Fronthaul Transmitter Bit Rate Reduction via Channel Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fronthaul technologies, such as CPRI, have high bandwidth requirements for 5G mobile communication systems, and existing compression methods do not adequately reduce the bit rate for uplink fronthaul signals due to their dynamic range and peak-to-average ratio, limiting the centralization of Radio Access Networks.
Innovation Solution
A method that estimates channel information for frequency-domain resource elements to reduce the number of bits required for representation, using techniques like channel compensation and interpolation, and transmits this information over the fronthaul link, allowing the receiver to reverse the process and restore the original values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If frequency-domain sampled interface is used with attenuation, then fronthaul capacity requirements are reduced, but uplink fronthaul signals still experience large power variations requiring 10+10 bits or 16+16 bits per sample
Solution Approach 1:
The patent changes the representation parameters by transitioning from time-domain samples to frequency-domain resource elements, and further by applying channel information-based compression. This transforms the signal representation from requiring 16+16 bits per time-domain sample to requiring significantly fewer bits per frequency-domain resource element, while maintaining signal fidelity through the use of channel state information for accurate reconstruction.
Solution Approach 2:
The patent introduces channel information (channel state information) as an intermediary element that enables compression. The channel information acts as a mediator that allows the transmitter to compress frequency-domain resource elements to fewer bits while providing the receiver with the necessary information to accurately reconstruct the original signal, thus resolving the contradiction between compression and signal fidelity.
2Quantity of substance
If more functionality is moved from centralized baseband processing units to remote radio units, then fronthaul bit rate is reduced, but latency requirements restrict the opportunities for centralisation
Solution Approach 1:
The patent segments the baseband processing functions between the remote radio unit and the centralized baseband processing unit. The RRU performs FFT conversion and channel information-based compression of frequency-domain resource elements, while the centralized BPU performs channel estimation and signal reconstruction. This segmentation allows latency-sensitive functions to remain distributed while enabling efficient fronthaul compression.
Solution Approach 2:
The patent applies preliminary compression at the RRU by transforming time-domain samples to frequency-domain resource elements and applying channel information-based compression before transmission over the fronthaul link. This preliminary action reduces the data volume that needs to be transmitted, thereby reducing fronthaul bit rate requirements while maintaining the ability to perform centralized processing.
Data Source
AI summary
It is provided a method for reducing bit rate requirement over an uplink fronthaul link. The method is performed in a fronthaul transmitter of a radio network node and comprises the steps of: determining a group of frequency-domain resource elements having been received over a wireless channel (40); estimating channel information (42); reducing a number of bits for representing the respective frequency-domain resource element, to thereby obtain a modified resource element (44); transmitting the modified resource elements of the group over the fronthaul link to a fronthaul receiver (46); and transmitting compression metadata (48), indicating how the reduction was performed, over the fronthaul link to the fronthaul receiver which performs the steps of receiving the modified resource elements (50), receiving the compression metadata (52) and expanding the resource elements to increase the number of bits for representing the frequency-domain resource elements (54).