Abstract
|
|
---|---|
For the digital signal processing (DSP) part of mixed-signal systems, fixed-point arith-metic is favoured due to its high efficiency in terms of implementation cost, namely energyconsumption, area, and latency. Fixed-point arithmetic provides low-cost operators at theexpense of an unavoidable error between the ideal precision values and the finite precisionones. With a careful data word-length optimisation process, the designer can safely profitfrom the inherent trade-off between implementation cost and computation accuracy. Inthis chapter, we describe in detail the different stages comprising the fixed-point con-version process (i.e., dynamic range evaluation, word-length optimisation, and accuracyevaluation) and propose practical solutions to address each stage. | |
International
|
Si |
|
10.1049/PBCS040E_ch |
Book Edition
|
|
Book Publishing
|
IET Digital Library |
ISBN
|
978-1-78561-609-9 |
Series
|
|
Book title
|
Digitally Enhanced Mixed Signal Systems |
From page
|
289 |
To page
|
326 |