Intersil Corporation HSP50215 Datasheet

HSP50215
Data Sheet January 1999 File Number
Digital UpConverter
The HSP50215 Digital UpConverter (DUC) is a QASK/FM modulator/FDM upconverter designed for high dynamic range applications such as cellular basestations. The DUC combines shaping and interpolation filters, a complex modulator, and Timing and Carrier NCO’s into a single package. Each DUC can create a single FDM channel. Multiple DUC’s can be cascaded digitally for multi-channel applications.
The HSP50215 supports both vector and FM modulation. In vector modulation mode, the DUC accepts 16-bit I and Q samples to generate virtually any quadrature AM or PM modulation format. The DUC also has two FM modulation modes. In the FM with pulse shaping mode, the 16-bit frequency samples are pulse shaped/bandlimited prior to FM modulation. No bandlimiting filter follows the FM modulator . This FM mode is useful for GMSK type modulation formats. In the FM with bandlimiting filter mode, the 16-bit frequency samples directly drive the FM modulator. The FM modulator output is filtered to limit the spectral occupancy. This FM mode is useful for analog FM or FSK modulation formats.
The DUC includes an NCO driven interpolation filter, which allows the input and output sample rate to have anon­integer or variable relationship. This re-sampling feature simplifies cascading modulators with sample rates that do not have harmonic or integer frequency relationships.
The DUC offers digital output spectral purity that exceeds 85dB at the maximum output sample rate of 52 MSPS, for input sample rates as high as 300 KSPS.
A 16-bit microprocessor compatible interface is used to load configuration and baseband data. A programmableFIFO depth interrupt simplifies the interface to the I and Q input FIFOs.
4346.4
Features
• Output Sample Rates Up to 52 MSPS (48 MSPS Industrial); Input Data Rates Up to 3.25 MSPS
• I/Q Vector, FM, and Shaped FM Modulation Formats
• 32-Bit Programmable Carrier NCO; 30-Bit Programmable Symbol Timing NCO
• Programmable I and Q, 256 Tap, Shaping FIR Filters with Interpolation by 4, 8 or 16
• Interpolation Filter Up Samples Shaping Filter Output to Output Sample Rate Under NCO Control
• Processing Capable of >90dB SFDR
• Cascade Input for Multiple Channel Transmissions
• 16-Bit µProcessor Interface for Configuration and User Data Input
Applications
• Single or Multiple Channel Digital Software Radio Transmitters (Wide-Band or Narrow-Band)
• Base Station Transceivers
• Operates with HSP50214 in Software Radio Solutions
• Compatible with the HI5741 D/A Converter
• HSP50215EVAL Evaluation Board Available
Ordering Information
PART
NUMBER
HSP50215VC 0 to 70 100 Ld MQFP Q100 .14x20 HSP50215VI -40 to 85 s100 Ld MQFP Q100 .14x20
TEMP
RANGE (oC) PACKAGE PKG. NO
Block Diagram
CAS(15:0)
CASZ
IIN(15:0)
QIN(15:0)
REFCLK
SYNCIN
RST
WR
RD CE
C(15:0)
A(9:0)
CONTROL
3-422
FIFO
1-7 DEEP
MUX
FM
MOD
QIN(15:0) IIN(15:0) CF(31:0)
SF(29:0)
= µP CONTROL SIGNALS
IFIFO
QFIFO
MUX
SHAPING FILTER
IFM
Q FM
NCO
CAUTION: These devices are sensitive to electrostatic discharge; follow proper IC Handling Procedures.
GAIN
CTRL
MUX
http://www.intersil.com or 407-727-9207
FILTER
INTERPOLATION
SIN NCO
CF(31:0)
+
+
LIMITER
OUTPUT FORMATTER
COS
| Copyright © Intersil Corporation 1999
OUT(15:0)
OE OFM
SYNCOUT
FIFORDY SAMPCLK
Functional Block Diagram
CAS(1P5:0)
CASZ
3-423
REFCLK
RST
WR
RD CE
C(15:0)
A(9:0)
MOD (1:0)
IIN(15:0)
QIN(15:0)
CONTROL
RTH (2:0)
RST
FM
FIFO
MOD
IFIFO
QFIFO
MUX
1-7 DEEP
MOD (1:0) SR(29:0)
ICOEFFICIENTS(15:0) QCOEFFICIENTS(15:0) CF(31:0)
OUTGAIN (7:0) EN OUT DLYSEL
DS (3:0) IP (1:0)
STATUS
MUX
MOD (1:0)
COARSE PHASE
NCO
EnNCO
SHAPING
FILTER
I FM
Q FM
LIMITER
MOD (1:0)
SET CLOSE TO LIMIT
MUX
GAIN
CTRL
(-2dBFS)
0.8 MAX
INTERPOLATION
FINE
PHASE
FILTER
(-3dBFS)
0.707
SIN
COS
NCO
+
EN_OUT
+
SR CF EN_OUT MOD OUTGAIN FIFORDY IFIFOEMPTY QFIFOEMPTY
RTH EnNCO
SYNCPOL DS IP
OUT(15:0)
LIMITER
OUTPUT FORMATTER
OFM OE
HSP50215
FIFORDY SAMPCLK
SYNCIN
SYNCPOL SYNCSEL
SYNC
CW3 WR
SYNCOUT
= µP CONTROL SIGNALS
Pinout
HSP50215
100 LEAD MQFP
TOP VIEW
OFM
CASZ CAS0 CAS1
GND CAS2 CAS3 CAS4
CAS5 CAS6 CAS7
V
CC
GND
REFCLK
CAS8 CAS9
CAS10 CAS11
V
CC
CAS12 CAS13 CAS14 CAS15
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
OUT14
OUT15
99 98 97 96 95 94 93 91 89 87 85 84 83 818286889092100
OUT13
VCCOUT12
OUT10
OUT11
GND
OUT9
OE
OUT8
VCCOUT7
OUT5
OUT6
GND
OUT4
OUT3
OUT2
OUT1
80 79 78 77 76 75 74 73 72 71 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51
OUT0
C15 C14 C13
V
CC
C12 C11 C10 C9 GND C8
C7 C6 C5 V
CC
C4 SAMPCK C3 C2
GND C1 C0
SYNCOUT
32 33 34 35 36 37 38 40 42 44 46 47 48 50494543413931
A8
A0
A1
A2
A3
GND
FIFORDY
A4
A6
A5
CC
V
A7
GND
A9
WR
CE
CC
RD
RST
V
SYNCIN
3-424
HSP50215
Pin Descriptions
NAME TYPE DESCRIPTION
V
CC
GND - Power supply ground input.
C(15:0) I/O µP Bidirectional Data bus. The C(15:0) bus is used for loading the configuration data and sample vectors for mod-
A(9:0) I µP Address Bus. The A(9:0) bus is used for addressing the proper registers for loading the configuration data and
WR I µP Write Strobe. When CE is asserted, data on the C(15:0) data bus is loaded into the address location found on
RD I µP Read Control. When RD and CE are low, the data found in the address location defined by A(9:0) is routed to
CE I µP Chip Enable. Used to gate the WR and RD µP interface control signals.
FIFORDY O FIFO Ready. A FIFORDY assertion indicates that the I and Q FIFOs havereachedtheprogrammedFIFOdepthand
REFCLK I Reference Clock. REFCLK is the master clock for the DUC. All timing is relative to the REFCLK rising edge. The
CAS(15:0) I Cascade Input Bus. This input bus is used to cascade multiple parts by routing the digital modulated signal from
CASZ I Cascade Input Bus Zero. When CASZ is asserted (pulled high), the part places zeroes on the CAS(15:0) data path.
- +5V Power supply input.
ulation. C15 is the MSB.
sample vectors for modulation. A9 is the MSB.
the A(9:0) busontherisingedgeoftheWRsignal.Insomecases,thereisaninternalsynchronizationtothemaster clock that must be completed before the next data is written. See the µP interface section for more information.
the C(15:0) µP data bus on the next rising edge of REFCLK.
more samples are required to maintain that FIFO depth.
frequency of the reference clock is denoted f
one DUC into the output summer of a second DUC. CAS(15:0) is 2’s complement format and is sampled on the rising edge of REFCLK. CAS15 is the MSB.
CASZ is asynchronous (not registered) to REFCLK and should not be changed on the fly.When unused, pull high with a pull up resistor (~22k).
, and is the rate at which data is output from the part.
CLK
OUT(15:0) O Output Data Bus. OUT(15:0) contains the digital modulated DUC output samples and is updated on the rising edge
of the REFCLK. OUT15 is the MSB.
OFM I Output Data Bus Format. When OFM is asserted (pulled high), the output bus format is 2’s complement. When not
asserted, the output format is offset binary. The OFM input is asynchronous (not registered) to REFCLK and should not be changed on the fly.
OE I Output Data Bus Enable.When OE is asserted (dropped low), the output data bus OUT(15:0) is enabled. When OE
is not asserted (pulled high), the output data bus OUT(15:0) is placed in the high impedance state.
SYNCIN I Sync Input. The SYNCIN input is used to synchronize the processing of multiple parts. The SYNCOUT of one part
acts as a master and is connected to the SYNCIN of all of the DUC’s that are to by synchronized. The DUC can be programmed so that either rising or falling edge of this signal initiates the processing.
SYNCOUT O Sync Output. The SYNCOUT output is used to synchronize the processing of multiple parts. The SYNCOUT of one
part acts as a master, and is connected to the SYNCIN of all of the DUC’s that are to be synchronized.
SAMPCLK O Sample Clock. This clock is provided to the data source to indicate when data is being transferred from the FIFO to
the shaping filter. The SAMPCLK output is generated by the sample rate NCO when the digital filter takes a new sample. It has approximately 50% duty cycle. The sample is taken on the high-to-low transition. SAMPCLK may be used instead of FIFORDY.
RST I Reset. When the RST input is asserted (dropped low), the DUC is reset and all processing halts. The DUC may
also be reset on µP command. Processing remains halted until a sync is generated either by µP command or assertion of SYNCIN. See the Reset section details of the specific functions halted by this control signal.
3-425
HSP50215
Functional Description
The HSP50215 Digital UpConverter (DUC) converts digital baseband data into modulated or frequency translated digital samples. The DUC can be configured to create any quadrature amplitude shift-keyed (QASK) data modulated signal, including QPSK, BPSK, and m-ary QAM. The DUC can also be configured to create both shaped and unfiltered FM signals. A minimum of 16 bits of resolution is maintained throughout the internal processing.
The DUC is configured via the 16-bit microprocessor data bus, using the address bus and signals. Configuration data that is loaded via this bus includes the 30-bit Sample Rate NCO center frequency, the 32-bit Carrier NCO center frequency, the modulation format, gain control, FIFO control, reset control and sync control. The I and Q baseband channels each have a 256 tap FIR filter whose coefficients and configuration are also programmed via the µP interface. Similarly, the control signals for the I and Q channel interpolation filters are programmed via the µP interface. Once the operational configuration for the device has been set, the 16-bitµP interfaceis used to input the I and Q data into the associated FIFOs.
The FIFOs provide the data interface between the µP and either the FM modulator or the shaping filters. Multiplexers route the I data to the FM modulator in the FM with bandlimiting filter mode. Both I and Q are routed to the 256 tap FIR shaping filters in the QASK mode. The shaping filter serves to both shape and interpolate the sample rate to 4, 8, or 16 times the input sample rate. The I shaping filter output can also be routed to the FM modulator for the FM with pulse shaping mode. Multiplexers select either the FM modulator output or the shaping filter output to be scaled and routed to the interpolation filters.
RD,WR and CE control
A(000)
WR
IIN(15:0)
R E G
>
WR
ZERO’S
A(2:0)
RTH(2:0)
FM ENABLED
QIN(15:0)
R E G
>
WR
A(001) WR
DFF1 DFF2 DFF3 DFF4
R
R
R E G
>
>
WRITE SHIFT ENABLE
R
R
E
E
G
G
>
>
R
R
E
E
G
G
>
>
WRITE SHIFT ENABLE
DFF1 DFF2 DFF3
R E G
>
E
G
>
R E G
>
01
R E G
>
R E
G
>
E
G
COMP
COMP
8:1 MUX
>
R E G
>
8:1 MUX
R E G
>
R E
G
>
R
ALL REGISTERS
E
ARE CLOCKED AT
G
REFCLK UNLESS SHOWN OTHERWISE
R E G
>
DFF
R E G
>
DFF4
R E
G
>
R E G
>
R E G
>
R E G
>
IFIFO(15:0)
FIFORDY
QFIFO(15:0)
R E G
>
The I and Q interpolation filters allow a non-integer increase in sample rate, up to the reference clock r ate. The interpolation filter output data is upconverted or modulated by the Carrier NCO and multipliers. The modulated signal is added to modulated inputs from other cascaded DUC’s. The output formatter sets the output buffer state and the output data format.
Programmable FIFO
The Programmable FIFOs provide a data storage and interface between the microprocessor data write holding register and the shaping filter or the FM modulator. Signal routing out of the FIFO is set by the modulation format. Each FIFO has seven 16-bit registers. Figure 1 shows the conceptual details of the I and Q FIFOs.
3-426
FIGURE 1. I AND Q FIFO BLOCK DIAGRAM
WR
REFCLK
DLY DATA
DFF 1
DFF 2
DFF 3
DFF 4
WR SHFT EN
REG1
FIFORDY
FIFO NEEDS MORE DATA
1234
FIGURE 2. FIFORDY AND DATA DELAY TIMING
FIFO NEEDS MORE DATA
HSP50215
Data enters the FIFO with a write command to either Control Word 0 (for I data), or Control W ord 1 (for Q data). This transfersdata from the microprocessor holding register into the first 16-bit register of the FIFO. The FIFO counter is incremented every time data is written into the FIFO. F our REFCLK periods are required from the rising edge of a WR signal before another WR rising edge can occur, (i.e ., bef ore data can once again be written into either the I or Q FIFO). This limits the maximum data input (write) rate to 52MHz/4 = 13MHz.
NOTE: The Write rate is not the parameter that determines the
maximuminput rate, the shaping filter is. Themaximum in­put for the shaping filter is 52MHz/(IP)(DS), which is 52MHz/16 =3.25 MHz for a minimal shaping filter (DS = IP=
4). See the Shaping Filter Section for more details.
The timing details of these FIFO registers are shown in Figure
2. While the data for the I and Q inputs are independent, the Write cycle limitations of the FIFO constrains the maximum input symbol rate of quadrature symbols (both I and Q data) as noted.
When the Shaping Filter requires another sample of data, a request is made to the FIFO for data and the FIFO counter is decremented. Figure 3 indicates the timing of a request for data from the Shaping filter to the actual appearance of data at the FIFO output. The FIFO has circuitry for detecting an empty FIFO as well as a full FIFO. An “empty” FIFO detection causes “zero” data to be entered into the shaping filter. A “full” FIFO detection prevents data from being pushed out of the FIFO before the filter requests it.
NOTE: Do not write to a full FIFO. Writing to a full FIFO is an er-
ror condition and the part will be reset to prevent trans­mission of erroneous data over the air.
A programmable FIFO depth threshold sets when the FIFORD Ysignal is asserted, alerting the data source that more data is required. The
FIFORD Y signal assists a data source in maintaining the desired FIFO data depth. Control Word 18, bits 0-2 are used to set the data FIFO depth threshold for both I and Q inputs.
NOTE: SAMPCLK may be used instead of FIFORDY to indicate
that data is transferred from the FIFO to the shaping filter. See the Pin Description Tab le.
.
with post-modulation filtering, and 10:FM with pre­modulation pulse shaping. These modulation paths are defined in the following subsections.
Modulation Mode 00 - QASK
This modulation mode configures the PUC as a BPSK, QPSK, OQPSK, MSK or m-QAM modulator. The filter configuration is shown in Figure 4. The data FIFO outputs are routed to the shaping filters. Here the samples are interpolated by 4, 8, or 16 and shaped using a FIR filter with up to a 256 taps. The filter impulse response can span 4-16 input samples. A half (input) sample delay can be set in the I channel for implementing OQPSK modulation. The output of the shaping filter is routed through a gain adjust multiplier and into the interpolation filter. The interpolation filter interpolates by a factor set in the resampling NCO. The output of the interpolation filter is at the master clock frequency, REFCLK. The samples are then mixed with the carrier L.O. for quadrature upconversion. The output is then summed with the cascade input signal, saturated (in the case of overflow), and formatted for output.
I
SHAPING
Q
FILTER
INTERPOLATION
FILTER
FIGURE 4. QASK
TO MODULATOR
Modulation Mode 01 - FM with BANDLIMITING Filter
This mode configures the PUC as an FM modulator with post-modulation filtering. This mode provides for FSK and FM modulation schemes. In this mode, the I input samples drive the frequency control section of a quadrature NCO to produce a zero IF FM signal. The FM quadrature signals are then routed to the shaping FIR filter and into the interpolation filter for bandlimiting and interpolation up to the master clock rate as shown in Figure 5. The quadrature filtered FM signals are then upconverted to the carrier frequency by the carrier NCO and mixers. The output is then summed with the cascade input signal, saturated (in the case of overflow), and formatted for output. Note that pulse shaping in this mode must be provided prior to the PUC.
REFCLK
EnFIFO
IFIFO
FIGURE 3. FIFO DATA AND ENABLE TIMING
Data Modulation Path
Three data path options are provided, one for each modulation format. The modulation format is selected using Control Word 16 (See the Microprocessor Write section). Control Word 16 bits (1:0) are defined as: 00:QASK, 01:FM
3-427
I
FM
MODULATOR
FIGURE 5. FM WITH BANDLIMITING
SHAPING
FILTER
INTERPOLATION
FILTER
TO
Modulation Mode 10 - FM with Pulse Shaping
This mode configures the PUC as an FM modulator with pre­modulation baseband pulse shaping. The data from the FIFO (I channel only) is routed to the FIR shaping filter. The FIR shaping filter output drives the frequency control section
MODULATOR
HSP50215
of a quadrature NCO to produce a zero I.F.FM signal. These FM modulated quadrature samples are then up sampled in the interpolation filter to the output sample rate. The baseband modulated signal is then upconverted to the carrier frequency by the carrier NCO and mixers. The output is then summed with the cascade input signal, saturated, and formatted for output.
I
SHAPING
FILTER
FIGURE 6. FM WITH PULSE SHAPING
FM
MODULATOR
INTERPOLATION
FILTER
TO
In Mode 10, the amplitude out of the shaping filter needs to be limited in order to prevent frequency excursions that cannot be filtered out in the interpolation filter. The quality of the FM signal is affected by the amplitude slew rate out of the shaping filter.As a rule of thumb, limiting this slew rate to less than 1/8 the sample rate will minimize this distortion.
FM Modulator
The FM modulator provides for frequency modulation of the carrier center frequency by the PUC input data. The FM modulator is driven either directly by the PUC I input (Mode
1) or by the output of the FIR shaping filter (Mode 2). The input data to the FM Modulator, is defined as dφ(n)/dt, where φ(nT) is the phase of a theoretical sinusoid described by:
sn() A (cos φ nT()]+ j sin φ nT()[]); A 1 in Modulator[=
(EQ. 1)
modulator. The maximum phase step that can occur in one clock is ±180 degrees. Table 1 provides the change in phase weighting of the input bits.
TABLE 1. FM MODULATOR TRANSFER FUNCTION
dφ(nT)/dt DEGREES/SAMPLE
1000 0000 0000 0000 -180 0000 0000 0000 0000 0 0111 1111 1111 1111 ~+180
MODULATOR
Shaping Filter
The shaping filter provides the necessary pulse shaping required on the input data to implement various quadrature ASK and shaped FM modulation formats. Two identical shaping filters (one each for the I and Q channels) are provided. The filters can implement a 4-16 input sample span impulse response using up to 256 taps with 16 bits of resolution in the coefficients.
The range of valid digital values for the coefficients is from 8001 to 7FFF. The value 8000 is not allowed. The coefficient format is 2’s complement. The span of the Impulse response of the polyphase filter can be from 4-16 samples. The desired sample span value minus one is programmed into the Data Samples (DS) field in Control Word 19, bits 2-5. The filter has a programmable interpolation rate (IP) of 4, 8, or 16. This interpolation rate is programmed by Control Address 19, bits 0 and 1. Thus, the required number of coefficients (or filter span) becomes
Figure 7 illustrates the conceptual design of the FM modulator. The input to the FM modulator, dφ(n)/dt, is integ rated via the carrier NCO accumulator. The NCO accumulator output represents phase and is used to address a SIN/COS generator, synthesizing a sinusoid of the form described in Equation 1. The phase accumulator feedback of the NCO is 16 bits and sixteen bits of the phase word are routed to the SIN/COS generator. Sixteen bits of resolution are provided on the Sine and Cosine outputs.
16
16
dφ(nT)/dt
EnNCO
MODE
1 OR 2
FIGURE 7. FM MODULATOR BLOCK DIAGRAM
R E G
>
φ(nT)
16
16
ROM
SIN/COS
COS[φ(nT)]
SIN[φ(nT)]
The transfer function of the FM modulator is defined by the change in degrees per sample value, dφ(nT)/dt,where dφ(nT)/dt is a 16-bit, twos complement, fractionally notated frequency control word with a range from -F +F
SAMP
/2. F
is defined as the sample rate into FM
SAMP
SAMP
/2 to
# Coefficients = (DS)(IP)
(EQ. 2)
with 256 being the maximum number of coefficients. Note that
REFCLK DS()IP()fS()>
where f
S
is the input sample rate of the shaping filter. For a
(EQ. 3)
16 input sample impulse response span, the total impulse response is 64, 128 or 256 filter taps for interpolation rates of 4, 8 or 16, respectively. The filter structure precludes coefficient re-use for symmetric filters, so both asymmetric and symmetric filters have up to 256 taps available and are loaded in identical manner.
The maximum input sample rate is:
fS=f
CLK
where f
[IP() DS()]
is the frequency of the reference clock, IP is the
CLK
(EQ. 4)
shaping filter interpolate rate; and DS is the number of data samples in the filter span. For example,if f
= 52MHz, the
CLK
filter span is 16 samples, and the interpolation rate is 16, then the maximum input sample rate, f
is 52/256 = 203kHz.
S
Table 2 shows severalexamples of calculations for FIR input sample rates based on master reference clock rate, number of data samples, and interpolation rate.
3-428
Loading...
+ 14 hidden pages