A PUCCH processing device, receiver, and base station
By adopting an architecture design that combines shared and dedicated processing units, the problem of unreusable PUCCH processing resources in 5G NR systems is solved, achieving efficient utilization of hardware resources and low power consumption, and improving the integration and scalability of base station equipment.
Patent Information
- Application Number
- CN202511214307.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-28
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-08-28
AI Technical Summary
In 5G NR systems, the traditional PUCCH processing architecture results in the inability to reuse hardware resources, leading to a decrease in resource utilization. Power consumption and chip area increase linearly with the type of format, becoming a bottleneck that restricts the evolution of base station equipment towards high integration and low power consumption.
The architecture adopts a shared processing unit and a dedicated processing unit. The shared processing unit performs common operations on frequency domain complex value data of various PUCCH formats within the same transmission time interval, while the dedicated processing unit performs differentiated operations, forming a three-stage pipeline architecture. The common operation unit is multiplexed according to the cycle, and the differentiated operation module is expanded as needed.
The total number of computing units, storage units, and control logic has been reduced, improving hardware resource requirements, increasing resource reuse, reducing processing latency and power consumption, and enhancing the scalability of the device.
Smart Images

Figure CN121012604B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of wireless communication technology, and more specifically, to a PUCCH processing device and a base station. Background Technology
[0002] In 5G NR systems, the Physical Uplink Control Channel (PUCCH) carries uplink control information such as ACK / NACK and CSI (Channel State Information), and defines multiple formats (e.g., Format0, Format1, etc.), each differing in time-domain symbol length, frequency-domain resource granularity, and modulation and coding scheme. Traditional base station implementations typically design the processing links for different formats completely independently, configuring a dedicated set of demapping, buffering, channel estimation, equalization, and demodulation modules for each PUCCH format to ensure the independence of each format in terms of functionality, timing, and performance metrics.
[0003] However, this architecture, which uses one processing link for each format, results in the inability to reuse hardware resources. When multiple PUCCH formats need to be supported simultaneously, the number of storage units, computing units, and control logic in the FPGA or ASIC must be increased exponentially, causing a sharp drop in resource utilization. Power consumption and chip area increase linearly with the number of formats, becoming the main bottleneck restricting the evolution of base station equipment towards high integration and low power consumption. Summary of the Invention
[0004] The purpose of this application is to provide a PUCCH processing device and a base station to solve the above-mentioned problems.
[0005] In a first aspect, embodiments of this application provide a PUCCH processing apparatus, including a shared processing unit and a dedicated processing unit connected to the shared processing unit, wherein: The shared processing unit is used to perform common operations on the frequency domain complex value data of multiple PUCCH formats output by Low-PHY within the same transmission time interval; The dedicated processing unit is used to perform differential operations on data corresponding to various PUCCH formats; wherein, the common operation and the differential operation are used to characterize the common operation part and the differential operation part when operating on the frequency domain complex value data of various PUCCH formats, respectively.
[0006] In the implementation of the above scheme, the shared processing unit performs common operations on the frequency domain complex-valued data of multiple PUCCH formats with unified hardware resources within the same transmission time interval. This merges the operation paths that originally needed to be configured repeatedly for each format into a single path, reducing the total number of operation units, storage units, and control logic, and helping to improve the hardware resource requirements of the above PUCCH processing device. On the other hand, the dedicated processing unit only performs local extensions for the differentiated operations that cannot be shared by each format. Its functional boundaries are clear, so that new formats can be implemented through pluggable dedicated sub-modules without modifying the shared parts, which is beneficial to improving the scalability of the above PUCCH processing device.
[0007] In one implementation of the first aspect, the shared processing unit includes a data extraction subunit, a polling scheduling subunit, and a baseband processing subunit connected in sequence, wherein: The data extraction sub-unit is used to extract the frequency domain complex value data of each PUCCH format from the frequency domain complex value data of the various PUCCH formats output by the Low-PHY, and to configure a format identifier for the frequency domain complex value data; The polling scheduling sub-unit is used to poll and schedule the frequency domain complex value data extracted by the data extraction sub-unit at the user level, and input the frequency domain complex value data into the baseband operation sub-unit; The baseband operation sub-unit is used to perform common operations on the frequency domain complex-valued data.
[0008] In the implementation of the above scheme, by dividing the shared processing unit into a data extraction sub-unit, a polling scheduling sub-unit, and a baseband operation sub-unit connected in sequence, the above PUCCH processing device logically forms a three-stage pipeline architecture. The interfaces of the three stages of data extraction, scheduling control, and common operation are clear, and each level of operation unit can be started sequentially according to the clock cycle, avoiding waiting across segments, thereby shortening the clock cycle requirement of the overall processing path. On the other hand, the polling scheduling sub-unit completes the scheduling at the user level, ensuring that the frequency domain complex value data of each user enters the baseband operation sub-unit completely within a single arbitration cycle, eliminating the context switching requirements caused by subcarrier-level or symbol-level splitting, thereby reducing the number of storage accesses and the complexity of control logic.
[0009] In one implementation of the first aspect, the data extraction subunit includes a demapping module and a cache module connected to the demapping module, wherein: The demapping module is used to extract the frequency domain complex value data of each PUCCH format from the frequency domain complex value data of multiple PUCCH formats based on the resource location information issued by the base station processor, and send the frequency domain complex value data and its format identifier to the cache module; The caching module is used to partition and store the frequency domain complex value data of each of the PUCCH formats based on the format identifier.
[0010] In the implementation of the above scheme, the demapping module accurately extracts frequency domain complex value data of different PUCCH formats from a unified frequency domain resource grid based on the resource location information issued by the base station processor, and marks the data ownership by format identifier, thereby realizing parallel extraction of multiple formats with a single entry point, which is beneficial to reducing the front-end hardware area of the above PUCCH processing device. On the other hand, the caching module uses format identifiers to partition and store the frequency domain complex value data, so that data of different PUCCH formats are kept address isolated in the physical buffer, avoiding read and write conflicts and data overwriting. The partitioned storage mechanism is beneficial to improving the utilization rate of storage resources and simplifying the subsequent scheduling logic.
[0011] In one implementation of the first aspect, the baseband operation subunit includes a low peak-to-average power ratio (PAPR) sequence generation module connected to the polling scheduling subunit, wherein: The polling scheduling sub-unit is also used to send a low peak-to-average ratio sequence generation request to the low peak-to-average ratio sequence generation module; and to perform a complex multiplication operation on the low peak-to-average ratio sequence returned by the low peak-to-average ratio sequence generation module and the frequency domain complex value data, and send the frequency domain complex value data after the complex multiplication process to the baseband operation sub-unit. The low peak-to-average ratio sequence generation module is used to generate and return a low peak-to-average ratio sequence in the corresponding format after receiving the low peak-to-average ratio sequence generation request sent by the polling scheduling sub-unit.
[0012] In the implementation of the above scheme, the low peak-to-average power ratio (PAPR) sequence generation module is directly coupled with the polling scheduling sub-unit, so that sequence request, generation and complex multiplication are all completed within the same scheduling cycle. This avoids the extra handshake cycle required by the independent sequence generation unit in the traditional scheme, which is beneficial to improving the overall processing latency of the above PUCCH processing device. On the other hand, the polling scheduling sub-unit triggers the generation of the PAPR sequence based on the current user and format information, and performs complex multiplication processing on the returned sequence and the frequency domain complex value data. This allows the data of each user to complete the sequence weighting before entering the baseband operation sub-unit, reducing the power fluctuations introduced by sequence buffering or delay, which is beneficial to improving the consistency of the uplink signal peak-to-average power characteristics.
[0013] In one implementation of the first aspect, the baseband arithmetic subunit further includes a channel estimation preprocessing module, a peak search module, an equalization module, and a soft demodulation module connected in sequence, wherein: The channel estimation preprocessing module is used to perform a fast inverse Fourier transform on the frequency domain complex value data to obtain time domain channel response data, and send the time domain channel response data to the peak search module. The peak search module is used to determine the peak position from the time-domain channel response data and send the peak position to the equalization module; The equalization module is used to perform channel equalization processing on the data input to the equalization module based on the peak position, obtain equalized symbol data, and send the symbol data to the soft demodulation module; The soft demodulation module is used to perform soft demodulation processing on the symbol data to obtain log-likelihood ratio data.
[0014] In the implementation of the above scheme, the channel estimation preprocessing module performs an inverse fast Fourier transform on the frequency domain complex-valued data, directly mapping the pilot subcarriers to the time domain channel impulse response. This allows peak search to locate the main path in the discrete time domain samples without interpolation or filtering in the frequency domain, thereby reducing the number of complex multiplications and accumulations and shortening the critical path delay of peak detection. On the other hand, the equalization module selects equalization coefficients based on the peak position and performs channel equalization on the input data. The soft demodulation module completes soft decision-making in the same pipeline stage. The channel estimation preprocessing module, peak search module, equalization module, and soft demodulation module form a fixed-depth processing chain. This processing chain allows data of different PUCCH formats to proceed at the same pace, avoiding cross-stage waiting or write-back, which helps to reduce the processing delay of the above PUCCH processing device.
[0015] In one implementation of the first aspect, the polling scheduling sub-unit is further configured to generate an accompanying identifier for the frequency domain complex-valued data, and input the frequency domain complex-valued data and its corresponding accompanying identifier into the baseband operation sub-unit; wherein, the accompanying identifier is used to indicate the PUCCH format to which the frequency domain complex-valued data belongs and the corresponding user identifier.
[0016] In the implementation of the above scheme, the path identifier is generated before the data enters the baseband operation sub-unit and is transmitted synchronously with the frequency domain complex-valued data. This allows the channel estimation preprocessing module, peak search module, equalization module, and soft demodulation module to determine their respective operation parameters based on the identifier content, without waiting for external configuration updates. This eliminates the idle period caused by parameter distribution and helps improve the overall processing latency of the PUCCH processing device. On the other hand, the path identifier carries both the PUCCH format and the user identifier, enabling each level of operation unit within the baseband operation sub-unit to distinguish and process data from different users and in different formats on a single data path. This avoids maintaining a separate control state machine for each format or user, which helps improve the scale and complexity of the control logic. Furthermore, the path identifier and the frequency domain complex-valued data can be cascaded and transmitted using the same data bus field and synchronously latched at each stage of the pipeline. This ensures accurate data attribution recovery at any depth. This synchronization mechanism prevents operational errors caused by out-of-order delivery or buffer reordering and eliminates the need for additional registers or handshake signals used for cross-stage synchronization in traditional schemes, thereby further reducing hardware overhead while ensuring functional correctness.
[0017] In one implementation of the first aspect, the channel estimation preprocessing module is used to determine the transform coefficients based on the accompanying identifier, and to perform a fast inverse Fourier transform on the frequency domain complex value data using the transform coefficients to obtain time domain channel response data. The peak search module is used to determine a decision threshold and a search window length based on the accompanying identifier, and to determine the peak position in the time-domain channel response data using the decision threshold and the search window length. The equalization module is used to determine the equalization coefficient based on the associated identifier, and to perform channel equalization processing on the data input to the equalization module using the equalization coefficient and the peak position to obtain equalized symbol data. The soft demodulation module is used to determine the quantization accuracy based on the in-path identifier, and to perform soft demodulation processing on the symbol data using the quantization accuracy to obtain the log-likelihood ratio data.
[0018] In the implementation of the above scheme, the channel estimation preprocessing module, peak search module, equalization module, and soft demodulation module all dynamically select transform coefficients, decision thresholds, equalization coefficients, and quantization precision using the in-path identifier, without the need for independent configuration of the bus or register writing. This eliminates the additional delay cycle required for format switching and ensures a constant pipeline cycle time. On the other hand, each module directly calls the corresponding parameter set based on the in-path identifier, realizing the on-demand loading of operation coefficients, threshold values, and quantization step sizes. This allows the same set of multipliers, adders, and lookup table resources to be seamlessly shared between different PUCCH formats, reducing the redundant deployment of storage units and control logic. Furthermore, the parameter selection mechanism driven by the in-path identifier limits format differences to the parameter level. When adding a new format, only the parameters of the corresponding module need to be extended, without modifying the operation unit structure or control flow of the shared processing unit, which is beneficial to improving the scalability of the above PUCCH processing device.
[0019] In one implementation of the first aspect, the dedicated processing unit includes: The first dedicated processing sub-unit is used to perform differential operations on the data corresponding to format 0; The first dedicated processing subunit includes an output module connected to the soft demodulation module, wherein: The soft demodulation module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is format 0, send the log-likelihood ratio data to the output module; The output module is used to send the log-likelihood ratio data output by the soft demodulation module to the MAC layer interface.
[0020] In the implementation of the above scheme, the soft demodulation module determines that the current data is format 0 based on the path identifier before outputting, and directly sends the obtained log-likelihood ratio data to the MAC layer interface through the output module of the first dedicated processing sub-unit, which simplifies the data path and helps to improve the processing delay of the above PUCCH processing device.
[0021] In one implementation of the first aspect, the dedicated processing unit includes: The second dedicated processing sub-unit is used to perform differential operations on the data corresponding to format 1. The second dedicated processing subunit includes a Fast Fourier Transform (FFT) module, a detransform precoding module, and a descrambling and decoding module. The input of the FFT module is connected to the output of the peak search module, the output of the FFT module is connected to the input of the equalization module, the input of the detransform precoding module is connected to the output of the equalization module, the output of the detransform precoding module is connected to the input of the soft demodulation module, and the output of the soft demodulation module is connected to the descrambling and decoding module. Wherein: The peak search module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is format 1, to send the time-domain channel response data to the fast Fourier transform module. The Fast Fourier Transform module is used to convert the time-domain channel response data into frequency-domain complex-valued data, and send the frequency-domain complex-valued data to the equalization module. The equalization module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is the format 1, send the symbol data to the detransformation precoding module; The detransform precoding module is used to perform detransform precoding processing on the symbol data to obtain time-domain symbol data, and send the time-domain symbol data to the soft demodulation module; The soft demodulation module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is the format 1, send the log-likelihood ratio data to the descrambling and decoding module; The descrambling and decoding module is used to descramble and decode the log-likelihood ratio data, and send the log-likelihood ratio data after the descrambling and decoding processes to the MAC layer interface.
[0022] In the implementation of the above scheme, when the accompanying identifier indicates Format1, the Fast Fourier Transform (FFT) module, the detransform precoding module, and the descrambling and decoding module are activated sequentially. Each module completes the frequency domain-to-time conversion, detransform precoding, and descrambling and decoding on the data path. Since each module is only activated when the identifier is triggered and remains bypassed at other times, the continuous occupation of the shared processing unit's computing resources is avoided, thereby reducing ineffective power consumption. On the other hand, the FFT module, the detransform precoding module, and the descrambling and decoding module process data sequentially through the gating logic driven by the format identifier, without the need for additional arbitration or buffering. This allows Format1 data to complete differentiated operations within a single transmission time interval, meeting the protocol timing requirements and improving the processing latency of the above-mentioned PUCCH processing device. Furthermore, the interfaces of each module are consistent with the bus protocol of the shared processing unit, and the processing of Format1 format data can be achieved without modifying the architecture of the shared processing unit, which helps to reduce hardware expansion costs.
[0023] Secondly, embodiments of this application provide a receiver, including the PUCCH processing apparatus provided in the first aspect or any possible implementation of the first aspect.
[0024] Thirdly, embodiments of this application provide a base station, including a PUCCH processing device provided by the first aspect or any possible implementation of the first aspect, or a receiver provided by the second aspect or any possible implementation of the second aspect.
[0025] Fourthly, embodiments of this application provide a PUCCH processing method, the method comprising: Within the same transmission time interval, common operations and differential operations are performed on the frequency domain complex-valued data of various PUCCH formats output by Low-PHY to obtain the log-likelihood ratio data of the frequency domain complex-valued data; wherein, the common operations and the differential operations are used to characterize the common operation part and the differential operation part when performing operations on the frequency domain complex-valued data of various PUCCH formats, respectively. The log-likelihood ratio data is sent to the MAC layer interface.
[0026] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description
[0027] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0028] Figure 1 This is a schematic diagram of the structure of the PUCCH processing device provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the shared processing unit provided in an embodiment of this application; Figure 3 A schematic diagram of a PUCCH processing device that supports both Format0 and Format1 PUCCH formats in an application scenario provided in this application embodiment; Figure 4 This is a flowchart illustrating the PUCCH processing method provided in an embodiment of this application.
[0029] The numbers in the diagram are as follows: 100. PUCCH processing unit; 110. Shared processing unit; 120. Dedicated processing unit; 111. Data extraction subunit; 112. Polling scheduling subunit; 113. Baseband operation subunit; 1111. Demapping module; 1112. Buffer module; 1131. Low peak-to-average ratio sequence generation module; 1132. Channel estimation preprocessing module; 1133. Peak search module; 1134. Equalization module; 1135. Soft demodulation module; 121. First dedicated processing subunit; 1211. Output module; 122. Second dedicated processing subunit; 1221. Fast Fourier Transform module; 1222. Detransform precoding module; 1223. Descrambling and decoding module. Detailed Implementation
[0030] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of this application, and are therefore merely examples and should not be used to limit the scope of protection of this application.
[0031] In 5G NR base stations, PUCCH is used to transmit uplink control information. PUCCH supports multiple formats, such as: Format0: A short format based on time-domain symbols, used to transmit 1-2 bits of information; Format1: A long format based on frequency domain sequences, used to transmit multi-bit information.
[0032] Traditional methods for handling various PUCCH formats employ independent modular architectures. This design philosophy stems from an overemphasis on the functional differences between each PUCCH format, resulting in inefficient reuse of hardware resources. Each module requires independent hardware resources, creating two parallel processing chains. When supporting multiple PUCCH formats simultaneously, the number of storage units, processing units, and control logic needs to be multiplied. Furthermore, the scheduling module also needs to maintain multiple sets of queue management logic simultaneously, leading to stricter timing constraints and a higher risk of metastability.
[0033] In view of this, this application provides a PUCCH processing device. The shared processing unit in this device performs common operations on frequency domain complex-valued data of multiple PUCCH formats using unified hardware resources within the same transmission time interval. This merges the operation paths that originally needed to be configured repeatedly for each format into a single path, reducing the total number of operation units, storage units, and control logic, and helping to improve the hardware resource requirements of the aforementioned PUCCH processing device. On the other hand, the dedicated processing unit only performs local extensions for the differentiated operations that cannot be shared by each format. Its functional boundaries are clear, so that new formats can be implemented through pluggable dedicated sub-modules without modifying the shared parts, which is beneficial to improving the scalability of the aforementioned PUCCH processing device.
[0034] The PUCCH processing device provided in this application can be applied to the base station side of a communication system. Before describing the technical solution of this application, the communication system described above will be described first.
[0035] The aforementioned communication system may include access devices and terminals. In the specific implementation of this application's embodiments, the terminal may be a computer, smartphone, telephone, cable TV set-top box, digital subscriber line router, or other similar devices. The access device may be one of a terrestrial base station, a high-altitude base station, a low-Earth orbit satellite, a medium-Earth orbit satellite, or a high-Earth orbit satellite. It should be noted that in practical applications, the number of access devices and terminals may be one or more, and this application does not limit this.
[0036] The aforementioned communication systems can be applied to Long Term Evolution (LTE) systems or NR systems (also known as 5th Generation (5G) systems), LTE and NR hybrid networking systems, Vehicle to Everything (V2X) systems, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) communication systems, Internet of Things (IoT) systems (such as Narrow Band Internet of Things (NB-IoT) systems), 6G systems and other systems evolving after 5G, as well as other next-generation communication systems. Alternatively, the communication system can also be an Open Radio Access Network (O-RAN or ORAN), a Cloud Radio Access Network (CRAN), or a Wireless Fidelity (Wi-Fi) system, without limitation.
[0037] Furthermore, the access equipment can be used to support terminal access. For example, it can be ground-based equipment such as Base Transceiver Stations (BTS) and Base Station Controllers (BSCs) in 2G access technology communication systems, Node Bs (RNCs) and Radio Network Controllers (RNCs) in 3G access technology communication systems, Evolved Node Bs (eNBs) in 4G access technology communication systems, Next Generation Node Bs (gNBs), Transmission Reception Points (TRPs), Relay Nodes, and Access Points (APs) in 5G access technology communication systems. It can also be non-ground-based equipment: high-altitude base stations, such as hot air balloons that provide wireless access functionality to terminals, low-Earth orbit satellites, medium-Earth orbit satellites, high-Earth orbit satellites, etc. For ease of description, in all embodiments of this application, the device providing wireless communication functionality to the terminal is collectively referred to as a base station.
[0038] A terminal can be a device that provides voice or data connectivity to a user, and may also be referred to as a mobile station, subscriber unit, station, or terminal equipment (TE). Terminals can be cellular phones, personal digital assistants (PDAs), modems, handheld devices, laptop computers, cordless phones, wireless local loop (WLL) stations, tablets, etc. With the development of wireless communication technology, any device that can access a communication system, communicate with the network side of the communication system, or communicate with other objects through the communication system can be a terminal in the embodiments of this application. Examples include terminals and vehicles in intelligent transportation systems, home appliances in smart homes, electricity meter reading instruments, voltage monitoring instruments, environmental monitoring instruments in smart grids, video surveillance instruments in smart security networks, cash registers, etc. In the embodiments of this application, the terminal can communicate with a base station. Multiple terminals can also communicate with each other. The terminal can be static or mobile.
[0039] The communication architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of communication architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.
[0040] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.
[0041] Please see Figure 1 This application provides a PUCCH processing device 100, which includes a shared processing unit 110 and a dedicated processing unit 120 connected to the shared processing unit 110, wherein: The shared processing unit 110 is used to perform common operations on frequency domain complex value data of multiple PUCCH formats output by Low-PHY within the same transmission time interval; The dedicated processing unit 120 is used to perform differential operations on data corresponding to multiple PUCCH formats; wherein, the common operation and the differential operation are used to characterize the common operation part and the differential operation part when performing operations on the frequency domain complex value data of multiple PUCCH formats, respectively.
[0042] The aforementioned Low-PHY refers to the logical entity responsible for low-level physical layer processing in the base station physical layer protocol stack, located between the radio frequency front-end and the higher-level physical layer High-PHY. Typical functions of Low-PHY include: (1) performing fast inverse Fourier transform (IFFT), cyclic prefix (CP) insertion, digital upconversion, and pre-correction on the downlink; (2) performing fast Fourier transform (FFT), cyclic prefix (CP) removal, digital downconversion, and power normalization on the uplink; and (3) performing resource unit (RE) level mapping and demapping, and exchanging IQ sampling streams with the radio frequency unit through a high-speed interface.
[0043] The frequency domain complex value data output by the Low-PHY is the complex value symbol stream arranged by subcarrier after the FFT demodulation, which is used by the PUCCH processing device 100 to further extract and decode control information.
[0044] The aforementioned Transmission Time Interval (TTI) refers to the smallest time unit that the physical layer divides for a complete uplink or downlink transmission. The MAC layer scheduling entity typically issues uplink grants with the TTI as the smallest time granularity, expecting to obtain the complete HARQ-ACK, CSI, or SR bitstream before the end of the TTI. If common and differential operations are performed within the same transmission time interval, the MAC layer can obtain the complete HARQ-ACK, CSI, or SR bitstream at the end of the current TTI, thereby completing closed-loop retransmission judgment, link adaptive update, and uplink scheduling decision within the same uplink grant cycle. This synchronization mechanism maximizes time-domain resource utilization with minimal delay between the physical and MAC layer control cycles, and prevents subsequent symbols from remaining idle while waiting for control information, improving overall processing efficiency.
[0045] The aforementioned common operations refer to the universal baseband signal processing steps that must be executed for all PUCCH formats, such as resource demapping and channel estimation preprocessing. These common operations are uniformly implemented in hardware by the shared processing unit 110, with computational resources reused on a per-cycle basis, avoiding repeated instantiation due to format differences. The differentiated operations refer to additional processing required only under specific PUCCH formats, such as detransformation precoding when processing Format1 data. These differentiated operations are implemented by a format-specific subunit within the dedicated processing unit 120, activated only when the corresponding format is triggered, and otherwise bypassed. This ensures that hardware expansion only involves local module additions or deletions, without affecting the shared operations.
[0046] The following describes the structure of the shared processing unit 110: Please see Figure 2 Optionally, the shared processing unit 110 includes a data extraction subunit 111, a polling scheduling subunit 112, and a baseband processing subunit 113 connected in sequence, wherein: The data extraction sub-unit 111 is used to extract the frequency domain complex value data of each PUCCH format from the frequency domain complex value data of multiple PUCCH formats output by Low-PHY. The polling scheduling sub-unit 112 is used to poll and schedule the frequency domain complex value data extracted by the data extraction sub-unit 111 at the user level, and input the frequency domain complex value data into the baseband operation sub-unit 113; The baseband operation sub-unit 113 is used to perform common operations on frequency domain complex-valued data.
[0047] The data extraction sub-unit 111, polling and scheduling sub-unit 112, and baseband operation sub-unit 113 are cascaded to form a three-stage pipeline architecture. The data extraction sub-unit 111 demaps and buffers complex-valued data of various PUCCH formats from the frequency domain resource grid output by the Low-PHY. The polling and scheduling sub-unit 112 uses users as the granularity for polling and arbitration, pushing complete data blocks to the baseband operation sub-unit 113. The baseband operation sub-unit 113 performs common operations such as equalization and soft demodulation according to the clock cycle and outputs the results within the current TTI. It is understandable that traditional solutions use a parallel structure where one format corresponds to one link. Each PUCCH format is independently configured with complete demapping, buffering, channel estimation, equalization, and soft demodulation modules, with no data interaction or resource sharing between them. This results in a linear increase in hardware units with the number of formats, a simultaneous increase in chip area and static power consumption, and a multiplied increase in the number of schedulers and state machines. The aforementioned three-stage pipeline architecture, through a cascaded design of data extraction, polling scheduling, and common operations, merges the common steps originally scattered across various links into a single hardware resource, which is then time-division multiplexed according to user granularity. This improves the resource reuse rate of the aforementioned PUCCH processing device 100, and the total number of multipliers, memory, and control logic no longer increases linearly with the number of formats.
[0048] The polling scheduling sub-unit 112 described above performs polling scheduling at the user granularity. This means that within each scheduling cycle, the polling scheduling sub-unit 112 submits all frequency domain complex-valued data belonging to the same user to the baseband processing sub-unit 113 at once, processes it, and then moves on to the next user. Within a single user, data is no longer split into subcarriers or symbol-level sub-blocks. Each user's pilot, UCI, and control metadata are treated as indivisible data blocks at the physical layer. Processing them all at once avoids saving and restoring context across granularities, reducing buffering and control signals. Furthermore, user-level polling only requires maintaining a single user counter and format identifier, reducing hardware complexity and processing latency.
[0049] The above scheme divides the shared processing unit 110 into a data extraction subunit 111, a polling scheduling subunit 112, and a baseband operation subunit 113 connected in sequence. This enables the PUCCH processing device 100 to form a logical three-stage pipeline data architecture. The interfaces of the three stages of data extraction, scheduling control, and common operation are clear, and each level of operation unit can be started sequentially according to the clock cycle, avoiding waiting across stages, thereby shortening the clock cycle requirement of the overall processing path. On the other hand, the polling scheduling subunit 112 completes the scheduling at the user level, ensuring that the frequency domain complex value data of each user enters the baseband operation subunit 113 completely within a single arbitration cycle. This eliminates the context switching requirements caused by subcarrier-level or symbol-level splitting, thereby reducing the number of storage accesses and the complexity of control logic.
[0050] The following is a description of the data extraction sub-section 111 mentioned above: Optionally, the data extraction sub-unit 111 mentioned above includes a demapping module 1111 and a cache module 1112 connected to the demapping module 1111, wherein: The demapping module 1111 is used to extract frequency domain complex value data of each PUCCH format from the frequency domain complex value data of multiple PUCCH formats based on the resource location information issued by the base station processor, and send the frequency domain complex value data and its format identifier to the cache module 1112. The cache module 1112 is used to store frequency domain complex value data of each PUCCH format in partitions based on the format identifier.
[0051] The aforementioned resource location information may include time-domain location and frequency-domain location, wherein: Time domain location: Provides the starting symbol index and the number of consecutive symbols occupied by the current PUCCH, used to accurately locate the time domain window of the uplink control channel within the time slot; Frequency domain location: Provides the starting physical resource block index and the number of resource blocks occupied, used to mark the frequency domain range of the uplink control channel in the frequency domain resource grid.
[0052] By uniquely determining the specific region of the PUCCH resource in the time-frequency two-dimensional resource grid through the joint time-domain and frequency-domain positions, the demapping module 1111 extracts the corresponding subcarrier set from the overall frequency-domain complex-valued data output by the Low-PHY, thereby completing the accurate extraction of the corresponding PUCCH format data.
[0053] When extracting frequency domain complex-valued data, the demapping module 1111 directly obtains the format identifier by parsing the format indication field sent by the base station processor along with the resource location information. The format indication field is located in the configuration register or control packet header, and its length can be 1–2 bits, with encoding methods such as: 00: Format0; 01: Format1, etc. The demapping module 1111 reads the format indication field to obtain the format identifier during the register write cycle.
[0054] The aforementioned cache module 1112 can receive the format identifier field and frequency domain complex-valued data in parallel at the write port. Several bits of the format identifier are directly fed into the address decoding logic, and the decoding result generates a partition pointer: if the identifier indicates Format0, the pointer points to the lower address partition; if the identifier indicates Format1, the pointer points to the higher address partition. Within the same clock cycle, the pointer is added to the write address to complete partition addressing. The frequency domain complex-valued data is written to the corresponding partition, and the read end also selects the partition address according to the format identifier, achieving conflict-free and aliasing-free partitioned storage.
[0055] The demapping module 1111 in the above scheme accurately extracts frequency domain complex value data of different PUCCH formats from a unified frequency domain resource grid based on the resource location information issued by the base station processor, and marks the data ownership by format identifier, thereby realizing parallel extraction of multiple formats with a single entry point, which is beneficial to reducing the front-end hardware area of the above PUCCH processing device 100. On the other hand, the caching module 1112 uses the format identifier to partition and store the frequency domain complex value data, so that the data of different PUCCH formats are kept at address isolation in the physical buffer, avoiding read and write conflicts and data overwriting. The partitioned storage mechanism is beneficial to improving the utilization rate of storage resources and simplifying the subsequent scheduling logic.
[0056] Optionally, the baseband operation sub-unit 113 mentioned above includes a low peak-to-average ratio sequence generation module 1131 connected to the polling scheduling sub-unit 112, wherein: The polling scheduling sub-unit 112 is also used to send a low peak-to-average ratio sequence generation request to the low peak-to-average ratio sequence generation module 1131; and to perform a complex multiplication operation on the low peak-to-average ratio sequence returned by the low peak-to-average ratio sequence generation module 1131 and the frequency domain complex value data, and to send the frequency domain complex value data after the complex multiplication process to the baseband operation sub-unit 113. The low peak-to-average ratio sequence generation module 1131 is used to generate and return a low peak-to-average ratio sequence in the corresponding format after receiving a low peak-to-average ratio sequence generation request sent by the polling scheduling sub-unit 112.
[0057] The aforementioned low peak-to-average power ratio (PAPR) sequence refers to a complex-valued sequence with a low PAPR, used at the transmitting end to scramble or spread uplink control information symbols to reduce instantaneous peak power, improve power amplifier efficiency, and reduce nonlinear distortion. In 5G NR, PUCCH Format 0 and Format 1 use ZC (Zadoff-Chu) sequences or orthogonal cover code (OCC) sequences of different lengths and root sequences as low PAPR sequences, with their specific parameters determined by higher-layer configuration.
[0058] The aforementioned low peak-to-average ratio (PAPR) sequence generation module 1131 is a configurable sequence generator that internally stores sequence generation parameters corresponding to various PUCCH formats, such as root index, cyclic shift, and sequence length. When the polling scheduling sub-unit 112 indicates the current user and format via a request signal, the module immediately looks up the table or performs real-time calculations according to the corresponding indication to generate a complex value sequence of the required length, and returns it within the same clock cycle for subsequent complex multiplication operations.
[0059] It is understandable that the above scheme advances the processing of sequence weighting, which is format-related but has a relatively low computational load, to be performed after the data leaves the buffer. The purpose is to: (1) reduce the number of input ports of the downstream baseband operation sub-unit 113: the low peak-to-average ratio sequence frequency domain complex value data has been fused in the scheduling stage, and the subsequent modules only need to face a single data stream; (2) avoid reserving independent sequence multipliers for each format inside the baseband operation sub-unit 113, and further compress the hardware scale; (3) take advantage of the natural alignment of the polling clock cycle to complete the pairing of user-level data and corresponding sequences in the same clock cycle, eliminate cross-level synchronization waiting, and ensure that the pipeline clock cycle is constant.
[0060] In the above scheme, the low peak-to-average power ratio (PAPR) sequence generation module 1131 is directly coupled with the polling scheduling sub-unit 112, so that sequence request, generation and complex multiplication are all completed within the same scheduling cycle. This avoids the extra handshake cycle required by the independent sequence generation unit in the traditional scheme, which is beneficial to improving the overall processing delay of the above PUCCH processing device 100. On the other hand, the polling scheduling sub-unit 112 triggers the generation of the PAPR sequence according to the current user and format information, and performs complex multiplication processing on the returned sequence and the frequency domain complex value data, so that the data of each user is weighted before entering the baseband operation sub-unit 113, reducing the power fluctuations introduced by sequence buffering or delay, which is beneficial to improving the consistency of the uplink signal peak-to-average power characteristics.
[0061] The remaining parts of the baseband operation sub-section 113 are described below: Optionally, the baseband operation sub-unit 113 further includes a channel estimation preprocessing module 1132, a peak search module 1133, an equalization module 1134, and a soft demodulation module 1135 connected in sequence, wherein: The channel estimation preprocessing module 1132 is used to perform a fast inverse Fourier transform on the frequency domain complex value data to obtain the time domain channel response data, and send the time domain channel response data to the peak search module 1133. The peak search module 1133 is used to determine the peak position in the time domain channel response data and send the peak position to the equalization module 1134; Equalization module 1134 is used to perform channel equalization processing on the data input to equalization module 1134 based on the peak position, obtain equalized symbol data, and send the symbol data to soft demodulation module 1135; The soft demodulation module 1135 is used to perform soft demodulation processing on symbol data to obtain log-likelihood ratio data.
[0062] After receiving the frequency domain complex-valued data, the channel estimation preprocessing module 1132 performs a fast Fourier transform on the pilot subcarriers to obtain time-domain channel response data. This data, in the form of discrete sampling, characterizes the amplitude-delay distribution of the wireless channel in the time domain, providing precise location and intensity information for the main path and multipath components. The peak search module 1133 selects the peak position accordingly, which is then used by the equalization module 1134 to generate equalization coefficients, thereby achieving channel distortion compensation for the data symbols.
[0063] The peak search module 1133 described above can use a sliding energy window mechanism to determine the peak position in the time-domain channel response data. The main steps include: first, in the time-domain channel response data, a rectangular window with a fixed step size and a length of N samples is moved, and the energy sum of the complex-valued samples within the window is calculated; then, the energy sum is compared with a preset threshold. If the current window energy is higher than the threshold and is a local maximum, the center index of the window is recorded as the peak position; finally, only this single index value is output to the equalization module 1134 for indexing the equalization coefficient table.
[0064] After receiving the peak position, the equalization module 1134 can use the position as an index to read the corresponding complex weighted value from the internal equalization coefficient table. Then, it performs complex multiplication on the data input to the equalization module 1134 subcarrier to compensate for the amplitude fading and phase rotation introduced by the channel, and outputs the equalized symbol data.
[0065] The aforementioned Log-Likelihood Ratio (LLR) is a commonly used soft demodulation method, defined as the natural logarithm of the ratio of the probability of a bit being 0 to the probability of it being 1 under the condition of receiving a symbol. After receiving the equalized symbol data, the aforementioned soft demodulation module 1135 can map the real and imaginary parts of each symbol to the corresponding coordinates of the constellation diagram, calculate the likelihood based on the Euclidean distance between the symbol and the ideal constellation point, and quantize the log-likelihood ratio into an LLR value with a fixed quantization bit width. The obtained LLR sequence is the log-likelihood ratio data.
[0066] In the above scheme, the channel estimation preprocessing module 1132 performs an inverse fast Fourier transform on the frequency domain complex-valued data, directly mapping the pilot subcarrier to the time domain channel impulse response. This allows peak search to locate the main path in the discrete time domain samples without interpolation or filtering in the frequency domain, thereby reducing the number of complex multiplications and accumulations and shortening the critical path delay of peak detection. On the other hand, the equalization module 1134 selects equalization coefficients based on the peak position and performs channel equalization on the input data. The soft demodulation module 1135 completes soft decision-making in the same pipeline stage. The channel estimation preprocessing module 1132, peak search module 1133, equalization module 1134, and soft demodulation module 1135 form a fixed-depth processing chain. This processing chain allows data of different PUCCH formats to proceed at the same pace, avoiding cross-stage waiting or write-back, which helps to reduce the processing delay of the above PUCCH processing device 100.
[0067] Optionally, the polling scheduling sub-unit 112 is also used to generate the associated identifier of the frequency domain complex value data and input the frequency domain complex value data and its corresponding associated identifier into the baseband operation sub-unit 113; wherein, the associated identifier is used to indicate the PUCCH format to which the frequency domain complex value data belongs and the corresponding user identifier.
[0068] When the polling scheduling sub-unit 112 reads frequency domain complex-valued data from the cache module 1112, it can simultaneously read the format identifier corresponding to the frequency domain complex-valued data. After obtaining the format identifier, it can concatenate the format identifier with the user index in the current scheduling entry to form an accompanying identifier. In scenarios that support Format0 and Format1, the accompanying identifier takes the following form: in the accompanying identifier, the format identifier occupies 1 bit (0=Format0, 1=Format1), and the user index occupies 7 bits (supporting 128 users), which are combined into an 8-bit field, which is the accompanying identifier. The accompanying identifier and the frequency domain complex-valued data are latched on the same clock edge and cascaded with the data stream to the baseband operation sub-unit 113 for subsequent modules to parse the format and user affiliation in real time.
[0069] In the above scheme, the in-path identifier is generated before the data enters the baseband operation sub-unit 113 and is transmitted synchronously with the frequency domain complex-valued data. This allows the channel estimation preprocessing module 1132, peak search module 1133, equalization module 1134, and soft demodulation module 1135 to determine their respective operation parameters based on the identifier content, without waiting for external configuration updates. This eliminates the idle period caused by parameter distribution and helps improve the overall processing latency of the PUCCH processing device 100. On the other hand, the in-path identifier carries both the PUCCH format and the user identifier, enabling the operation of each level of the operation unit within the baseband operation sub-unit 113 to... Distinguishing and processing data from different users and in different formats on a single data path avoids the need to maintain a separate control state machine for each format or user, which helps to improve the scale and complexity of the control logic. On the other hand, the accompanying identifier and frequency domain complex value data can be cascaded and transmitted using the same data bus field, and are synchronously latched at each stage of the pipeline to ensure that the data ownership can be accurately restored at any depth. This synchronization mechanism can prevent calculation errors caused by out-of-order or buffer reordering, and at the same time eliminates the need for additional registers or handshake signals used for cross-stage synchronization in traditional solutions, thereby further reducing hardware overhead while ensuring functional correctness.
[0070] Optionally, the channel estimation preprocessing module 1132 is used to determine the transform coefficients based on the path identifier, and to perform a fast inverse Fourier transform on the frequency domain complex value data using the transform coefficients to obtain the time domain channel response data. The peak search module 1133 is used to determine the decision threshold and search window length based on the in-path identifier, and to determine the peak position in the time-domain channel response data using the decision threshold and search window length. Equalization module 1134 is used to determine the equalization coefficient based on the in-path identifier, and to perform channel equalization processing on the data input to equalization module 1134 using the equalization coefficient and peak position to obtain equalized symbol data. The soft demodulation module 1135 is used to determine the quantization accuracy based on the in-path identifier and to perform soft demodulation processing on the symbol data using the quantization accuracy to obtain the log-likelihood ratio data.
[0071] The aforementioned channel estimation preprocessing module 1132 synchronously parses the accompanying identifier when receiving data packets. If the accompanying identifier indicates that the PUCCH format is Format0, the channel estimation preprocessing module 1132 selects 128 IFFT coefficients from the local coefficient table; if the accompanying identifier indicates that the PUCCH format is Format1, the channel estimation preprocessing module 1132 selects 256 IFFT coefficients from the local coefficient table. After selecting the coefficients, the channel estimation preprocessing module 1132 performs an inverse fast Fourier transform on the frequency domain complex-valued data according to the corresponding length, and outputs discrete-time domain channel response data for use by subsequent modules.
[0072] After receiving the accompanying identifier, the peak search module 1133 can complete the peak location according to the following steps: (1) Parse the format field in the accompanying identifier: if the accompanying identifier indicates that the PUCCH format is Format0, then read the threshold T0 and the window length W0 from the local register; if the accompanying identifier indicates that the PUCCH format is Format1, then read the threshold T1 and the window length W1. (2) In the time domain channel response data, slide with the selected window length W as the width and the step size as 1 sample, and calculate the sum of energy of the complex-valued samples in the window. (3) Compare the sum of energy with the corresponding threshold T. If the current window energy is higher than the threshold and is a local maximum, then record the index of the center sample of the window as the peak position and output the index value to the equalization module 1134.
[0073] Upon receiving the accompanying identifier, the equalization module 1134 first parses the format field and user index, and locates the corresponding entry in the internal equalization coefficient table. Then, using the peak position as the address offset, it reads the channel gain estimate at that position as the equalization coefficient. It performs a complex multiplication operation on the input symbol data subcarrier by subcarrier to counteract the amplitude attenuation and phase rotation introduced by the channel, and the output is the equalized symbol data.
[0074] The aforementioned soft demodulation module 1135 parses the accompanying identifier while receiving symbol data and selects the corresponding quantization precision based on the format field therein. A narrow quantization bit width is used when the accompanying identifier indicates the PUCCH format is Format0, and a wide quantization bit width is used when the accompanying identifier indicates the PUCCH format is Format1. The soft demodulation module 1135 maps the symbols to a constellation diagram, calculates the Euclidean distance to the ideal constellation points, and linearly maps the distance values to the log-likelihood ratio according to the selected quantization precision, outputting an LLR sequence.
[0075] In the above scheme, the channel estimation preprocessing module 1132, peak search module 1133, equalization module 1134, and soft demodulation module 1135 all dynamically select transform coefficients, decision thresholds, equalization coefficients, and quantization precision using in-path identifiers. This eliminates the need for independent configuration of the bus or register writing, thereby eliminating the additional delay cycle required for format switching and ensuring a constant pipeline cycle time. On the other hand, each module directly calls the corresponding parameter set based on the in-path identifier, realizing the on-demand loading of operation coefficients, threshold values, and quantization step sizes. This allows the same set of multipliers, adders, and lookup table resources to be seamlessly shared between different PUCCH formats, reducing the redundant deployment of storage units and control logic. Furthermore, the parameter selection mechanism driven by the in-path identifier limits format differences to the parameter level. When adding a new format, only the parameters of the corresponding module need to be extended, without modifying the operation unit structure or control flow of the shared processing unit 110, which is beneficial to improving the scalability of the above-mentioned PUCCH processing device 100.
[0076] The following is a description of the structure of the dedicated processing unit 120: Optionally, the dedicated processing unit 120 includes: The first dedicated processing sub-unit 121 is used to perform differential operations on the data corresponding to format 0; The first dedicated processing sub-unit 121 includes an output module 1211 connected to the soft demodulation module 1135, wherein: The soft demodulation module 1135 is also used to determine the PUCCH format based on the associated identifier; and, when the PUCCH format is format 0, to send the log-likelihood ratio data to the output module 1211. Output module 1211 is used to send the log-likelihood ratio data output by soft demodulation module 1135 to the MAC layer interface.
[0077] The aforementioned output module 1211 can adopt a single-port FIFO (First Input First Output) interface, with a bit width consistent with the log-likelihood ratio bit width output by the soft demodulation module 1135. When the accompanying identifier indicates format 0, the soft demodulation module 1135 writes LLR data into the FIFO when the valid signal is pulled high. The output module 1211 pushes the data to the MAC layer interface in a fixed-cycle, first-in-first-out order, completing the delivery of format 0 bit-level information.
[0078] It is understandable that the first dedicated processing subunit 121 may only include the output module 1211, because the complete bit-level information of Format 0 has already been obtained through common operations in the baseband operation subunit 113 of the shared processing unit 110. Specifically, the log-likelihood ratio is directly generated after equalization and soft demodulation of the frequency domain complex-valued data, and no further additional operations such as FFT, IDFT, or decoding are required. Therefore, the differentiated processing of Format 0 only requires one path to forward the log-likelihood ratio to the MAC layer interface, and the output module 1211 is sufficient to complete this function without the need for additional hardware units.
[0079] Before outputting, the soft demodulation module 1135 in the above scheme determines that the current data is format 0 based on the accompanying identifier, and directly sends the obtained log-likelihood ratio data to the MAC layer interface through the output module 1211 of the first dedicated processing sub-unit 121, which simplifies the data path and helps to improve the processing delay of the above PUCCH processing device 100.
[0080] Optionally, the dedicated processing unit 120 includes: The second dedicated processing sub-unit 122 is used to perform differential operations on the data corresponding to format 1; The second dedicated processing subunit 122 includes a Fast Fourier Transform (FFT) module 1221, a detransform precoding module 1222, and a descrambling and decoding module 1223. The input of the FFT module 1221 is connected to the output of the peak search module 1133, and the output of the FFT module 1221 is connected to the input of the equalization module 1134. The input of the detransform precoding module 1222 is connected to the output of the equalization module 1134, and the output of the detransform precoding module 1222 is connected to the input of the soft demodulation module 1135. The peak search module 1133 is also used to determine the PUCCH format based on the associated identifier; and when the PUCCH format is format 1, to send the time-domain channel response data to the fast Fourier transform module 1221. The Fast Fourier Transform module 1221 is used to convert the time-domain channel response data into frequency-domain complex-valued data and send the frequency-domain complex-valued data to the equalization module 1134. The equalization module 1134 is also used to determine the PUCCH format based on the in-path identifier; and when the PUCCH format is format 1, to send the symbol data to the detransformation precoding module 1222. The detransform precoding module 1222 is used to perform detransform precoding processing on symbol data, obtain time-domain symbol data, and send the time-domain symbol data to the soft demodulation module 1135; The soft demodulation module 1135 is also used to determine the PUCCH format based on the associated identifier; and, when the PUCCH format is format 1, to send the log-likelihood ratio data to the descrambling and decoding module 1223. The descrambling and decoding module 1223 is used to descramble and decode the log-likelihood ratio data, and send the descrambled and decoded log-likelihood ratio data to the MAC layer interface.
[0081] After receiving the time-domain channel response data sent by the peak search module 1133, the Fast Fourier Transform (FFT) module 1221 can treat this data as an N-point discrete-time signal and directly perform an N-point complex FFT operation on it. The operation can adopt an in-situ butterfly structure, where each stage performs multiplication, addition, and rotation factor weighting on the input complex number, and combines the results stage by stage to obtain the frequency domain coefficients. After completing all cascaded operations, the module outputs N-point complex value data arranged in subcarrier order, which is the corresponding frequency domain channel response, and can be directly sent to the equalization module 1134 for subsequent processing.
[0082] The aforementioned detransform precoding module 1222 receives the frequency domain symbol data output by the equalization module 1134, which can be regarded as an N-point complex sequence, and performs an N-point inverse discrete Fourier transform (IDFT). The transform can adopt an in-situ butterfly structure, performing complex multiplication and addition and rotation factor weighting step by step, and finally mapping the frequency domain symbols back to the time domain. After completing the operation, the module outputs N-point time domain symbol data in chronological order, and sends it to the soft demodulation module 1135 through register cascading for subsequent soft decision use.
[0083] The steps of the descrambling and decoding module 1223 for descrambling and decoding the log-likelihood ratio data can include: first, determining the scrambling sequence index according to the accompanying identifier, reading the corresponding complex scrambling sequence from the internal register, and performing a sign-level XOR operation on the sequence with the input log-likelihood ratio data bit by bit to complete the descrambling. Subsequently, the descrambling and decoding module 1223 can send the descrambled LLR sequence to a decoder matching the PUCCH format, such as a Reed-Muller decoder, to convert the LLR into hard bits through soft-decision maximum likelihood decoding. After decoding, the resulting bit stream includes a CRC check result, which is then output to the MAC layer interface via a FIFO interface at a fixed interval after parallel-to-serial conversion.
[0084] In the above scheme, when the accompanying identifier indicates Format1, the Fast Fourier Transform module 1221, the Detransform Precoding module 1222, and the Descrambling and Decoding module 1223 are activated sequentially. Each module completes the frequency domain to time conversion, detransform precoding, and descrambling and decoding on the data path. Since each module is only activated when the identifier is triggered and remains bypassed at other times, the continuous occupation of the computing resources of the shared processing unit 110 is avoided, thereby reducing ineffective power consumption. On the other hand, the Fast Fourier Transform module 1221, the Detransform Precoding module 1222, and the Descrambling and Decoding module 1223 process data sequentially through the gating logic driven by the format identifier, without the need for additional arbitration or buffering. This allows Format1 data to complete differentiated operations within a single transmission time interval, meeting the protocol timing requirements and improving the processing latency of the PUCCH processing device 100. Furthermore, the interfaces of each module are consistent with the bus protocol of the shared processing unit 110, and the processing of Format1 format data can be achieved without modifying the architecture of the shared processing unit 110, which helps to reduce hardware expansion costs.
[0085] To facilitate understanding, a specific application example of the PUCCH processing device 100 in a certain application scenario is provided below. In this scenario, the PUCCH processing device 100 supports both Format0 and Format1 formats. Please refer to... Figure 3 The PUCCH processing unit 100 mainly includes: Shared processing unit 110 and dedicated processing unit 120, wherein the shared processing unit 110 mainly includes: The demapping module 1111 is used to extract the data corresponding to Format0 and Format1 from the frequency domain data output by Low-PHY based on the time domain position and frequency domain position in the parameters of Format0 and Format1 sent by the base station CPU.
[0086] The cache module 1112 is used to cache the data output by the demapping module 1111. It can be understood that once the cache module 1112 contains data, it will send a request signal to the polling scheduling sub-unit 112 and then wait for the polling scheduling sub-unit 112 to read it.
[0087] The polling scheduling sub-unit 112 is used to poll and schedule user data in the buffer module 1112. The scheduling granularity is user. Only after one user has been scheduled can another user be scheduled. It will determine whether the user is Format0 or Format1 and generate a path identifier to be output to the channel estimation preprocessing module 1132. The data flows to the last module together.
[0088] In addition, while reading data from the buffer module 1112, the polling scheduling sub-unit 112 also sends a low peak-to-average ratio (LPRR) sequence generation request to the low peak-to-average ratio (LPRR) sequence generation module 1131, driving the LPRR sequence generation module 1131 to generate a low peak-to-average ratio (LPRR) sequence, which is then multiplied by a complex number with the data read from the buffer module 1112 and output to the channel estimation preprocessing module 1132.
[0089] Low peak-to-average ratio sequence generation module 1131: used to receive requests from polling scheduling sub-unit 112 and generate a low peak-to-average ratio sequence to output to polling scheduling sub-unit 112.
[0090] Channel estimation preprocessing module 1132: used to perform inverse fast Fourier transform (IFFT) processing to convert the frequency domain signal back to the time domain signal.
[0091] Peak search module 1133: used to find the position of the peak from the data output by the channel estimation preprocessing module 1132.
[0092] Balanced module 1134: Used to implement the balanced function between Format0 and Format1.
[0093] Soft demodulation module 1135: Used to receive the output data of equalization module 1134, then soft demodulate it, and output the corresponding log-likelihood ratio (LLR) data.
[0094] In the dedicated processing unit 120, the first dedicated processing subunit 121 for performing differential operations on the frequency domain complex value data of Format0 mainly includes: Output module 1211: Used to receive the Format0 data output by the soft demodulation module 1135 and output it to the L2 interface, that is, the MAC layer interface.
[0095] In the dedicated processing unit 120, the second dedicated processing subunit 122, which performs differential operations on the data corresponding to the Format1 format, mainly includes: The Fast Fourier Transform module 1221 is used to perform Fast Fourier Transform on the data corresponding to Format1 format, converting the time domain data into frequency domain data.
[0096] The detransformation precoding module 1222 is used to perform detransformation precoding on the data corresponding to Format1 format, which can be implemented by IDFT.
[0097] The descrambling and decoding module 1223 is used to receive the data corresponding to the Format1 format output by the soft demodulation module 1135, and perform descrambling and RM decoding before outputting it to the MAC layer interface.
[0098] The working principle of the above-mentioned PUCCH processing device 100 is described below, mainly including the following stages: Data extraction stage: The demapping module 1111 receives the unified parameter packet sent by the base station CPU and parses out the time domain / frequency domain location information of Format 0 / 1. The demapping module 1111 reads the frequency domain data output by the Low-PHY according to the parameter configuration and merges the two formats of data into the buffer module 1112. This stage can achieve single-module compatibility for extracting data in both formats through dynamic parameter switching.
[0099] Unified scheduling phase: The polling scheduling sub-unit 112 polls the cache module 1112 at the user granularity to generate a path-specific format identifier. Simultaneously, the polling scheduling sub-unit 112 requests the low-peak-to-average ratio sequence generation module 1131 to generate a low-peak-to-average ratio sequence in the corresponding format. This phase achieves coordination between unified scheduling and differentiated processing through a path-specific identifier-driven mechanism.
[0100] Basic Processing Stage: The polling scheduling sub-unit 112 multiplies the data retrieved from the buffer module 1112 with the low peak-to-average power ratio (PAPR) sequence and inputs it into the channel estimation preprocessing module 1132. The channel estimation preprocessing module 1132 alternately processes the two data formats and outputs a time-domain signal. The peak search module 1133 adjusts the search parameters according to the path identifier to locate the peak position. The equalization module 1134 loads the equalization coefficients of the corresponding format to complete channel equalization. The soft demodulation module 1135 selects the quantization precision according to the identifier bits and outputs LLR data. This stage achieves resource reuse of the basic processing stage through the dynamic configuration of the shared module.
[0101] Differentiation Processing Stage: Format1 path activates the Fast Fourier Transform module 1221 to convert the time-domain signal into a frequency-domain signal. After de-transform precoding module 1222 removes the precoding, the data is returned to the soft demodulation module 1135. Format0 path directly outputs LLR data to the output module 1211. Format1 path inputs the LLR data into the descrambling and decoding module 1223 to complete the final processing. This stage achieves differentiation functionality through the precise insertion of dedicated modules.
[0102] Experiments have shown that the shared architecture in the above scheme, compared to the existing independent architecture where one format corresponds to one link: (1) Improved resource utilization: When processing the same number of Format0 and Format1 users, the lookup table resource utilization rate of the shared architecture is reduced by 40% compared with the independent architecture. This is mainly due to the hardware reuse of modules such as demapping, caching, and channel estimation preprocessing.
[0103] (2) Reduced power consumption: Module reuse in the shared architecture can reduce hardware resource consumption and directly reduce static power consumption. For example, the single-channel estimation preprocessing module 1132 replaces the original dual-module architecture, reducing the number of multipliers and adders used and reducing chip power consumption. Another example is that the unified cache pool reduces the number of registers, further reducing power consumption.
[0104] (3) Significantly enhanced scalability: The modular design of the shared architecture allows for the addition of new Format types by expanding only the dedicated module group without modifying the shared module group. For example, if Format2 needs to be introduced, only a specific module needs to be inserted into the dedicated processing unit 120 of the PUCCH processing device 100, and the shared processing unit 110 can continue to be reused. This design enables the PUCCH processing device 100 to flexibly adapt to the evolution requirements of the 5G NR standard.
[0105] Based on the same inventive concept, this application also provides a receiver, including the PUCCH processing device 100 provided in this application embodiment.
[0106] Based on the same inventive concept, this application also provides a base station, including the PUCCH processing device 100 provided in this application embodiment, or the receiver provided in this application embodiment.
[0107] Please see Figure 4 Based on the same inventive concept, this application also provides a PUCCH processing method, including: Step S210: Within the same transmission time interval, commonality operations and differential operations are performed on the frequency domain complex-valued data of various PUCCH formats output by Low-PHY to obtain the log-likelihood ratio data of the frequency domain complex-valued data; wherein, the commonality operation and the differential operation are used to characterize the commonality operation part and the differential operation part when performing operations on the frequency domain complex-valued data of various PUCCH formats, respectively; Step S220: Send the log-likelihood ratio data to the MAC layer interface.
[0108] The PUCCH processing method provided in this application embodiment can realize all the functions of the above-mentioned PUCCH processing device 100. The implementation principle and technical effects of the PUCCH processing method have been described in the foregoing device embodiments. For the sake of brevity, any part not mentioned in the method embodiment can be referred to the corresponding content in any of the foregoing device embodiments.
[0109] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.
[0110] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0111] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.
[0112] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.
[0113] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.
Claims
1. A PUCCH processing apparatus, characterized in that, It includes a shared processing unit and a dedicated processing unit connected to the shared processing unit, wherein: The shared processing unit is used to perform common operations on frequency domain complex value data of various PUCCH formats output by Low-PHY within the same transmission time interval; the shared processing unit includes a data extraction subunit, a polling scheduling subunit, and a baseband operation subunit connected in sequence; The dedicated processing unit is used to perform differential operations on data corresponding to various PUCCH formats; wherein, the common operation and the differential operation are used to characterize the common operation part and the differential operation part when operating on the frequency domain complex value data of various PUCCH formats, respectively; The polling scheduling sub-unit is used to generate the accompanying identifier of the frequency domain complex value data, and input the frequency domain complex value data and the corresponding accompanying identifier into the baseband operation sub-unit; wherein, the accompanying identifier is used to indicate the PUCCH format to which the frequency domain complex value data belongs and the corresponding user identifier; the accompanying identifier and the frequency domain complex value data are cascaded and transmitted using the same data bus field, and are synchronously latched at each stage of the pipeline of the baseband operation sub-unit.
2. The PUCCH processing apparatus according to claim 1, characterized in that, The data extraction sub-unit is used to extract the frequency domain complex value data of each PUCCH format from the frequency domain complex value data of the various PUCCH formats output by the Low-PHY. The polling scheduling sub-unit is used to poll and schedule the frequency domain complex value data extracted by the data extraction sub-unit at the user level, and input the frequency domain complex value data into the baseband operation sub-unit; The baseband operation sub-unit is used to perform common operations on the frequency domain complex-valued data.
3. The PUCCH processing apparatus according to claim 2, characterized in that, The data extraction subunit includes a demapping module and a cache module connected to the demapping module, wherein: The demapping module is used to extract the frequency domain complex value data of each PUCCH format from the frequency domain complex value data of multiple PUCCH formats based on the resource location information issued by the base station processor, and send the frequency domain complex value data and its format identifier to the cache module; The caching module is used to partition and store the frequency domain complex value data of each of the PUCCH formats based on the format identifier.
4. The PUCCH processing apparatus according to claim 2, characterized in that, The baseband operation subunit includes a low peak-to-average ratio sequence generation module connected to the polling scheduling subunit, wherein: The polling scheduling sub-unit is also used to send a low peak-to-average ratio sequence generation request to the low peak-to-average ratio sequence generation module; and to perform a complex multiplication operation on the low peak-to-average ratio sequence returned by the low peak-to-average ratio sequence generation module and the frequency domain complex value data, and send the frequency domain complex value data after the complex multiplication process to the baseband operation sub-unit. The low peak-to-average ratio sequence generation module is used to generate and return the low peak-to-average ratio sequence in the corresponding format after receiving the low peak-to-average ratio sequence generation request sent by the polling scheduling sub-unit.
5. The PUCCH processing apparatus according to claim 2, characterized in that, The baseband arithmetic subunit further includes a channel estimation preprocessing module, a peak search module, an equalization module, and a soft demodulation module connected in sequence, wherein: The channel estimation preprocessing module is used to perform a fast inverse Fourier transform on the frequency domain complex value data to obtain time domain channel response data, and send the time domain channel response data to the peak search module. The peak search module is used to determine the peak position from the time-domain channel response data and send the peak position to the equalization module; The equalization module is used to perform channel equalization processing on the data input to the equalization module based on the peak position, obtain equalized symbol data, and send the symbol data to the soft demodulation module; The soft demodulation module is used to perform soft demodulation processing on the symbol data to obtain log-likelihood ratio data.
6. The PUCCH processing apparatus according to claim 5, characterized in that, The channel estimation preprocessing module is used to determine the transformation coefficients based on the path identifier, and to perform a fast inverse Fourier transform on the frequency domain complex value data using the transformation coefficients to obtain time domain channel response data. The peak search module is used to determine a decision threshold and a search window length based on the accompanying identifier, and to determine the peak position in the time-domain channel response data using the decision threshold and the search window length. The equalization module is used to determine the equalization coefficient based on the associated identifier, and to perform channel equalization processing on the data input to the equalization module using the equalization coefficient and the peak position to obtain equalized symbol data. The soft demodulation module is used to determine the quantization accuracy based on the in-path identifier, and to perform soft demodulation processing on the symbol data using the quantization accuracy to obtain the log-likelihood ratio data.
7. The PUCCH processing apparatus according to claim 5, characterized in that, The dedicated processing unit includes: The first dedicated processing sub-unit is used to perform differential operations on the data corresponding to format 0; The first dedicated processing subunit includes an output module connected to the soft demodulation module, wherein: The soft demodulation module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is format 0, send the log-likelihood ratio data to the output module; The output module is used to send the log-likelihood ratio data output by the soft demodulation module to the MAC layer interface.
8. The PUCCH processing apparatus according to claim 5, characterized in that, The dedicated processing unit includes: The second dedicated processing sub-unit is used to perform differential operations on the data corresponding to format 1. The second dedicated processing subunit includes a Fast Fourier Transform (FFT) module, a detransform precoding module, and a descrambling and decoding module. The input of the FFT module is connected to the output of the peak search module, the output of the FFT module is connected to the input of the equalization module, the input of the detransform precoding module is connected to the output of the equalization module, the output of the detransform precoding module is connected to the input of the soft demodulation module, and the output of the soft demodulation module is connected to the descrambling and decoding module. Wherein: The peak search module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is format 1, to send the time-domain channel response data to the fast Fourier transform module. The Fast Fourier Transform module is used to convert the time-domain channel response data into frequency-domain complex-valued data, and send the frequency-domain complex-valued data to the equalization module. The equalization module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is the format 1, send the symbol data to the detransformation precoding module; The detransform precoding module is used to perform detransform precoding processing on the symbol data to obtain time-domain symbol data, and send the time-domain symbol data to the soft demodulation module; The soft demodulation module is further configured to determine the PUCCH format based on the accompanying identifier; and, when the PUCCH format is the format 1, send the log-likelihood ratio data to the descrambling and decoding module; The descrambling and decoding module is used to descramble and decode the log-likelihood ratio data, and send the log-likelihood ratio data after the descrambling and decoding processes to the MAC layer interface.
9. A receiver, characterized in that, include: The PUCCH processing apparatus as described in any one of claims 1 to 8.
10. A base station, characterized in that, include: The PUCCH processing apparatus as described in any one of claims 1 to 8; Alternatively, the receiver as described in claim 9.
11. A PUCCH processing method, characterized in that, The method includes: Within the same transmission time interval, common operations and differential operations are performed on the frequency domain complex-valued data of various PUCCH formats output by Low-PHY to obtain the log-likelihood ratio data of the frequency domain complex-valued data; wherein, the common operations and the differential operations are used to characterize the common operation part and the differential operation part when performing operations on the frequency domain complex-valued data of various PUCCH formats, respectively. Generate an accompanying identifier for the frequency domain complex-valued data; wherein, the accompanying identifier is used to indicate the PUCCH format to which the frequency domain complex-valued data belongs and the corresponding user identifier, and the accompanying identifier and the frequency domain complex-valued data are cascaded and transmitted through the same data bus field and are synchronously latched at each stage of the pipeline; The log-likelihood ratio data is sent to the MAC layer interface.
Citation Information
Patent Citations
Terminal, base station, wireless communication system, and wireless communication method
CN118511552A
