Back to results
US 8,948,303 B1Granted
Communication device and method of crest factor reduction using amplitude compression
AssigneeMicroelectronics Technology Inc.
InventorKhiem Van Cai
Life of this patent
Estimating the term…
Abstract
A method of crest factor reduction using amplitude compression for multi carrier signals. Major elements of the inter-modulation distortion caused by amplitude suppression are removed to be outside of the desired signal band, and the peak amplitude is compressed, thus producing a signal with lower peak to average ratio for a required maximum error vector magnitude.
Claims12
A method of crest factor reduction (CFR) using amplitude compression, executable by a communication device, the method comprising:
- receiving a complex input signal x(t) at a first sampling rate Rs;
- up-sampling the complex input signal x(t) to a second sampling rate M·Rs that is larger than an instantaneous bandwidth of the complex input signal x(t), to generate an up-sampled complex signal y(t);
- computing an amplitude signal a(t) representing an amplitude of the up-sampled signal y(t) and sending the amplitude signal a(t) to an amplitude compressor of the communication device to extract the compressed gain g1(t);
- multiplying the gain g1(t) with a correction gain g2(t) to obtain a final gain signal g(t);
- performing amplitude compression of the up-sampled complex signal y(t) by multiplying the gain signal g(t) with the up-sampled complex signal y(t) to produce the amplitude compressed signal z(t);
- performing signal filtering by the filter of the communication device on the amplitude compressed signal z(t) to generate a post-CFR signal u(t), wherein the post-CFR signal u(t) is at a higher sampling rate and with a lower peak to average (PAR) ratio than the complex input signal x(t).
3 more elements
The method as claimed in clam 1, further comprising:
- assigning the compressed gain g1(t) a value substantially equal to an identity element of an operation of the step of multiplying if 0<a(t)<T1;
- assigning the compressed gain g1(t) a value substantially equal to T1/a(t) if T1<a(t)<T2;
- and
- assigning the compressed gain g1(t) a value less than T1/a(t) if T2<a(t), wherein T1 and T2 are two amplitude threshold values of the amplitude signal.
1 more element
A wireless communication device, comprising:
- an up-sampling stage operable to receive a complex input signal x(t) at a first sampling rate Rs and up-sample the complex input signal x(t) to a second sampling rate M·Rs that is larger than an instantaneous bandwidth of the complex input signal x(t), to generate an up-sampled complex signal y(t);
- a first amplitude computing stage operable to compute an amplitude signal a(t) representing an amplitude of the up-sampled signal y(t) and sending the amplitude signal a(t) to an amplitude compressor of the communication device to extract the compressed gain g1(t);
- a first multiplier operable to multiply the gain g1(t) with a correction gain g2(t) to obtain a final gain signal g(t);
- a second multiplier operable to performing amplitude compression of the up-sampled complex signal y(t) by multiplying the gain signal g(t) with the up-sampled complex signal y(t) to produce the amplitude compressed signal z(t);
- a filter operable to perform signal filtering on the amplitude compressed signal z(t) to generate a post-CFR signal u(t), wherein the post-CFR signal u(t) is at a higher sampling rate and with a lower peak to average (PAR) ratio than the complex input signal x(t).
2 more elements
The wireless communication device as claimed in clam 7, wherein in obtaining the gain g2(t), the method further comprises:
- a first amplitude computing stage operable to compute an amplitude b(t) of the post-CFR signal u(t);
- and
- a gain correction stage operable to execute a gain correction process comprising:
- obtaining sample N points b1, b2, to bN of the amplitude b(t);
- obtaining sample N points a1, a2 to aN of the amplitude a(t);
- obtaining short term power ratio G, wherein and
- G
- =
- P
- y
- P
- u
- ,
- P
- y
- =
- ∑
- i
- =
- 1
- N
-
- a
- i
- 2
- ,
- Pu=Σi=1Nbi2, and i is a integer variable ranging from 1 to N;
- obtaining an average power ratio Gm from the short term power ratio G;
- and
- determining the correction gain g2(t) based on square root of the average power ratio Gm.
27 more elements
Dependencies are inferred from the claim text.