Placement and scheduling of radio signal processing dataflow operations

Inventors

O'Shea, Timothy James

Assignees

DeepSig Inc

Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for placement and scheduling of radio signal processing dataflow operations. An example method provides a primitive radio signal processing computational dataflow graph that comprises nodes representing operations and directed edges representing data flow. The nodes and directed edges of the primitive radio signal processing computational dataflow graph are partitioned to produce a set of software kernels that, when executed on processing units of a target hardware platform, achieve a specific optimization objective. Runtime resource scheduling, including data placement for individual software kernels in the set of software kernels to efficiently execute operations on the processing units of the target hardware platform. The resources of the processing units in the target hardware platform are then allocated according to the defined runtime resource scheduling.

Core Innovation

A primitive radio signal processing computational dataflow graph corresponding to a radio signal processing application is provided, where nodes represent operations and directed edges represent data flow. The radio signal processing application includes at least one of RF synchronization, RF demodulation, error correction decoding, RF beam-steering, or multi-antenna RF-signal combining. The nodes and directed edges of the primitive radio signal processing computational dataflow graph are partitioned to produce a set of software kernels that, when executed using a baseband processing unit, achieve a specific optimization objective.

Runtime resource scheduling data is determined, the runtime resource scheduling data including data placement for individual software kernels in the set of software kernels across processing resources of the baseband processing unit. The method performs at least one of RF signal transmission or RF signal reception by allocating the processing resources of the baseband processing unit according to the data placement for the individual software kernels. The runtime resource scheduling and the allocation are based on the partitioning and the specific optimization objective.

A computational dataflow graph corresponding to a radio signal processing application is also partitioned into a set of software kernels for execution using a plurality of processing units of a target hardware platform. Data placement for individual software kernels across the plurality of processing units is determined based on resource consumption by at least one other process that is running or will run concurrently, while executing the operations using the target hardware platform. Resource use is measured and, based on the measured resource use, at least some of the nodes and directed edges are re-partitioned to produce an updated set of software kernels.

Claims Coverage

The document includes three independent claims, each centered on partitioning a radio signal processing computational dataflow graph into software kernels and performing execution using data placement/resource allocation, with claim-specific constraints on concurrent processes, measurement-driven re-partitioning, and optimization-objective targeting.

Across the independent claims, the core coverage is the end-to-end flow from providing a radio signal processing computational dataflow graph, partitioning it into software kernels to meet a specific optimization objective, determining data placement across processing resources or units, executing by allocating resources according to the placement, and, in one claim family, using concurrent-process resource consumption and measured resource use to drive scheduling and re-partitioning.

Stated Advantages

Documented Applications