarded to R2 at time slot t3 at the fastest. Subsequent RSs perform similar operations. When UE decodes correctly, it will feed back ACK at t8, and pass it to BS via R3, R2, and R1. Data1 transfer ends correctly. In the case of Example 1, the UE can receive Data1 only at time slot t7 in the fastest case, and the end-to-end delay is 7 time slots.

Example 2: The first transmission of data Data2 in R1 and R3 fails. As shown in Figure 2, the process is as follows:

R1 receives the data Data2 in time slot t1, and the decoding error occurs, then at t2, it feeds back NACK ₀ to request retransmission to the BS, and no longer forwards the data. The BS retransmits Data2 at t3 at the earliest. R1 decodes correctly after merging the two received versions, and forwards Data2 to R2 at t5. After R2 decodes correctly, it is forwarded to R3 at t7. At this time, R3 has a decoding error, and at t8, NACK ₀ is fed back to request retransmission. When R2 receives NACK ₀ , it re-encodes and forwards NACK ₁ , and when R1 receives NACK ₁ , it re-encodes and forwards NACK ₂ . The BS receives NACK ₂ , and the BS knows that there is an error in the link from R2 to R3, then reschedules resources for all subsequent links of this link, and notifies relevant nodes through signaling. R2 retransmits Data2 at t13 after receiving the resource allocation signaling, R3 combines the two received versions and decodes them correctly, and forwards them to UE at t15. The UE decodes correctly and feeds back ACK at t16, which is passed to BS via R3, R2, and R1. Data2 transfer ends correctly. In Example 2, the fastest end-to-end delay of Data2 is 15 time slots.

It is worth noting that the time slot arrangements in the above examples all represent the fastest possible situation, and the actual system may have different time slots according to system load conditions and scheduling conditions.

Contents of the invention

Aiming at the problems of the centralized scheduling multi-hop relay downlink system and the defects of the existing HARQ method, the present invention provides a new type of HARQ method in the centralized scheduling multi-hop relay downlink system, and specifically provides the following three technologies plan:

A HARQ method in a multi-hop relay downlink system of centralized scheduling, comprising the following steps:

First, the base station allocates bandwidth for all links from the base station to the user terminal, and reserves this resource until the data packet is successfully transmitted;

Then, the relay station on the path receives the downlink data packet and decodes it. If the decoding fails, the relay station sends an uplink orthogonally coded NACK to the previous hop and stops the erroneous data packet from continuing to forward, waiting for retransmission. If the decoding is successful, the relay station forwards the data packet to the next hop, and sends an uplink orthogonal coded ACK ₀ to feed back to the previous hop;

When the uplink feedback received by the relay station is ACK _k , it sends another orthogonally coded ACK _k+1 to the previous hop; when the uplink feedback received by the relay station is NACK, the relay station that receives the NACK is reserved Packet retransmission on the resource block;

Next, when the user terminal receives the data packet and decodes it correctly, it will feed back ACK ₀ to the previous hop relay station, otherwise it will feed back NACK;

Finally, when the base station receives the ACK _K feedback, indicating that the data packet has succeeded at the K+1 hop, the resources reserved for the K+1 hop and all previous links are released. When the base station receives the NACK feedback, the Data packet retransmission is performed on the reserved resource blocks.

A HARQ method in a multi-hop relay downlink system of centralized scheduling, comprising the following steps:

First, the base station allocates bandwidth for all links from the base station to the user terminal for the first HARQ transmission of the data packet;

Then, each relay station located on the path receives the downlink data packet and forwards it in the mode of amplification and forwarding (Amplif-and-Forward, abbreviated as "AF") or demodulation and forwarding (Demodulate-and-Forward, abbreviated as "DmF") For the next hop, the relay station monitors the signaling sent by the previous hop. If it hears ACK ₀ , it means that the previous hop decodes correctly, and the relay station also decodes the data packet; if it hears NACK ₀ , it means that the previous hop If the decoding is wrong, the relay station buffers the data packet;

When the relay station performs decoding, if the decoding fails, the relay station sends an orthogonally coded NACK ₀ , otherwise it sends an orthogonally coded ACK ₀ , and its previous hop and next hop nodes monitor the signaling at the same time;

If the relay station receives NACK _k from the next hop, it sends another orthogonally coded NACK _k+1 to the previous hop; if the relay station receives ACK ₀ from the next hop, it does not take Any operation; when the relay station receives the signaling from the next hop as ACK ₁ , it still forwards ACK ₁ to the previous hop;

Then, when the user terminal receives the data packet, it firstly monitors the signaling of the previous hop, if it hears ACK ₀ , it means that the decoding of the previous hop is correct, then the user terminal also decodes the data packet; if it hears NACK ₀ , indicating that the previous hop was incorrectly decoded, and the user terminal only needs to buffer the data packet and no longer decode it. When the user terminal decodes, if it is correct, then feed back ACK ₁ to the relay station of the previous hop, otherwise feed back NACK ₀ ;

Finally, when the base station receives NACK _K feedback, it indicates that the data packet has an error at the K+1 hop, so the base station re-allocates resources for the K+1 hop and all subsequent links for retransmission, and notifies the corresponding relay station through downlink signaling, When the corresponding relay station receives the downlink resource allocation signaling, it adopts the Decode-and-Forward (Decode-and-Forward, abbreviated as "DcF") mode to retransmit the data packet. When the base station receives ACK ₀ feedback, it does not take any action; when the base station receives If ACK ₁ is fed back, it indicates that the data packet has been correctly received by the user terminal, and the transmission of the data packet ends.

A HARQ method in a multi-hop relay downlink system of centralized scheduling, comprising the following steps:

First, the base station allocates bandwidth for all links from the base station to the user terminal, and reserves this resource until the data packet is successfully transmitted;

Then, each relay station on the path receives the downlink data packet and forwards it to the next hop in the mode of amplification and forwarding or demodulation and forwarding. The relay station monitors the signaling sent by the previous hop. If it hears ACK ₀ , it means that the previous hop If the code is correct, the relay station will also decode the data packet; if it hears NACK, it means that the previous hop decoding error, then the relay station only needs to buffer the data packet and no longer decode it;

When the relay station performs decoding, if the decoding fails, the relay station sends an orthogonally encoded NACK, otherwise it sends an orthogonally encoded ACK ₀ , and its previous hop and next hop nodes monitor the signaling at the same time;

If the relay station receives the signaling from the next hop as ACK _k , it sends another orthogonally coded ACK _k+1 to the previous hop; if the relay station receives the signaling from the next hop as NACK, it sends On the resource block of , the data packet is retransmitted in decoding and forwarding mode;

Then, when the user terminal receives the data packet, it first monitors the signaling of the previous hop. If it hears ACK ₀ , it means that the decoding of the previous hop is correct, and the user terminal also decodes the data packet; if it hears NACK, It means that the decoding error in the previous hop, the user terminal only needs to buffer the data packet and no longer decode it. When the user terminal performs decoding, if it is correct, it will feed back ACK ₀ to the relay station of the previous hop, otherwise it will feed back NACK;

Finally, when the base station receives the ACK _K feedback, it indicates that the data packet has succeeded at the K+1 hop, thereby releasing the resources reserved for the K+1 hop and all previous links. When the base station receives the NACK feedback, it Data packet retransmission is performed on the remaining resource blocks.

The HARQ method in the centralized scheduling multi-hop relay downlink system proposed by the present invention is applied to the HARQ transmission protocol of the downlink MAC layer of the centralized scheduling multi-hop relay system. Compared with the prior art, the present invention can effectively reduce the end-to-end transmission delay of the multi-hop relay system with centralized scheduling, and make up for the defects and deficiencies of the existing solutions. Provide better services for real-time business.

Description of drawings

Fig. 1, Fig. 2 are the flow charts of the HARQ method of the multi-hop relay downlink system of the existing centralized scheduling in two cases;

Fig. 3, Fig. 4 are the example flow charts in two cases of embodiment one of the present invention;

Fig. 5 and Fig. 6 are example flowcharts in two cases of Embodiment 2 of the present invention;

Fig. 7 and Fig. 8 are example flowcharts of Embodiment 3 of the present invention in two cases.

Detailed ways

Three embodiments of a HARQ method in a centralized scheduling multi-hop relay downlink system are:

Embodiment 1, a non-adaptive HARQ method for selectively forwarding RS, the steps are as follows:

First, the BS allocates bandwidth for all links from the BS to the UE, and the resource is reserved until the data packet is successfully transmitted.

Then, the RS located on the path receives the downlink data packet and decodes it. If the decoding fails (such as CRC check), the RS will send an uplink orthogonally coded NACK to the previous hop and stop the continued forwarding of the wrong data packet, waiting for retransmission; if the decoding is successful, the RS will send the NACK to the previous hop The data packet is forwarded to the next hop, and an uplink orthogonal coded ACK ₀ is sent back to the previous hop.

When the uplink feedback received by the RS is ACK _k , it sends another orthogonally coded ACK _k+1 to the previous hop; when the uplink feedback received by the RS is NACK, the RS that receives NACK is in the reserved Data packet retransmission is performed on the resource block, and NACK does not need to be forwarded to the previous hop.

Next, when the UE receives the data packet and decodes it correctly, it will feed back ACK ₀ to the RS of the previous hop, otherwise it will feed back NACK.

Finally, when the BS receives the ACK _K feedback, it indicates that the data packet has succeeded at the (K+1)th hop, thereby releasing the resources reserved by the (K+1)th hop and all previous links. When the BS receives NACK feedback, it retransmits the data packet on the resource block reserved by the BS.

In this method, if any link error occurs, the corresponding sending node will perform HARQ retransmission on the reserved resource block in the specified time slot without re-requesting resources from the BS; in order to improve resource utilization, the successful link The path will feed back the orthogonally coded ACK signaling to notify the BS to release the corresponding resources as early as possible, thereby greatly reducing the retransmission delay.

The specific examples are as follows:

Example 1: The data Data1 is transmitted successfully for the first time. As shown in Figure 3, the process is as follows:

R1 receives Data1 at time slot t1, decodes it correctly, and feeds back ACK ₀ to BS at time slot t2. Due to the decoding delay, R1 can only forward Data1 to R2 at time slot t3 at the fastest. BS receives ACK ₀ and knows that R1 has received it correctly, then releases the resources reserved for the link from BS to R1. R2 receives Data1, decodes it correctly, feeds back ACK ₀ to R1 at t4, and forwards Data1 to R3 at t5. After receiving ACK ₀ , R1 re-encodes and feeds back ACK ₁ to BS at t5. BS receives ACK ₁ and knows that R2 has received it correctly, then releases the resources reserved for R1 to R2. Subsequent RSs perform similar operations. The UE decodes correctly, feeds back ACK ₀ at t8, re-encodes through R3, R2, and R1 respectively, and finally transmits ACK ₃ to the BS, and the BS knows that the UE has successfully received it, and releases all resources. Data1 transfer ends correctly. In Example 1, the end-to-end delay is 7 time slots.

Example 2: The first transmission of data Data2 between R1 and R3 fails. As shown in Figure 4, the process is as follows:

R1 receives Data2 in time slot t1, and the decoding is wrong, then feedbacks NACK to BS to request retransmission at t2, and no longer forwards the data. BS receives NACK, then retransmits Data2 at t3 on the reserved resource block. R1 combines the two received versions and decodes correctly, feeds back ACK ₀ to BS at t4, and forwards Data2 to R2 at t5. BS receives ACK ₀ and knows that R1 has received it correctly, then releases the resources reserved for the link from BS to R1. R2 receives Data2 and decodes it correctly, then feeds back ACK ₀ to R1 at t6 and forwards Data2 to R3 at t7. After R1 receives ACK ₀ , it re-encodes and feeds back ACK ₁ to BS at t5. BS receives ACK ₁ and knows that R2 has received it correctly, then releases the resources reserved for the link from R1 to R2. After receiving Data2, R3 makes a decoding error, and returns NACK at t8 to request retransmission. R2 receives NACK and retransmits Data2 on the reserved resource block at t9. R3 decodes correctly after combining the two received versions, then feeds back ACK ₀ to R2 at t10 and forwards Data2 to UE at t11. R2 receives ACK ₀ , re-encodes and forwards ACK ₁ to R1, and R1 receives ACK ₁ , re-encodes and forwards ACK ₂ to BS. BS receives ACK ₂ and knows that R3 has received it correctly, then releases the resource reserved for the link from R2 to R3. The UE receives Data2, decodes it correctly, feeds back ACK ₀ at t12, re-encodes it through R3, R2, and R1, and finally transmits ACK ₃ to the BS. The BS knows that the UE has successfully received it, releases all resources, and the transmission of Data2 ends correctly. In Example 2, the end-to-end delay is 11 time slots.

Embodiment 2, the adaptive HARQ method of pipelined RS, the steps are as follows:

First, the BS allocates bandwidth to all links from the BS to the UE for the first HARQ transmission of data packets.

Then, each RS located on the path receives the downlink data packet, and can forward it to the next hop in the AF or DmF mode at the fastest in the next time slot. The RS listens to the signaling sent by the previous hop. If it hears ACK ₀ , it means that the previous hop decoded correctly, and the RS also decodes the data packet; if it hears NACK ₀ , it means that the previous hop decoded incorrectly. The RS only needs to buffer the data packets without decoding them.

When the RS is decoding, if the decoding fails (such as CRC check), the RS sends an orthogonally coded NACK ₀ , otherwise it sends an orthogonally coded ACK ₀ , and its previous hop and next hop nodes monitor the signal at the same time make.

If the RS receives the signaling from the next hop as NACK _k , it sends another orthogonally coded NACK _k+1 to the previous hop; if the RS receives the signaling from the next hop as ACK ₀ , it does not take Any operation; when RS receives the signaling from the next hop as ACK ₁ , it still forwards ACK ₁ to the previous hop.

Next, when the UE receives the data packet, it first monitors the signaling of the previous hop. If it hears ACK ₀ , it means that the decoding of the previous hop is correct, and the UE also decodes the data packet; if it hears NACK ₀ , it means If the previous hop decodes incorrectly, the UE only needs to buffer the data packet and no longer decode it. When the UE performs decoding, if it is correct, it will feed back ACK ₁ to the RS of the previous hop, otherwise it will feed back NACK ₀ .

Finally, when the BS receives NACK _K feedback, it indicates that the data packet has an error at the (K+1) hop, so that the BS re-allocates resources for the (K+1) hop and all subsequent links for retransmission, and transmits the resource through the downlink signaling The corresponding RS is notified, and the corresponding RS adopts the DCF mode to retransmit the data packet after receiving the downlink resource allocation signaling. When the BS receives ACK ₀ feedback, it does not take any action; when the BS receives ACK ₁ feedback, it indicates that the data packet has been correctly received by the UE, and the transmission of the data packet ends.

In this method, the RS can forward the data packet while decoding it after receiving the data packet. Like the pipeline mode, the RS adopts DCF when the time slot allows decoding and the decoding is successful, otherwise it adopts the amplification and forwarding AF Or demodulate and forward the DmF, so that the end-to-end delay is greatly reduced, and it is not affected by the decoding delay.

The specific examples are as follows:

Example 1: The data Data1 is transmitted successfully for the first time. As shown in Figure 5, the process is as follows:

R1 receives Data1 in time slot t1, and uses amplification and forwarding or demodulation forwarding to forward the data packet in time slot t2 while decoding. R1 decodes correctly and feeds back ACK ₀ , and BS and R2 monitor at the same time. When R2 receives Data1, it listens to the ACK ₀ signaling of R1, and decodes Data1 while using AF or DmF to forward the data packet in time slot t3. Subsequent RSs perform similar operations. The UE decodes it correctly, and feeds back ACK ₁ at t5, and transmits it to the BS via R3, R2, and R1. Data1 transfer ends correctly. In Example 1, the end-to-end delay is 4 time slots. Among them, ACK ₀ is sent by each RS, and ACK ₁ is sent by UE.

Example 2: The first transmission of data Data2 in R1 and R3 fails. As shown in FIG. 6 , the hollow circle indicates that the RS does not decode because the previous hop is NACK ₀ . The process is as follows:

R1 receives Data2 in time slot t1, and uses AF or DmF to forward the data packet in time slot t2 while decoding. R1 makes a decoding error, and feeds back NACK ₀ at t2, and BS and R2 monitor at the same time. When R2 receives Data2 and listens to the NACK ₀ signaling of R1, it buffers the received Data2 without decoding it, and uses AF or DmF to forward Data2 and NACK ₀ in time slot t3. Subsequent RSs perform similar operations. BS receives NACK ₀ , then reschedules resources for all links, and retransmits Data2 at t3. R1 merges and decodes the two received versions and at the same time uses AF or DmF to forward the data packet in time slot t4. R1 decodes correctly and feeds back ACK ₀ , and BS and R2 monitor at the same time. R2 listens to R1's ACK ₀ signaling when receiving Data2, then merges and decodes the two received versions of Data2, and uses AF or DmF to forward the data packet in time slot t5. R2 decodes correctly and feeds back ACK ₀ , and R1 and R3 monitor at the same time. R3 listens to R2's ACK ₀ signaling when receiving Data2, then merges and decodes the two received versions of Data2, and uses AF or DmF to forward the data packet in time slot t6. R3 makes a decoding error, and feeds back NACK ₀ at t6, and R2 and UE monitor at the same time. When the UE receives Data2 and listens to the NACK ₀ signaling of R3, it buffers the received Data2 without decoding it. When R2 receives NACK ₀ , it re-encodes and forwards NACK ₁ , and when R1 receives NACK ₁ , it re-encodes and forwards NACK ₂ . The BS receives NACK ₂ and knows that the link from R2 to R3 is faulty, then reschedules resources for all subsequent links of this link, and notifies relevant nodes through signaling. After receiving the resource allocation signaling, R2 retransmits Data2 in DcF mode at t11, R3 combines and decodes the three received versions, and forwards it to UE in AF or DmF mode at t12. R3 decodes correctly, feeds back ACK ₀ at t12, and R2 and UE monitor at the same time. When the UE receives Data2 and listens to the ACK ₀ signaling of R3, it combines and decodes the three received versions. UE decodes correctly, feeds back ACK ₁ , and transmits it to BS via R3, R2, and R1. Data2 transfer ends correctly. In Example 2, the end-to-end delay is 12 time slots.

Embodiment 3, non-adaptive HARQ method of pipelined RS

First, the BS allocates bandwidth for all links from the BS to the UE, and the resource is reserved until the data packet is successfully transmitted.

Then, each RS located on the path receives the downlink data packet, and can forward it to the next hop in the AF or DmF mode at the fastest in the next time slot. The RS listens to the signaling sent by the previous hop. If it hears ACK ₀ , it means that the previous hop decoded correctly, and the RS also decodes the data packet; if it hears NACK, it means that the previous hop decoded incorrectly. RS only needs to buffer the data packet without decoding it.

When the RS decodes, if the decoding fails (such as CRC check), the RS sends an orthogonally coded NACK, otherwise it sends an orthogonally coded ACK ₀ , and its previous hop and next hop nodes monitor the signal at the same time make.

If the RS receives the signaling from the next hop as ACK _k , it sends another orthogonally coded ACK _k+1 to the previous hop; if the RS receives the signaling from the next hop as NACK, it The DcF mode is used to retransmit the data packet on the resource block.

Next, when the UE receives the data packet, it first monitors the signaling of the previous hop. If it hears ACK ₀ , it means that the decoding of the previous hop is correct, and the UE also decodes the data packet; if it hears NACK, it means that the previous hop If there is a decoding error in one hop, the UE only needs to buffer the data packet and no longer decode it. When the UE performs decoding, if it is correct, it will feed back ACK ₀ to the RS of the previous hop, otherwise it will feed back NACK.

Finally, when the BS receives the ACK _K feedback, it indicates that the data packet has succeeded at the (K+1)th hop, thereby releasing the resources reserved by the (K+1)th hop and all previous links. When the BS receives NACK feedback, it retransmits the data packet on the reserved resource block.

This method is a comprehensive scheme of the first method and the second method, and has the advantages of the first and second methods at the same time.

The specific examples are as follows:

Example 1: The data Data1 is transmitted successfully for the first time. As shown in Figure 7, the process is as follows:

The third solution is a comprehensive solution of the first method and the second method, and has the advantages of the first and second methods at the same time. For example, as shown in FIG. 7 , the end-to-end delay is only 4 time slots. The second example is shown in FIG. 8 , where the end-to-end delay is only 8 time slots.

Claims (5)

1. the HARQ method in the multi-hop relay downlink system of a centralized scheduling is characterized in that may further comprise the steps:

At first, the base station is to the good bandwidth of all link assignment from the base station to the user terminal, and resource reservation arrives till the data packet transmission success;

Then, be positioned at relay station downlink data receiving bag on the path row decoding of going forward side by side, if decoding failure, the continuation that the NACK that this relay station then sends the uplink orthogonal coding feeds back to previous dive and stops error data packets being transmitted, and waits re-transmission; If successfully decoded, relay station then is transmitted to next to this packet and jumps, and sends the ACK of uplink orthogonal coding ₀Feed back to previous dive;

The up ACK that is fed back to orthogonal coding that receives when this relay station _k, then send the ACK of another orthogonal coding _K+1Give previous dive; As the up NACK that is fed back to orthogonal coding that this relay station receives, the relay station that then receives the NACK of orthogonal coding retransmits at the enterprising line data bag of reserved resource piece; Wherein, k represents jumping figure;

Then, when user terminal receives packet and decoding is correct, then feed back the ACK of orthogonal coding ₀Give the relay station of previous dive, otherwise the NACK of feedback orthogonal coding;

At last, receive ACK when the base station _KFeedback shows that packet in K+1 jumping success, then discharges K+1 and jumps and all link institute reserved resource of front, receives the NACK feedback when the base station, then retransmits at the enterprising line data bag of base station reserved resource piece.

2. the HARQ method in the multi-hop relay downlink system of a centralized scheduling is characterized in that may further comprise the steps:

At first, the base station is used for the HARQ transmission first time of packet to the good bandwidth of all link assignment from the base station to the user terminal;

Then, be positioned at each the relay station downlink data receiving bag on the path, adopt the pattern of amplifying forwarding or demodulation forwarding to be transmitted to next and jump, relay station is monitored the signaling that previous dive sends, if hear the ACK of orthogonal coding ₀, illustrate that previous dive decoding is correct, then this relay station is also deciphered packet; If hear the NACK of orthogonal coding ₀, the previous dive decoding error is described, then this relay station is data pack buffer;

When relay station is deciphered, if decoding failure, this relay station sends the NACK of orthogonal coding ₀, otherwise the ACK of transmission orthogonal coding ₀, its previous dive and next-hop node are monitored signaling simultaneously;

If it is the NACK of orthogonal coding that relay station is received the signaling from next jumping _k, then send the NACK of another orthogonal coding _K+1Give previous dive; If it is the ACK of orthogonal coding that relay station is received the signaling from next jumping ₀, then do not take any operation; Relay station receives that the signaling from next jumping is the ACK of orthogonal coding ₁, then still transmit the ACK of orthogonal coding ₁Give previous dive;

Then, when user terminal receives packet, it at first monitors the signaling of previous dive, if hear the ACK of orthogonal coding ₀, illustrate that previous dive decoding is correct, then user terminal is also deciphered packet; If hear the NACK of orthogonal coding ₀, the previous dive decoding error is described, then user terminal only needs data pack buffer is no longer deciphered, and when user terminal is deciphered, if correctly, then feeds back the ACK of orthogonal coding ₁Give the relay station of previous dive, otherwise the NACK of feedback orthogonal coding ₀

At last, receive the NACK of orthogonal coding when the base station _KFeedback, show that packet jumps out mistake at K+1, thereby jump K+1 the base station and all links of back are redistributed resource and retransmitted, and notify corresponding relay station by downlink signaling, the downlink resource assignment signalling of receiving corresponding relay station then adopts decoding forward mode retransmission data packet may, receives the ACK of orthogonal coding when the base station ₀Feedback is not then taked any operation; Receive the ACK of orthogonal coding when the base station ₁Feedback shows that then packet is correctly received by user terminal, and this data packet transmission finishes; Wherein, k represents jumping figure.

3. the HARQ method in the multi-hop relay downlink system of centralized scheduling according to claim 2, it is characterized in that: behind the described relay station downlink data receiving bag, adopt the pattern of amplifying forwarding or demodulation forwarding to be transmitted to next jumping at next time slot that is right after.

4. the HARQ method in the multi-hop relay downlink system of a centralized scheduling is characterized in that may further comprise the steps:

At first, the base station is to the good bandwidth of all link assignment from the base station to the user terminal, and resource reservation arrives till the data packet transmission success;

Then, be positioned at each the relay station downlink data receiving bag on the path, adopt the pattern of amplifying forwarding or demodulation forwarding to be transmitted to next and jump, relay station is monitored the signaling that previous dive sends, if hear the ACK of orthogonal coding ₀, illustrate that previous dive decoding is correct, then this relay station is also deciphered packet; If hear the NACK of orthogonal coding, the previous dive decoding error is described, then this relay station only needs data pack buffer is no longer deciphered;

When relay station is deciphered, if decoding failure, this relay station then sends the NACK of orthogonal coding, otherwise sends the ACK of orthogonal coding ₀, its previous dive and next-hop node are monitored signaling simultaneously;

If it is the ACK of orthogonal coding that relay station is received the signaling from next jumping _k, then send the ACK of another orthogonal coding _K+1Give previous dive; If it is the NACK of orthogonal coding that relay station is received the signaling from next jumping, then on the reserved resource piece, adopt decoding forward mode retransmission data packet may;

Then, when user terminal receives packet, its monitors signaling of previous dive earlier, if hear the ACK of orthogonal coding ₀, illustrate that previous dive decoding is correct, then user terminal is also deciphered packet; If hear the NACK of orthogonal coding, the previous dive decoding error is described, then user terminal only needs data pack buffer is no longer deciphered, and when user terminal is deciphered, if correctly, then feeds back the ACK of orthogonal coding ₀Give the relay station of previous dive, otherwise the NACK of feedback orthogonal coding;

At last, receive ACK when the base station _KFeedback shows packet in K+1 jumping success, thereby discharges K+1 jumping and all link institute reserved resource of front, receives the NACK feedback when the base station, then retransmits at the enterprising line data bag of reserved resource piece; Wherein, k represents jumping figure.

5. the HARQ method in the multi-hop relay downlink system of centralized scheduling according to claim 4, it is characterized in that: behind the described relay station downlink data receiving bag, adopt the pattern of amplifying forwarding or demodulation forwarding to be transmitted to next jumping at next time slot that is right after.

Images