Hybrid architecture for signal processing and signal processing accelerator
Systems and methods for configuring a SPA are disclosed. The SPA comprises a plurality of input ports, a plurality of data memory units, signal processing circuitry, and an enable block including at least two counters. Each counter determines an amount of unprocessed data that is stored in a respective one of the plurality of data memory units, and the enable block is configured to disable the signal processing circuitry until a predetermined amount of data is received over the input ports.
1. An integrated circuit system, comprising:
programmable logic circuitry in a first region of the integrated circuit system;
a two-dimensional array of instruction set architecture (ISA)-based signal processors in a second region of the integrated circuit system different from the first region;
an interconnect network that enables communication between ISA-based signal processors of the two-dimensional array of ISA-based signal processors; and
a hardened processing system configured to transmit instructions over the interconnect network to the two-dimensional array of ISA-based signal processors to configure the contents of one or more ISA-based signal processors of the two-dimensional array of ISA-based signal processors.
2. The integrated circuit system of claim 1 , wherein the programmable logic circuitry comprises field programmable gate array (FPGA) circuitry configured to send data to or receive data from the two-dimensional array of ISA-based signal processors.
3. The integrated circuit system of claim 1 , wherein the two-dimensional array of ISA-based signal processors has vector processing functionality.
4. The integrated circuit system of claim 1 , wherein the two-dimensional array of ISA-based signal processors is programmable to perform digital pre-distortion (DPD) processing.
5. The integrated circuit system of claim 1 , wherein the two-dimensional array of ISA-based signal processors comprises distributed memory, wherein each ISA-based signal processor of the array of ISA-based signal processors is associated with at least one memory block of the distributed memory.
6. The integrated circuit system of claim 1 , wherein the interconnect network connects a first ISA-based signal processor of the array of ISA-based signal processors to a second ISA-based signal processor of the two-dimensional array of ISA-based signal processors to skip a third ISA-based signal processor located between the first ISA-based signal processor and the second ISA-based signal processor in the two-dimensional array of ISA-based signal processors.
7. The integrated circuit system of claim 6 , wherein the two-dimensional array of ISA-based signal processors comprises at least a first row, a second row, and a third row of ISA-based signal processors, and a first column, a second column, and a third column of ISA-based signal processors.
8. The integrated circuit system of claim 1 , wherein an ISA-based signal processor of the two-dimensional array of ISA-based signal processors comprises an instruction unit, a program counter, or a combination thereof.
9. The integrated circuit system of claim 8 , wherein the instruction unit stores microcode for the ISA-based signal processor.
10. The integrated circuit system of claim 8 , wherein the program counter controls a sequence of processing steps in the ISA-based signal processor.
11. The integrated circuit system of claim 1 , wherein the two-dimensional array of ISA-based signal processors comprises at least a first row, a second row, and a third row of ISA-based signal processors, and a first column, a second column, and a third column of ISA-based signal processors.
12. An integrated circuit system, comprising:
programmable logic circuitry in a first region of the integrated circuit system;
a two-dimensional array of instruction set architecture (ISA)-based signal processors in a second region of the integrated circuit system different from the first region;
an interconnect network that enables communication between ISA-based signal processors of the two-dimensional array of ISA-based signal processors to bypass at least one ISA-based signal processors of the two-dimensional array of ISA-based signal processors in transmitting at least one signal in the communication between other ISA-based signal processors of the two-dimensional array of ISA-based signal processors; and
a hardened processing system configured to transmit instructions over the interconnect network to the two-dimensional array of ISA-based signal processors to configure contents of one or more ISA-based signal processors of the two-dimensional array of ISA-based signal processors.
13. The integrated circuit system of claim 12 , wherein the two-dimensional array of ISA-based signal processors comprises at least a first row, a second row, and a third row of ISA-based signal processors, and a first column, a second column, and a third column of ISA-based signal processors.
14. The integrated circuit system of claim 12 , wherein the two-dimensional array of ISA-based signal processors is programmable to perform digital pre-distortion (DPD) processing.
15. The integrated circuit system of claim 12 , wherein the interconnect network is configured to enable a first ISA-based signal processor of the two-dimensional array of ISA-based signal processors to stream data directly to a second ISA-based signal processor of the two-dimensional array of ISA-based signal processors.
16. The integrated circuit system of claim 15 , wherein the interconnect network communicatively connects the first ISA-based signal processor directly to the second ISA-based signal processor using unidirectional routing circuitry.
17. The integrated circuit system of claim 12 , wherein an ISA-based signal processor of the two-dimensional array of ISA-based signal processors comprises an instruction unit, a program counter, or a combination thereof.
18. The integrated circuit system of claim 17 , wherein the instruction unit stores microcode for the ISA-based signal processor.
19. A method, comprising:
receiving input data at a two-dimensional array of instruction set architecture (ISA)-based signal processors in a first region of an integrated circuit separate from a second region comprising programmable logic circuitry of the integrated circuit;
routing communication between at least a portion of the two-dimensional array of ISA-based signal processors via an interconnect network; and
transmitting instructions, via a hardened processing system, to the two-dimensional array of ISA-based signal processors to cause the two-dimensional array of ISA-based signal processors to perform a corresponding function.
20. The method of claim 19 , wherein the two-dimensional array of ISA-based signal processors has vector processing functionality.