Device and method for carrying out complex multiplication and butterfly calculation by virtue of floating point unit of processor
The invention provides a device and a method for carrying out complex multiplication and butterfly calculation by virtue of a floating point unit of a processor. The floating point unit is divided into a floating point addition unit and a floating point multiplication unit, wherein a few interconnec...
Gespeichert in:
Hauptverfasser: | , , , , |
---|---|
Format: | Patent |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention provides a device and a method for carrying out complex multiplication and butterfly calculation by virtue of a floating point unit of a processor. The floating point unit is divided into a floating point addition unit and a floating point multiplication unit, wherein a few interconnection lines are additionally arranged between data ports of the floating point addition unit and the floating point multiplication unit; a special data path is formed between the floating point addition unit and the floating point multiplication unit; real and imaginary components of a complex are input into a data port of the floating point unit according to a determined ordering rule; the floating point complex multiplication and floating point complex butterfly calculation can be achieved under the control of decoding signals generated by a decoding unit. According to the method and the device, hardware resources are saved; compared with the existing complex multiplication and butterfly calculation function unit, the flexibility is further improved; meanwhile, parallel and flow calculation of a large amount of data can be well implemented; high real-time performance and high resource utilization rate are obtained. |
---|