CN115878558A - Universal SDR platform supporting mixed granularity reconstruction - Google Patents

Universal SDR platform supporting mixed granularity reconstruction Download PDF

Info

Publication number
CN115878558A
CN115878558A CN202211506363.3A CN202211506363A CN115878558A CN 115878558 A CN115878558 A CN 115878558A CN 202211506363 A CN202211506363 A CN 202211506363A CN 115878558 A CN115878558 A CN 115878558A
Authority
CN
China
Prior art keywords
granularity
cgra
fpga
block
mixed
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211506363.3A
Other languages
Chinese (zh)
Inventor
吴雪桐
陆芳
赵峰
何宇童
贺聪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
White Box Shanghai Microelectronics Technology Co ltd
Original Assignee
White Box Shanghai Microelectronics Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by White Box Shanghai Microelectronics Technology Co ltd filed Critical White Box Shanghai Microelectronics Technology Co ltd
Priority to CN202211506363.3A priority Critical patent/CN115878558A/en
Publication of CN115878558A publication Critical patent/CN115878558A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Logic Circuits (AREA)

Abstract

The invention relates to a mixed granularity reconfigurable digital signal processing module which comprises a bottom layer and an intermediate layer, wherein the bottom layer is a fine-granularity FPGA basic processing unit, the intermediate layer is a coarse-granularity CGRA basic processing unit, and the coarse-granularity CGRA basic processing unit comprises an FPGA basic unit FPGABlock, a CGRA basic unit CGRABlock and a DSP engine basic unit which are sequentially increased in granularity, and an SRAM. The invention uses the programmable logic resource with FPGA fine granularity and the DSP resource to physically realize the mixed-granularity reconfigurable digital signal processing and accelerate the reconfigurable capability of the system.

Description

Universal SDR platform supporting mixed granularity reconstruction
Technical Field
The invention relates to the technical field of SDR (software defined disc) platforms, in particular to a general SDR platform supporting mixed granularity reconstruction.
Background
In a conventional FPGA development model, a user generally models an application scenario using a Hardware Description Language (HDL), and then maps the hardware scenario onto an FPGA through a specific FPGA development tool, thereby finally generating an FPGA image that can be run. This traditional FPGA development model causes hardware resources to be idle for a large portion of the time. In order to improve the development efficiency of the FPGA and better utilize the logic resources of the FPGA, a top-level user does not need to pay much attention to the implementation mode and details of the FPGA hardware logic. Thus, the FPGA virtualization technology is produced. Briefly, an FPGA Overlay (virtualization) is a virtual programmable architecture located on top of the FPGA hardware layer and connected to top-level applications.
The CGRA (Coarse-Grained Reconfigurable Array) as a programmable processing unit supporting a higher-level programming model can implement the FPGA virtualization technology, and is a Coarse-Grained Reconfigurable architecture. It improves efficiency by building modules internally, reducing interconnections. And the driving of the data stream is realized by configuring different functional areas. And when the next moment, the unit can be quickly configured with different functions, and the data flow is driven again, which is a data driving and dynamic reconstruction calculation mode, so that the data is high-speed parallel, the layout, wiring and compiling speed is high, and the system performance can be improved.
Software Defined Radio (SDR) is a Radio broadcast communication technology that is based on Software Defined wireless communication protocols rather than being implemented by hardwire. The frequency bands, air interface protocols and functions may be upgraded by software downloads and updates without complete hardware replacement. A software defined radio is a system and architecture that must have the ability to be reprogrammable and reconfigurable to enable devices to be used with a variety of standards, a variety of frequency bands and to perform a variety of functions, and that will not only use programmable devices to implement baseband digital signal processing, but will also program and reconfigure the analog circuitry for radio and intermediate frequencies. The requirements of the functionality of software defined radio include: the ability to reprogram and reset, the ability to provide and change services, the ability to support multiple standards, and the ability to intelligently utilize spectrum, among other things. It should be appreciated that SDR is not a stand alone technology, but a common platform that can be used for all technologies.
SDR baseband processing typically requires a processor for system control and configuration and an FPGA to implement the data path and control of the large computational signal processing, minimizing the delay of the system. The SDR system developed based on the FPGA platform has strong real-time processing capability, but has high development difficulty and high development cost.
Disclosure of Invention
The invention aims to solve the technical problem of providing a general SDR platform supporting mixed granularity reconfiguration, which uses programmable logic resources and DSP resources of FPGA fine granularity to physically realize mixed granularity reconfigurable digital signal processing and accelerate the system reconfigurable capability.
The technical scheme adopted by the invention for solving the technical problem is as follows: the mixed-granularity reconfigurable digital signal processing module comprises a bottom layer and an intermediate layer, wherein the bottom layer is a fine-granularity FPGA basic processing unit, the intermediate layer is a coarse-granularity CGRA basic processing unit, and the coarse-granularity CGRA basic processing unit comprises an FPGA basic unit FPGA Block, a CGRA basic unit CGRA Block and a DSP engine basic unit, and SRAM, wherein the sizes of the FPGA basic unit FPGA Block, the CGRA basic unit CGRA Block and the DSP engine basic unit are sequentially increased.
The FPGA basic unit FPGA Block comprises a Block random access memory BRAM and a configurable logic Block CLB.
The CGRA basic unit CGRA Block comprises a multiplied by b computing units PE which are arranged according to an array, two adjacent computing units PE are interconnected to form a grid structure, and the CGRA basic unit CGRA Block achieves different functions through the computing units PE with different connection architectures.
The DSP engine basic unit comprises a plurality of DSPu, and the DSPu is realized through a configurable logic block CLB and a DSP48Slice in the bottom layer.
The fine-grained FPGA basic processing unit comprises a configurable logic block CLB, an input/output port IO, a block random access memory BRAM and a DSP48Slice, wherein the block random access memory BRAM and the DSP48Slice are both surrounded by different configurable logic blocks CLB.
The technical scheme adopted by the invention for solving the technical problem is as follows: there is provided a general purpose SDR platform comprising hardware logic including a mixed-granularity reconfigurable digital signal processing module as described above.
Advantageous effects
Due to the adoption of the technical scheme, compared with the prior art, the invention has the following advantages and positive effects: the invention uses the programmable logic resource and DSP resource of FPGA fine grain to realize the mixed grain reconfigurable digital signal processing physically, and quickens the system reconfigurable ability; the user of the invention can use C language with higher flexibility to process digital signal, and then the invention is assisted by HDL which describes hardware more accurately, thereby the invention is more convenient for the user to use and the circuit design is more accurate.
Drawings
FIG. 1 is an overall architecture diagram of a generic SDR platform supporting mixed-granularity reconfigurability according to an embodiment of the invention;
FIG. 2 is a block diagram of a mixed-granularity reconfigurable digital signal processing module according to an embodiment of the present invention;
fig. 3 is a signal processing flow diagram of a mixed-granularity reconfigurable digital signal processing module according to an embodiment of the present invention.
Detailed Description
The invention will be further illustrated with reference to the following specific examples. It should be understood that these examples are for illustrative purposes only and are not intended to limit the scope of the present invention. Further, it should be understood that various changes or modifications of the present invention can be made by those skilled in the art after reading the teaching of the present invention, and these equivalents also fall within the scope of the claims appended to the present application.
The embodiment of the invention relates to a mixed granularity reconfigurable digital signal processing module which can be applied to a general SDR platform, and the embodiment is described in detail as follows:
as shown in fig. 1, the figure depicts the overall architecture of a general SDR platform supporting mixed-granularity reconfigurability, which includes two parts, a processing system and hardware logic. The processing system comprises an Application Processing Unit (APU), a real-time processing unit (RPU), a Graphic Processing Unit (GPU), a platform management unit, a DDR controller, a system control unit, a first high-speed connection, a general connection and the like; the hardware logic includes the key points of the embodiment, i.e., a mixed-granularity reconfigurable digital signal processing module (MGR-DSP), a digital-to-analog conversion unit, a second high-speed connection, a general I/O interface, and the like. When different RF analog signals enter and exit the SDR platform from the peripheral interface, the specific RF signal processing is completed by hardware logic including MGR-DSP, and the protocol stack is partially completed by the processing system including APU and RPU. The first high-speed connection mainly comprises connection interfaces such as USB, PCle and SATA, the universal connection mainly comprises universal connection interfaces such as UART, SPI, NAND and GigE, and the second high-speed connection mainly comprises interfaces such as Ethernet MAC and PCle.
As shown in fig. 2, the figure describes a specific architecture of a mixed-granularity reconfigurable digital signal processing module in a general SDR platform, and the architecture is composed of two layers, including a bottom layer and a middle layer, wherein the middle layer is mapped to the bottom layer through a mapping tool. The basic processing unit of the FPGA with the fine granularity at the bottom layer comprises a Configurable Logic Block (CLB), an input/output port (IO), a Block Random Access Memory (BRAM), a DSP48Slice and some interconnections. The intermediate layer comprises a fine-grained FPGA basic unit (FPGA Block), a coarse-grained CGRA basic unit (CGRABlock), a coarser-grained DSP engine basic unit and an SRAM (static random access memory), wherein the fine-grained FPGA basic unit FPGABlock comprises a Block random access memory BRAM and a configurable logic Block CLB; the coarse-grained CGRA basic units CGRABlock comprise a multiplied by b computing units PE arranged in an array, two adjacent computing units PE are interconnected to form a grid structure, and the coarse-grained CGRA basic units CGRABlock realize different functions through the computing units PE with different connection architectures; the basic unit in the DSP engine is physically realized by the programmable logic unit CLB of the bottom layer FPGA and DSP48Slice resources, wherein a simple realization mode is to utilize DSP extension in a RISC-V instruction set.
As shown in fig. 3, the flow chart describes the process of controlling signal processing based on the hardware logic area of mixed-granularity reconfigurable digital signal processing module (MGR-DSP), which includes control program, used languages C and HDL, MGR-DSP, ADC/DAC and RF Block. The control program controls MGR-DSP jointly through UHD library and HDL in C language. Parameters such as sampling frequency of signals, gain of transmitting or receiving signals, clock frequency and the like can be modified through the UHD library, wherein specific functions are written by HDL with higher description accuracy. The two languages are converted to each other through a certain language conversion, so that a user can modify different signal parameters by calling the written function without more finely writing hardware, only needs to transfer some work (such as register operation) to the HDL (hardware description language), and is simple to develop and easy to master. The processed digital signal is processed by digital-to-analog conversion (ADC/DAC) and other devices to obtain a radio frequency signal.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications. It is intended that the scope of the invention be defined by the claims and their equivalents.

Claims (6)

1. The mixed-granularity reconfigurable digital signal processing module is characterized by comprising a bottom layer and an intermediate layer, wherein the bottom layer is a fine-granularity FPGA basic processing unit, the intermediate layer is a coarse-granularity CGRA basic processing unit, and the coarse-granularity CGRA basic processing unit comprises an FPGA basic unit FPGABlock, a CGRA basic unit CGRA Block and a DSP engine basic unit which are sequentially increased in granularity, and an SRAM.
2. The mixed-granularity reconfigurable digital signal processing module according to claim 1, characterized in that the FPGA basic unit FPGA Block comprises a Block random access memory BRAM and a configurable logic Block CLB.
3. The mixed-granularity reconfigurable digital signal processing module according to claim 1, wherein the CGRA basic unit CGRA Block includes a × b arithmetic units PE arranged in an array, two adjacent arithmetic units PE are interconnected to form a grid structure, and the CGRA basic unit CGRA Block implements different functions through arithmetic units PE of different connection architectures.
4. The mixed-granularity reconfigurable digital signal processing module according to claim 1, wherein the DSP engine basic unit comprises a plurality of DSPu's implemented by configurable logic blocks CLB and DSP48Slice in an underlying layer.
5. The mixed-granularity reconfigurable digital signal processing module according to claim 1, wherein the fine-granularity FPGA basic processing unit comprises a configurable logic block CLB, an input/output port IO, a block random access memory BRAM, and a DSP48Slice, wherein each bit of the block random access memory BRAM and the DSP48Slice is surrounded by a different configurable logic block CLB.
6. A general purpose SDR platform comprising hardware logic comprising the mixed granularity reconfigurable digital signal processing module of any one of claims 1-5.
CN202211506363.3A 2022-11-29 2022-11-29 Universal SDR platform supporting mixed granularity reconstruction Pending CN115878558A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211506363.3A CN115878558A (en) 2022-11-29 2022-11-29 Universal SDR platform supporting mixed granularity reconstruction

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211506363.3A CN115878558A (en) 2022-11-29 2022-11-29 Universal SDR platform supporting mixed granularity reconstruction

Publications (1)

Publication Number Publication Date
CN115878558A true CN115878558A (en) 2023-03-31

Family

ID=85764431

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211506363.3A Pending CN115878558A (en) 2022-11-29 2022-11-29 Universal SDR platform supporting mixed granularity reconstruction

Country Status (1)

Country Link
CN (1) CN115878558A (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060211387A1 (en) * 2005-02-17 2006-09-21 Samsung Electronics Co., Ltd. Multistandard SDR architecture using context-based operation reconfigurable instruction set processors
CN112084139A (en) * 2020-08-25 2020-12-15 上海交通大学 Multi-emission mixed granularity reconfigurable array processor based on data flow driving
CN214045680U (en) * 2020-12-29 2021-08-24 上海瀚芯实业发展合伙企业(有限合伙) Coarse-grained reconfigurable OFDM transmitting end, receiving end and communication system
CN113468102A (en) * 2021-07-22 2021-10-01 无锡沐创集成电路设计有限公司 Mixed-granularity computing circuit module and computing system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060211387A1 (en) * 2005-02-17 2006-09-21 Samsung Electronics Co., Ltd. Multistandard SDR architecture using context-based operation reconfigurable instruction set processors
CN112084139A (en) * 2020-08-25 2020-12-15 上海交通大学 Multi-emission mixed granularity reconfigurable array processor based on data flow driving
CN214045680U (en) * 2020-12-29 2021-08-24 上海瀚芯实业发展合伙企业(有限合伙) Coarse-grained reconfigurable OFDM transmitting end, receiving end and communication system
CN113468102A (en) * 2021-07-22 2021-10-01 无锡沐创集成电路设计有限公司 Mixed-granularity computing circuit module and computing system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
邹晨 等: "基于FPGA的动态可重配置方法研究", 《航空计算技术》, vol. 42, no. 3, 31 May 2012 (2012-05-31), pages 125 - 129 *
郭彪: "面向可重构SoC的软件无线电架构与调度算法设计", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 9, 15 September 2022 (2022-09-15), pages 136 - 66 *

Similar Documents

Publication Publication Date Title
US10634722B1 (en) Combined analog architecture and functionality in a mixed-signal array
Majer et al. The Erlangen Slot Machine: A dynamically reconfigurable FPGA-based computer
US7895416B2 (en) Reconfigurable integrated circuit
US20170177536A1 (en) Microcontroller programmable system on a chip
US10210914B2 (en) Programmable logic accelerator in system on chip
EP1220108A2 (en) Programmable circuit
KR20030096283A (en) Adaptive integrated circuitry with heterogeneous and reconfigurable matrices of diverse and adaptive computational units having fixed, application specific computational elements
KR20220157377A (en) Scalable Array Architecture for In-Memory Computing
EP1556760A2 (en) Common interface framework for developing field programmable based applications independent of target circuit board
CN107092577B (en) Integrated circuit device with programmable block and analog circuit control
JP2018530040A (en) Self-adaptive chip and placement method
Ramacher et al. Architecture and implementation of a Software-Defined Radio baseband processor
EP1290546B1 (en) Programmable single-chip device and related development environment
CN115878558A (en) Universal SDR platform supporting mixed granularity reconstruction
US20060101232A1 (en) Semiconductor integrated circuit
US7620678B1 (en) Method and system for reducing the time-to-market concerns for embedded system design
CN111209246B (en) Miniature programmable on-chip computer based on multi-chip packaging technology
CN114722001A (en) Software-defined analog-digital hybrid SoC chip architecture
KR102253414B1 (en) Method for design accelerator of cnn-rnn architecture conversion type
Baskaran et al. A hardware operating system based approach for run-time reconfigurable platform of embedded devices
Kaneko et al. An integrated machine code monitor for a RISC-V processor on an FPGA
CN117391035A (en) Chip with self-defined pin function and implementation method
CN116303215A (en) Master-slave type multi-DSP processing system, method, equipment and storage medium
Mar et al. An architecture for a programmable mixed-signal device
Li et al. Research on Reconfigurable Technology and a Design of Application

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination