欢迎访问ic37.com |
会员登录 免费注册
发布采购

PDSP16510AA0GC 参数 Datasheet PDF下载

PDSP16510AA0GC图片预览
型号: PDSP16510AA0GC
PDF下载: 下载PDF文件 查看货源
内容描述: 单机FFT处理器 [Stand Alone FFT Processor]
分类和应用: 外围集成电路
文件页数/大小: 25 页 / 273 K
品牌: MITEL [ MITEL NETWORKS CORPORATION ]
 浏览型号PDSP16510AA0GC的Datasheet PDF文件第8页浏览型号PDSP16510AA0GC的Datasheet PDF文件第9页浏览型号PDSP16510AA0GC的Datasheet PDF文件第10页浏览型号PDSP16510AA0GC的Datasheet PDF文件第11页浏览型号PDSP16510AA0GC的Datasheet PDF文件第13页浏览型号PDSP16510AA0GC的Datasheet PDF文件第14页浏览型号PDSP16510AA0GC的Datasheet PDF文件第15页浏览型号PDSP16510AA0GC的Datasheet PDF文件第16页  
PDSP16510  
given previously.  
This is loaded at the sampling rate and then data is transferred  
to the PDSP16510 at a user defined rate. The time taken to  
load this external buffer must be at least equal to the sum of  
the time to transfer data in and out of the FFT processor and  
the transform time itself. When data blocks are overlapped by  
50% or 75%, no more than one half or one quarter of the block,  
respectively, must have been loaded in the same time. In the  
1024 point modes the dump time can be any user defined  
value, and need not be increased to allow for block overlap-  
ping. The dump time, however , does directly effect the  
maximum sampling rates which can be accommodated with-  
out loss of incoming data.  
The time taken to dump the transformed data must be no  
more than the load time, if continuous inputs are to be  
supported and I/O operations are concurrent with transforms.  
With block overlapping the dump time must be reduced to the  
time taken to load the partial block. This dump time must  
include four extra DOS strobes needed to prime the output  
circuitry when a transform is complete. These, in effect, can be  
added to the transform time such that with concurrent I/O and  
0%, 50%, or 75% overlapping;  
nS or (nS)/2 or (nS)/4 must be gtr than or equal to PK + 4W  
The maximum sampling rates for 1024 point transforms  
at any load and dump rate can be calculated from the following  
relationship:  
where n is the transform size, S is the input DIS period, P is  
the number of clock periods given in Table 4, K is the system  
clock period, and W is the DOS period which can be less than  
S if necessary. Note also that S must be synchronous to  
SCLK, and if an asynchronous ratio is required then a  
pdsp16540 input buffer should be used.  
When DIS and DOS are produced from a common source  
the minimum allowable sampling period must be increased to  
allow for the extra dumping time. Thus when DIS and DOS  
have equal periods and, for example, there is no overlapping;  
1024S or 512S or 256S > 1024B + PK + D  
for 0%, 50%, or 75% overlapping respectively. S, P, and K  
were defined opposite. B is the clock period in which data is  
read from the input buffer and loaded into the device, D is the  
total dump time allowing for the four extra DOS periods. The  
periods of the load and dump clocks cannot be less than the  
system clock period. The maximum sampling rates given in  
Table 5 assume that a 40 MHz I/O rate is used, and that all  
results are dumped.  
(n - 4)S must be greater than or equal to PK  
The maximum sampling rates given in Table 5 allow for the  
extra dumping time.  
MULTIPLE DEVICE SYSTEMS  
The load and dump operations are not concurrent with  
transforms in the 1024 point modes, and an external input  
buffer will be needed if loss of incoming data is to be avoided.  
In real time applications several devices may be used in  
parallel in order to increase the sampling rate, but not to  
increase the transform size. When all outputs are commoned  
together, and feed a single output processor, then the data  
dump time must always be less than or equal to the time taken  
to load the data block ( or 50% or 25% of the time with block  
overlapping ). In most configurations with block overlapping  
the dump rate requirements will limit the maximum input rate,  
if only one output processor is provided. This can be avoided  
if the system provides separate output processors for every  
device. The system clock used for internal calculations then  
ultimately imposes a limit on the maximum sampling rate  
possible.  
A multiple device system performing complex transforms  
with a single output processor is shown in Figure 9. The INEN/  
LFLG signals are used to co-ordinate the segmentation of  
data between devices. The in-active going edge of LFLG  
instigates the load procedure in the next device, and, since  
this edge can be programmed to occur either 25%, 50%, or  
100% through the load operation, it can cause the next device  
to commence loading before the previous one has finished. In  
this manner data block overlapping is achieved. When mul-  
tiple concurrent transforms are performed ( for example 4 x 64  
or 8 x 64 ) two LFLG transitions are sometimes needed to  
support block overlapping. This is fully explained in the section  
on Mode 1 sampling rates.  
Configuration  
Parameters  
Power on  
Reset  
Output  
Clock  
Complex Data  
Input  
IMAG  
REAL  
O/P  
S
MAG'  
PDSP16330  
PDSP16510  
PDSP16510  
PDSP16510  
PHASE  
CLK  
SCALE  
TAG  
IMAG  
REAL  
O/P  
S
DATA  
AVAIL'  
IMAG  
REAL  
O/P  
S
In any of the multiple device modes an INEN edge  
transition is needed to start a new load procedure when the  
previous one has finished. When the LFLG output from the last  
device is fed back to the INEN input of the first device,  
continuous transforms will be executed. This continuous  
sequence can be started by the rising edge of DEF if Control  
Register Bit 12 is set in the first device (see section on Loading  
INPUT CLOCK  
Fig 9. Multiple Device Configuration  
12