2009 International Symposium on System-on-Chip 2009
DOI: 10.1109/socc.2009.5335658
|View full text |Cite
|
Sign up to set email alerts
|

A DSP architecture optimized for wireless baseband

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

0
3
0

Year Published

2010
2010
2024
2024

Publication Types

Select...
4
2

Relationship

0
6

Authors

Journals

citations
Cited by 12 publications
(3 citation statements)
references
References 2 publications
0
3
0
Order By: Relevance
“…However in the case of high DLP algorithms our architecture has performances in the range of SIMD solutions. For example: Tensilica's DSP [8] computes a FIR operation (256 complex input) in about 520 clock cycles; a programmable baseband processor described in [22] performs the same operation in about 1,000 clock cycles. The performance of our solution is somewhere in the middle, performing a correlation point (FIR operation with 256 complex inputs) in 688 clock cycles (see Table 2).…”
Section: Cell Search Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…However in the case of high DLP algorithms our architecture has performances in the range of SIMD solutions. For example: Tensilica's DSP [8] computes a FIR operation (256 complex input) in about 520 clock cycles; a programmable baseband processor described in [22] performs the same operation in about 1,000 clock cycles. The performance of our solution is somewhere in the middle, performing a correlation point (FIR operation with 256 complex inputs) in 688 clock cycles (see Table 2).…”
Section: Cell Search Resultsmentioning
confidence: 99%
“…A similar approach is used by Tensilica in their ConnX Baseband Engine processor [8]. The processor uses a 3-way VLIW instructions set and supports 8-way SIMD operations.…”
Section: Introductionmentioning
confidence: 99%
“…Developer can vectorize the program himself using ConnX BBE's data type and intrinsic function. In addition Xtensa C and C++ compiler can automatically do this vectorization with little or no human intervention ( Figure 4 [19].…”
Section: Connx Bbe By Tensilicamentioning
confidence: 99%