Method and device for managing hybrid automatic repeat request cache

Abstract

The invention discloses a method and device for managing hybrid automatic repeat request (HARQ) cache. The method comprises the following steps of: storing received data into the HARQ cache, and/or reading the data from the HARQ cache, and/or deleting the data from the HARQ cache by taking a code block (CB) as a unit. The invention also discloses a device for managing the HARQ cache. By adopting the method and the device, rational distribution of the HARQ cache can be realized, so that the receiving performance of a receiver is guaranteed.

Description

technical field [0001] The present invention relates to a hybrid automatic repeat request (HARQ, Hybrid Automatic Repeat Request) technology, in particular to a method and device for managing a HARQ buffer (buffer). Background technique [0002] In the Long Term Evolution (LTE, Long Term Evolution) protocol, the size of the HARQ buffer corresponding to the capability level of the user equipment (UE, User Equipment) is defined. For example, the size of the HARQ buffer that a UE with a capability level of 3 needs to support is 1237248 storage units; in order to improve the performance of the turbo decoder, the communication receiver generally uses the logarithmic likelihood ratio (LLR, Log-Likelihood Ratio) to represent a bit (bit), so that the turbo decoder can improve the decoding performance by about 3dB . At present, the mainstream LLR bit width in the industry is 8 bits or 6 bits. Correspondingly, the bit width stored in the HARQ buffer can be 4 bits. If each storage un...

AI Technical Summary

Problems solved by technology

After evenly distributing the entire HARQ buffer, it can support the maximum number of HARQ processes, that is, when the number of HARQ processes is the maximum number of HARQ processes, it can still be guaranteed that each HARQ process can be allocated to a certain HARQ buffer. However, because the HARQ process stores data The unit used in this process is the transmission block (TB, Transfer Block), and the size of the TB may be larger than the size of the HARQ buffer allocated to the HARQ process. In this way, the TB cyclic redundancy check (CRC, Cyclic Redundancy Check) fails, only part of the data of the TB can be saved, in other words, part of the data of the TB will be discarded, and when the TB is received next time, only part of the data of the TB will be saved, so only part of the data of the TB can be saved. Merge the saved data of the TB with the data of the TB received next time, but cannot completely merge the data of the TB transmitted this time with the data of the TB received next time. Therefore, it cannot be effectively used The merged data of the TB gets the correct data of the TB, so that the receiving performance will be degraded

[0007] 2. According to K MIMO × M limit Evenly distribute the entire HARQ buffer for each process. After evenly allocating the entire HARQ buffer, although the TB data of each HARQ process can be completely merged, because the maximum number of supported processes is 8, when there are more than 9 HARQ processes When merging is required, HARQ blocking will occur, which will also cause reception performance degradation

[0008] To sum up, there is currently no technical solution for reasonably allocating and managing HARQ buffers

Images