 |
|
| If you can't view the Datasheet, Please click here to try to view without PDF Reader . |
|
|
|
| If you can't view the Datasheet, Please click here to try to view without PDF Reader . |
|
|
| Datasheet File OCR Text: |
| cml semiconductor products tetra baseband processor fx980 ? 1997 consumer microcircuits limited d/980/3 november 1997 1.0 features advance information rrc filters for both tx and rx 4 x10-bit d-a and 4 input 10-bit a-d p p /4 dqpsk modulation transmit output power control 2x 13-bit resolution sigma delta d-a low power 3.0 - 5.5volt operation 2x 16-bit resolution sigma delta a-d effective power down modes 1.1 brief description this device is intended to act as an interface between the analogue and digital sections of a digital radio system, and performs many critical and dsp-intensive functions. the chip is designed with the necessary capability to meet the requirements for use in both mobile and base station applications in terrestrial trunked radio (tetra) systems. the transmit path comprises all the circuitry required to convert digital data into suitably filtered analogue i and q signals for subsequent up-conversion and transmission. this includes digital control of the output amplitudes, digital control of the output offsets and fully programmable digital filters: default coefficients provide the rrc response required for tetra. the receive section accepts differential analogue i and q signals at baseband and converts these into a suitably filtered digital form for further processing and data extraction. a facility is provided for digital offset correction and the digital filters are fully programmable with default coefficients providing the rrc response required for tetra. auxiliary dac and adc functions are included for the control and measurement of the rf section of the radio system. this may include afc, agc, rssi, or may be used as part of the control system for a cartesian loop.
tetra baseband processor fx980 ? 1997 consumer microcircuits limited 2 d/980/3 contents section page 1.0 feature s ....................................................................................................................................... 1 . 1.1 brief descriptio n ...................................................................................................................... . 1 1.2 block diagra m ........................................................................................................................... . 3 1.3 signal lis t ................................................................................................................................... . 4 1.4 external component s ............................................................................................................ . 6 1.5 general descriptio n ............................................................................................................... . 7 1.5.1 connection and decoupling of power supplie s .................................................................. . 7 1.5.2 tx data pat h ........................................................................................................................... . 8 1.5.2.1 modulato r ............................................................................................................................. . 8 1.5.2.2 filter s ................................................................................................................................... . 8 1.5.2.3 gain multiplie r ....................................................................................................................... . 8 1.5.2.4 offset adjus t ......................................................................................................................... . 8 1.5.2.5 sigma-delta d-a converters and reconstruction filter s ........................................................ . 8 1.5.2.6 phase pre-distortio n .............................................................................................................. . 8 1.5.2.7 ramping output amplitud e ................................................................................................... . 8 1.5.3 rx data pat h ........................................................................................................................... . 9 1.5.3.1 anti-alias filtering and sigma-delta a-d converter s ............................................................. . 9 1.5.3.2 filter s .................................................................................................................................... . 9 1.5.3.3 offset register s .................................................................................................................... . 9 1.5.3.4 i and q channel gain ............................................................................................................ . 9 1.5.4 auxiliary circuit s ................................................................................................................... . 9 1.5.4.1 10-bit dac s .......................................................................................................................... . 9 1.5.4.2 10-bit ad c ............................................................................................................................ . 9 1.5.4.3 power ramping and contro l ................................................................................................ . 10 1.5.5 irq functio n ........................................................................................................................ . 10 1.5.6 serial interfac e ..................................................................................................................... . 10 1.5.6.1 command interfac e ............................................................................................................. . 11 1.5.6.2 command read interfac e ................................................................................................... . 12 1.5.6.3 rx data interfac e ................................................................................................................ . 12 1.5.6.4 transmission of dat a .......................................................................................................... . 12 1.5.6.5 command control serial wor d ............................................................................................ . 13 1.5.7 register descriptio n ............................................................................................................ . 14 1.5.7.1 register and access point summar y .................................................................................. . 16 1.6 application note s .................................................................................................................. . 33 1.6.1 genera l .................................................................................................................................. . 33 1.6.2 transmitte r ............................................................................................................................ . 33 1.6.3 receive r ................................................................................................................................ . 33 1.6.4 timin g .................................................................................................................................... . 33 1.7 performance specificatio n ................................................................................................ . 33 1.7.1 electrical performanc e ........................................................................................................... . 33 1.7.2 packagin g .............................................................................................................................. . 33 note: as this product is still in development, it is likely that a number of changes and additions will be made to this specification. items marked tbd or left blank will be included in later issues. information in this data sheet should not be relied upon for final product design. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 3 d/980/3 1.2 block diagram figure 1 block diagram tetra baseband processor fx980 ? 1997 consumer microcircuits limited 4 d/980/3 1.3 signal list l6 package 44 plcc package # signal description pin no. pin no. name type 15 mclk i/p master clock input (typically 9.216mhz) 16 sclk o/p serial interface clock 17 cmddat bi command serial interface data 18 cmdfs i/p command serial interface frame 19 cmdrddat o/p command serial interface read data 20 cmdrdfs o/p command serial interface read frame 11 rxdat o/p receive serial interface data 12 rxfs o/p receive serial interface strobe 23 n_irq o/p interrupt request 14 n_reset i/p chip reset 24 scansel i/p scan select (normally tied low) 25 itxp o/p transmit "i" channel, positive output 26 itxn o/p transmit "i" channel, negative output 30 qtxp o/p transmit "q" channel, positive output 29 qtxn o/p transmit "q" channel, negative output 42 irxp i/p receive "i" channel, positive input 41 irxn i/p receive "i" channel, negative input 38 qrxp i/p receive "q" channel, positive input 37 qrxn i/p receive "q" channel, negative input 43 auxadc1 i/p auxiliary adc channel 1 44 auxadc2 i/p auxiliary adc channel 2 1 auxadc3 i/p auxiliary adc channel 3 2 auxadc4 i/p auxiliary adc channel 4 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 5 d/980/3 1.3 signal list (continued) l6 package 44 plcc package # signal description pin no. pin no. name type 10 auxdac1 o/p auxiliary dac channel 1 9 auxdac2 o/p auxiliary dac channel 2 8 auxdac3 o/p auxiliary dac channel 3 7 auxdac4 o/p auxiliary dac channel 4 36 bias1 bi analogue bias level. this pin should be de- coupled to v ssb. 35 bias2 bi analogue bias level. this pin should be de- coupled to v ssb . 32 v cc1 power i channel analogue positive supply rail. this pin should be de-coupled to v ss1. 33 v cc2 power q channel analogue positive supply rail. this pin should be de-coupled to v ss2. 34 v cc3 power analogue bias positive supply rail. levels and voltages are dependent upon this supply. this pin should be de-coupled to v ssb. 6 v dd1 power auxiliary analogue positive supply rail. this pin should be de-coupled to v ssa. 3,21 v dd power digital positive supply rail. this pin should be de-coupled to v ss. 27,40 v ss1 ground i channel analogue negative supply rail. 28,39 v ss2 ground q channel analogue negative supply rail. 31 v ssb ground analogue bias negative supply rail. 5 v ssa ground auxiliary analogue negative supply rail. 4,13,22 v ss ground primary digital negative supply rail. notes: i/p = input o/p = output bi = bi-directional tetra baseband processor fx980 ? 1997 consumer microcircuits limited 6 d/980/3 1.4 external components rx inputs when using the internal anti-alias filter, the following is suggested example values: r1 = 220 w c1 = 1.5nf (r1, c1 precise values are not critical) (-3db at 240khz) r2 = 408 w c2 = 3.9nf (r2 x c2 time constant should be preserved) (-3db at 50khz) when not using the internal anti alias filter, it is suggested that the user should follow the guidelines in section 1.5.3.1. in both cases, there should be at least one filter pole close to the chip inputs. figure 2a recommended external components - rx inputs tx outputs example values: r3 = 220 w c3 = 1nf decoupling capacitors should be employed as detailed in section 1.5.1 figure 2b recommended external components - tx outputs agc irxp irxn qrxp qrxn filter 1 r2 r2 c2 r2 r2 c2 r1 r1 r3 r1 c1 c1 r1 c3 itxp itxn qtxp qtxn r3 c3 r3 r3 c3 c3 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 7 d/980/3 1.5 general description 1.5.1 connection and decoupling of power supplies optimum performance from the fx980 can only be obtained by the use of adequate decoupling and the separation of analogue and digital signals, including the use of separate ground planes. printed circuit board layout should follow the recommendations shown in figure 3. figure 3 recommended decoupling components tetra baseband processor fx980 ? 1997 consumer microcircuits limited 8 d/980/3 1.5.2 tx data path the features described below give a high degr ee of flexibility for the user to compensate in the baseband processing for non-ideal performance in the if, rf and rf linear amplifier sections. 1.5.2.1 modulator this takes the 2-bit symbols, performs a gray code conversion and uses a recursive adder to generate a 3-bit code representing the 8 possible phase states. a look up table provides the digitally encoded i and q values for each phase state. the modulator function can be by-passed if required; in this case the 3-bit code representing the 8 possible phase states which are passed to the look up table is provided directly via the serial interface. 1.5.2.2 filters digital filtering is applied to the data from the modulator; the coefficients are set as default to give a root raised cosine response with roll-off factor of 0.35. these fir filters operate at 8x the incoming symbol rate and are configured, for each channel, as two filters in cascade: the first filter has 79 taps and the second filter has 49 taps. the first filter is used to enhance stop-band rejection and act as a sampling correction filter and the second filter provides the primary shaping. coefficients for the filters may also be downloaded to the device via the serial interface; this gives the opportunity, if required, to fine tune the frequency response of a complete system so as to minimise the ber or to use the device in other applications. the filters can also be by-passed if required. 1.5.2.3 gain multipl ier this circuitry allows independent external control of the digital amplitudes in the i and q channels to 12 bits of resolution. extra circuits allow a mode of operation which will enable linear ramping up to a maximum value, stay at this value for a specified duration, then ramp back down to zero. the maximum value for each channel, the duration at maximum, the ramping up rate and the ramping down rate are all programmable via the serial interface. 1.5.2.4 of fset adjust offset registers allow any offsets introduced in the analogue sections of the transmit path to be corrected digitally via the serial interface. the offset adjust has a resolution of 1 lsb and a maximum value of 0.25x full scale. 1.5.2.5 sigma-delta d-a converters and reconstruction filters the converters are designed to have low distortion and >80db dynamic range. these 3rd order converters operate at a frequency of 128x symbol rate so as to over-sample the data at their inputs a further 16 times. the reconstruction filters are 5th order, switched capacitor, low pass filters designed to work in conjunction with an external rc. 1.5.2.6 phase pre-distortion a further feature allows the user to compensate for a non-orthogonal carrier phase in the external quadrature modulator by adding a programmable fraction of up to 1/8 of the filtered i and q channel signals to each other immediately prior to the dac input. 1.5.2.7 ramping output amplitude a facility is provided to allow linear ramping of the outputs. this is accomplished, if enabled, by multiplying the gain multiplier words by the ramping control register (rcr) value. the rcr is a 12-bit word, representing a value from 0 to 1, which is designed to increment by an amount (inc) until its maximum value. this value is held until a number of symbol times from the start of transmission (trd) when rcr decrements by an amount (dec) until zero. inc, dec and trd are all 12-bit words input via the serial interface prior to the start of a transmission. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 9 d/980/3 1.5.3 rx data path 1.5.3.1 anti-alias filtering and sigma-delta a-d converters the sampling frequency of the sigma-delta a-d is 128x symbol rate. the high oversampling rate relaxes the design requirements on the anti-alias filter. however, to achieve optimum performance the anti-alias filter must reject the sampling frequency to about -110db, of which at least 40db must be provided externally. additionally, in order to ease the complexity of the subsequent digital filters, there is a further requirement that the anti-alias filter suppress 8x symbol rate to about -30db. the on-chip anti-alias filter is designed to achieve this when used in conjunction with some external filtering. if required, the on-chip anti-alias filter can be by-passed and powered down, although external anti- aliasing must then be provided. the 4th order sigma-delta a-d converters are designed to have low distortion and >96db dynamic range. the baseband i and q channels must be provided as differential signals; this minimises in-band pick up both on and off the chip. 1.5.3.2 filters digital filtering is applied to the data from the sigma-delta a-d converters; the default coefficients are set to give a root raised cosine response with roll-off factor of 0.35. these fir filters are configured, for each channel, as three filters in cascade. the first filter gives sufficient rejection at 8x symbol rate to permit decimation at that frequency (note that -30db is provided by the primary anti-alias filters). the second filter has 63 taps and is used to enhance stop-band rejection. the third filter has 49 taps and provides the primary shaping requirements. coefficients for the second and third filters are programmable via the serial interface. this gives the opportunity, if required, to fine tune the frequency response of a complete system so as to minimise the ber or to use the device in other applications. the filters can also be by-passed if required, by setting the centre coefficient to maximum and all other coefficients to zero. 1.5.3.3 offset registers system generated offsets may be removed by control of the offset register via the serial interface. 1.5.3.4 i and q channel gain programmable gain modules are provided in both i and q channels. these blocks allow the user to adjust the dynamic range of the received data within the digital filters, thus optimising the filter signal to noise performance for a range of levels at the rx input pins. the two channels are independently programmable. this enables differential gain corrections to be made within the digital domain. 1.5.4 auxiliary circuits 1.5.4.1 10-bit dacs four 10-bit dacs are provided to assist in a variety of control functions. the dacs are designed to provide an output as a proportion of the supply voltage, depending on the digital input. they are monotonic with an absolute accuracy of better than 1%. control and data for these come via the serial interface. 1.5.4.2 10-bit adc a 10-bit adc is provided to assist in a variety of measurement and control functions. the adc is designed to produce a digital output proportional to the input voltage; full scale being the positive supply. it is monotonic with an absolute accuracy of about 1%. an input multiplexer allows the input to be selected from one of four sources. control and digital data output is via the serial interface. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 10 d/980/3 1.5.4.3 power ramping and control one of the dacs has an additional feature which enables a set of values to be sequenced out at a pre-selected frequency. this is aimed at enabling power ramping of a rf output with a suitable profile. the sequence may be reversed for power down. the sequence of values is stored in a dedicated ram, which can be loaded via the serial interface. 1.5.5 irq function an interrupt request (irq) pin is provided for asynchronous communication with an external processor. the irq (asserted low) will be asserted when any of the error or user information flags are activated by an internal operation. some examples of operations which may generate an irq are: 1. an attempt by the user to write to a full tx data-input fifo 2. an attempt is made by the tx to read from the tx data-input fifo when it is empty. 3. an internal arithmetic overflow has occurred in an fir filter. the irq feature may also be used to establish the phasing of the received i and q channel data from the rxdat serial port should synchronisation be lost for any reason. the cause of the irq can be obtained by reading the error flags register. all possible causes of an irq are masked on reset. mask status can be altered by writing to the irq mask register. note that default coefficients and settings have been optimised to maximise performance and should not cause arithmetic overflows. however, use of non-default coefficients, large offset corrections or large tx phase adjustments may cause problems, which can be corrected by scaling down coefficients or via the gain multiplier feature. 1.5.6 serial interface all digital data i/o and control functions for the fx980 are via the serial interface. it is expected that the fx980 will be used in conjunction with a dsp and/or other processor. the device has three serial interface ports, each port is based on the industrial standard three wire serial interface. this interface allows communication with standard dsp ics using a minimum of external components. the three serial interface ports are: cmd command port, generally this is an input port receiving commands and data from the host, but may also be configured as a bi-directional i/o interface. cmdrd command read port, an output port to send command read data back to the host. read data is only sent on this port in response to a read command. rxdata receive data port, an output port to send receive data back to the host. data is only present on this interface when the rx data path is active. this port may also be configured as the cmdrd port. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 11 d/980/3 functions performed by the serial interface include: power up or down and optional bypassing of selected blocks setting digital filter coefficients loading ramp up and ramp down increments and burst lengths for tx loading and transmitting data loading offset correction, gain multiplier and phase adjustment registers enabling/disabling of output via the rx serial interface vary sampling time for rx data relative to the symbol (144khz) clock. loading data into auxiliary dacs initiating conversions using auxiliary adcs and reading results writing data to, and reading data from, the waveform generation sram power ramping time step control the three interfaces consist of the following signal pins: sclk output serial clock pin. this pin is common for all three interfaces. cmddat in/out command port data pin. this pin is by default an input, but may be configured as an open drain bi-directional pin. cmdfs input command port frame sync pin. this pin is used to mark the first bit in a serial frame. cmdrddat output command read port data pin. this pin only has active data on it in response to a read command. cmdrdfs output command read port frame sync pin. this pin is used to mark the first bit in a serial frame. rxdat output receive data port data pin. this pin is only active when the rx data path is active. rxfs output receive data port frame sync pin. this pin is used to mark the first bit in a serial frame. note: all frame sync strobe signals are actually coincident with the last bit of a dataframe. see figures 4 and 5 for further details. 1.5.6.1 command interface a serial command word consists of a 16-bit frame. each frame is marked by an active frame sync event which precedes the msb bit. a command word can be either a control word or a transmit data word. msb lsb r / w address data 15 14 8 7 0 command control serial word msb lsb 1 tx data address u/d 4/1 tx data 15 14 10 9 8 7 0 command transmit data serial word tetra baseband processor fx980 ? 1997 consumer microcircuits limited 12 d/980/3 1.5.6.2 command read interface command read data is either output on one of the serial read ports, or driven out in the last 8 bits (data field) on the cmd port. when command read data is output on a serial read port, the read address is put in the most significant half of the word, and the read data in the least significant half. msb lsb 0 read address data 15 14 8 7 0 command read serial word 1.5.6.3 rx data interface the rx data interface is used only for output of the i and q received data, unless it is operating in the mode where cmdrd data is directed to it. when data reception is enabled, i and q received data will be output at either 8x or 4x the symbol rate, under control of command register rxsetup2 . (see section 1.5.7). this is achieved by reducing the serial interface clock rate from mclk/2 to mclk/4 and discarding alternate data samples under control of command registers configctrl1 and rxsetup2 . 16-bit i and q data words are output at the rx data interface, i data and msb first (by default), on the rising edge of sclk. 1.5.6.4 transmission of data the address of the tx fifo is given consecutive locations ($0x04-$0x07), which allows the address bits a1 and a0 (bits 11 and 10) of the command transmit data serial word to be utilised as transmit control functions. data to be transmitted can be in either one or four (2-bit) symbol blocks, which are subsequently modulated into the dqpsk constellation, or in 3 -bit words, which map directly into constellation points according to the table shown below. 3 bit code 000 001 010 011 100 101 110 111 i q 1 0 0.7071 0.7071 0 1 -0.7071 0.7071 -1 0 -0.7071 -0.7071 0 -1 0.7071 -0.7071 constellation map the user initiates a transmit frame by asserting the txen bit in the txsetup register. however, internal transmission of the data will wait until specific conditions have been met. firstly, a valid data word must be written into the fifo with the txrampen bit of the txsetup register asserted. secondly, the internal symbol clock must be active. therefore there is a variable delay between asserting the txen bit and transmission starting. the user may poll the txpathen bit of the txfifostatus register to establish when transmission has started, and in this case the active state of txpathen in high. in general, the user will wish to know when the transmit frame has completed. this is indicated by txpathen returning low. to relieve the user of polling overheads when waiting for tx frame completion, an interrupt can be set up to occur on the transition of the txpathen bit from high to low. in such circumstances, the interrupt activation state of the txpathen can be considered low. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 13 d/980/3 two control bits are associated with each data transmission word. one controls the format of the word and the other initiates and terminates a transmission cycle. this close association enables precise control of the transmission frame. to relieve the user of the need to synchronise each txdata write with the internal transmit cycle, transmit data words are written into an internal 4 -word-deep fifo. symbols or constellation points are then read as needed from this fifo. it is necessary to make sure that there is always a word to be read, and also that the fifo is never written to when full. this may be accomplished by using one of two data interlock mechanisms. data interlock mechanisms there are two possible transmission data interlock mechanisms. it is recommended that the user should always use one of these methods. software polling. serial clock when ready. software polling requires the user to first check that the fifo is not full before writing each txdata word. this may be accomplished by inspecting the relevant fifo status bits before writing one or more txdata words. the serial clock when ready method is a hardware interlock mechanism (enabled by setting the txhandshakeen bit of configctrl1 register active). the mechanism allows the user to write txdata words without doing any fifo checks: the hardware handshake is implemented by stopping the serial port clock when the fifo is full. to prevent a serial port lockout-condition, the handshake is only enabled once the transmission frame has been initiated and is automatically disabled at the end of a frame. this mechanism should be used with care, because stopping the clock will freeze all other serial port transfers (the serial port clock sclk is common to all three serial ports), including access to auxiliary data converters and receive data. power ramping and frame interlock the rampup bit in the txdata word is used to control both the power ramping function and the frame activation. to start a transmission frame, a transmission word is written with the rampup bit active. all subsequent txdata words prior to frame termination must also have this bit active. the frame is terminated by writing transmit data words with the rampup bit inactive. subsequent txdata words must also have this bit inactive, until initiation of a new frame is required. while the power ramping is active (up or down) the user must supply transmission symbols or valid constellation points. once the ramp down operation has completed, all subsequent txdata writes with the rampup bit inactive will be ignored. 1.5.6.5 command control serial word a command word either directly accesses an internal register for a read or write operation, or addresses a memory access point to indirectly access a block of internal memory. for test purposes all registers that can be written may also be read. not all registers may be written, as some are just status registers. each register or memory access point is assigned a unique address: the whole (8- bit) address range is reserved for the fx980. indirect memory addressing all internal memory access is via an access point. first, a command word access is used to reset the internal address pointer, then data port access operations post-increment this address pointer. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 14 d/980/3 example: to program the fifth and sixth locations of the auxiliary sram with $0x01aa the commands would be: $0x0000 t cmd ; set configctrl1 all bits low ; use default conditions $0x0118 t cmd ; set configctrl2 bits 7 and 6 low ; required by default for these reserved bits ; set configctrl2 bit 4 high ; post-increment addresses on a read operation ; set configctrl2 bit 3 high ; enable read/write access to the auxiliary sram $0xf300 t cmd ; read sramdata lsb register ; read fourth memory location (lsb). post-increment pointer. cmdrd t $0xf3xx ; sramdata lsb register data returned ; discard this byte $0x7002 t cmd ; write sramdata lsb register ; write $0x02 to fifth memory location (lsb) $0x716a t cmd ; write sramdata msb register ; write $0x6a to sixth memory location (msb) $0xf000 t cmd ; read sramdata lsb register ; read fifth memory location (lsb) cmdrd t $0xf002 ; sramdata lsb register data returned ; check this byte is $0x02 $0xf100 t cmd ; read sramdata msb register ; read sixth memory location (msb) cmdrd t $0xf16a ; sramdata msb register data returned ; check this byte is $0x6a $0x0110 t cmd ; set configctrl2 bit 3 low ; disable read/write access to the auxiliary sram 1.5.6.6 coefficient memory the convention for naming filter coefficients is a1 to an, where n is given by (filter length + 1)/2, i.e. for the 15-tap filter, n = 8. this arises from the internal architecture of the filters and the fact that they are all ? odd ? and symmetrical. write or read operations beyond this coefficient number will be reflected about the central coefficient e.g. the tenth read operation from the 15-tap filter would access coefficient location a6. there is no practical reason to write or read beyond location n, but the user in any case must avoid write operations at the (filter length + 1) location. this location (a0) location must be zero for the filters to operate correctly. the global reset (n-reset pin) establishes this condition when taken low. 1.5.7 register description this section describes in detail each of the registers and access points addressed by the command control serial word. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 15 d/980/3 key to register map each section that follows describes in detail the operation and use of each of the registers in the device. the registers are split into their functional groups, grouping associated registers together. each section consists of a title, an address, a function reference field, a description, and a bit specification. the function reference field describes the overall access available to this section (rw/w/r, where r = read and w = write). the bit specification describes the function of each individual bit, or a range of bits within a register. there is a separate line for each distinct field of bits. the state column indicates the action available to each group of bits (rw/w/r). register reset state all i/o access points (both read and write) are reset to logic zero on taking n_reset low, except where explicitly shown in this document . the reset state of status bits will depend on the level of the status signal being monitored. other registers (both read and write) are not affected by taking n_reset low. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 16 d/980/3 1.5.7.1 register and access point summary control and status registers $0x00 configctrl1 configuration control register 1 $0x01 configctrl2 configuration control register 2 $0x02 powerdownctrl power control register $0x03 txsetup transmit set-up register $0x04-$0x07 txdata transmit data registers $0x08 rxsetup1 receive set-up control register 1 $0x09 rxsetup2 receive set-up control register 2 $0x0a anactrl analogue configuration control register $0x0b auxadcctrl auxiliary adc data converter control register $0x0c ramdacctrl ram dac control register $0x0d loopbackctrl loopback control register $0x0e txerrorstatus transmit error status register $0x0f txerrstatmask transmit error status interrupt mask register auxiliary function registers $0x10-$0x17 auxadcdata auxiliary adc data registers $0x18-$0x1f auxdacdata auxiliary dac data registers status and interrupt registers $0x20 rxerrorstatus receive error status register $0x21 rxerrorstatmask receive error status interrupt mask register $0x22 txfifostatus transmission data fifo status register $0x23 txfifostatmask tx data fifo status interrupt mask register memory i/o access points $0x24-$0x2d coefframdata coefficient memory i/o access addresses $0x2e-$0x2f not used. rx data path registers $0x30-$0x31 rxiqgainmult receive i channel gain attenuation registers $0x32-$0x33 rxiqoffset receive i channel offset correction registers $0x34-$0x35 rxiqgainmult receive q channel gain attenuation registers $0x36-$0x37 rxiqoffset receive q channel offset correction registers tetra baseband processor fx980 ? 1997 consumer microcircuits limited 17 d/980/3 rx data path access points $0x38-$0x39 rxdataaccess receive path data access point (i) $0x3a-$0x3b rxdataaccess receive path data access point (q) $0x3c-$0x3f not used tx data path registers $0x40-$0x41 txphase transmit i channel phase correction registers $0x42-$0x43 txiqgainmult transmit i channel gain attenuation registers $0x44-$0x45 txiqoffset transmit i channel offset correction registers $0x46-$0x47 txphase transmit q channel phase correction registers $0x48-$0x49 txiqgainmult transmit q channel gain attenuation registers $0x4a-$0x4b txiqoffset transmit q channel offset correction registers $0x4c-$0x4d txrampupinc transmit ramp-up increment registers $0x4e-$0x4f txrampdndec transmit ramp-down decrement registers tx data path access points $0x50-$0x51 txdataaccess transmit path data access point (i) $0x52-$0x53 txdataaccess transmit path data access point (q) $0x54-$0x5f not used self test registers $0x60-$0x61 bistprsg built-in self test pseudo-random sequence generator $0x62 bistcontrol built-in self test control register $0x63 not used $0x64-$0x6d bistcrcregisters built-in self test cyclic redundancy code checkers $0x6e-$0x6f not used sram memory access points $0x70-$0x73 sramdata auxiliary dac1 memory i/o access addresses $0x74-$0x7f not used note: addresses $0x80 to $0xff cannot be used as the msb controls the direction of data flow: ? 1 ? = high = read and ? 0 ? = low = write. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 18 d/980/3 configctrl1 title: configurati on control register address: $0x00 function: rw description: general configuration bits, together with operational control signal bits. bit name active state function 7 dataratehi high rw when set active all serial port data transfers will be at half of the master clock rate. when inactive, all serial port data rates will be at a quarter of the master clock rate. this has the effect of altering the rx sample output rate from 8 times the symbol rate when active to 4 times when inactive. 6 txhandshakeen high rw when set active enable the transmit hardware interlock protocol, thereby stopping the serial clock ( sclk) if the transmit path is enabled and the transmit fifo is full. 5 bidircmdporten high rw when this bit is set active the cmd port will drive its data line out of the chip for the last 8 bits of read operations. when set inactive command read data will be returned on either the rx or the cmdrd port (default). 4 rxdataforcmdrden high rw this bit only takes effect if the bidircmdporten bit is inactive. when set active this bit causes all command read operations to respond with data on the rx serial port. when set inactive the command read data will be output via the cmdrd port (default). (5,4) commandreaddatamode rw the bidircmdporten bit and rxdataforcmdrden bit together control the method by which command read data is returned to the user. 00 (default) read data returned on cmdrd port. 01 read data returned on rx port and cmdrd port 10,11 read data returned on cmd port. 3 lowrxrdfs high rw when set active both the cmdrdfs and the rxfs output pins will be driven active low, when set inactive the two frame sync's will be driven active high (default). 2 rxdataportdisable high rw when set active tristates the rxdat and rxfs pins. 1 rdcmdportdisable high rw when set active tristates the cmdrddat and cmdrdfs pins. 0 symbomodubypass high rw setting this bit bypasses the modulator, thereby taking the least significant 3 bits of each command transmit data serial word received via the serial interface to represent an absolute constellation mapping. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 19 d/980/3 address and data format for configctrl1 access data field [7:0] d7 d6 d5 d4 d3 d2 d1 d0 0 0 0 0 0 0 0 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 20 d/980/3 configctrl2 title: configuration control register address: $0x01 function: rw description: general configuration bits, together with operational control signal bits. bit name active state function 7 rw reserved. set this bit low. undefined on read. 6 rw user defined bit. this bit has no internal functionality and is reset low with the global n_reset pin. the user may employ this bit for any useful purpose. 5 n_slowdown low rw when active, this bit reduces the slew rate of digital output pins. this reduces power consumption, ground bounce and reflection problems associated with fast edges on poorly terminated lines. de-activation speeds up the digital outputs, but increases power consumption, ground bounce and reflection problems. it is anticipated that the latter mode will be used only in 3.3v systems. 4 sramiordinc high rw this bit determines whether a read or write operation to the auxiliary sram will increment the address pointers. when set active causes read operations to move the address pointer on, this would therefore allow an efficient write then read verify scheme to be used. when set inactive write operations increment the address pointer. 3 sramloen high rw when set active allows read/write access to the auxiliary sram. it is only valid to activate this bit when the sram is not being accessed by the ramdac. when this bit is set active, the first access to sramdata will access the first sram address location. subsequent read or write accesses will increment the address pointer to the next memory location. 2 coefframiordinc high rw this bit determines whether a read or write operation to a coefficient memory will increment the address pointers. when set active the address pointer is incremented by any coefficient ram read operation, thereby allowing a write then read verification. when set inactive, write operations increment the address pointer. 1 coefframloen high rw when set active allows read/write access to all the coefficient memories. this bit is valid only when the tx and rx data paths are inactive. when this bit is set active, the first access to any of the coefficient memories will access the first coefficient location (a1). subsequent read or write accesses to any coefficient memory will increment the address pointers for all the coefficient memories. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 21 d/980/3 0 n_bigenddata low rw when set active causes serial port read data, from the rx port to be generated with the msb data bit as the first serial word bit. if inactive, the lsb is first. on taking n_reset low this bit is active (i.e. the default is msb first). address and data format for configctrl2 access data field [7:0] r d6 d5 d4 d3 d2 d1 d0 1 0 0 0 0 0 0 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 22 d/980/3 powerdownctrl title: power control register address: $0x02 function: rw description: this register, together with the following bits, controls the power saving features: txen bit of register txsetup rxen bit of register rxsetup1 txclkstop bit of register txsetup rxclkstop bit of register rxsetup1 bit name active state function 7 rw reserved. set this bit low. undefined on read. 6 biaslctrl high rw when set active, increases tx and rx analogue bias currents. 5 biaspowdn low rw when set active powers down the analogue bias section. 4 auxdac4powdn low rw when set active powers down auxiliary dac4. 3 auxdac3powdn low rw when set active powers down auxiliary dac3. 2 auxdac2powdn low rw when set active powers down auxiliary dac2. 1 auxdac1powdn low rw when set active powers down auxiliary dac1. 0 rxaafpowdn low rw when set active powers down the receive analogue anti -alias filter (aaf). address and data format for powerdownctrl access data field [7:0] r d6 d5 d4 d3 d2 d1 d0 0 1 0 0 0 0 0 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 23 d/980/3 txsetup title: transmit set-up register address: $0x03 function: rw description: sets up the transmit functions. bit name active state function 7:4 rw reserved. set these bits low. undefined on read. 3 txclkstop high rw when set active causes the txen bit to also be used to gate the tx data path master clock. when inactive (default state) the tx data path master clock is always supplied. 2 txen high rw when set active, enables the tx data path, allowing transmission to start when the correct enable sequence has been seen. this bit may only be cleared when the txpathen status bit in the txfifostatus register is inactive, setting inactive during a transmission cycle will cause erroneous behaviour. this bit also acts as a transmit section power enable bit. 1 txrampen high rw when set active, this bit enables the transmit amplitude ramping function. ramping is then controlled by the txrampup bit of the txdata register when this bit is inactive, the txrampup bit will directly control the transmit amplitude (high meaning full amplitude, low meaning zero amplitude). 0 txfircoeffreset low rw when set active this bit forces all the tx data path filters to load their default coefficient values. this bit will be set active on taking n_reset low, and therefore needs to be deactivated before default filter coefficients can be overwritten. address and data format for txsetup access data field [7:0] r r r r d3 d2 d1 d0 1 1 0 0 0 0 0 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 24 d/980/3 txdata title: transmit data register address: $0x04 - $0 x07 (mapped over four locations, two address bits being used as data bits) function: w fifo input r fifo output description: this transmit data register is 10 bits wide. the two least significant bits of the address bus are used to drive bits 8 and 9, hence it can be considered to be mapped over four consecutive locations. this data word is written into a fifo. the function is only decoded when the fifo is read (there is an exception for the first data word). the fifo will be read when the tx data path demands data. this will only occur when the txen bit of the txsetup register is set active. for test purposes the fifo data output may be accessed by reading these registers. data write with symbol modulator not bypassed bit name active state function 9 txrampup high w this bit is written to the fifo. while the txen bit of the txsetup register is active, it controls the tx data path ramping. setting it active will cause the amplitude to ramp up to its full value, conversely setting the bit inactive will cause the amplitude to ramp down to its minimum value. if the bit is changed while the amplitude is being ramped, the ramp direction will change to the direction set by this bit. while the txrampen bit is inactive, the txrampup bit will directly control the transmit amplitude (high meaning full amplitude and low meaning zero amplitude). 8 multisymbol high w this bit is written to the fifo and when this bit is set active, the fifo symbol data will be marked as a four symbol word. when set inactive, the fifo symbol data will be marked as a single symbol word. this bit is inactive if the symbmodubypass bit of the configctrl1 register is active. 7:6 txrelsymbol4 data w fourth symbol in word to be written to fifo. 5:4 txrelsymbol3 data w third symbol in word to be written to fifo. 3:2 txrelsymbol2 data w second symbol in word to be written to fifo. 1:0 txrelsymbol1 data w first symbol in word to be written to fifo. data write with symbol modulator bypassed bit name active state function 9 txrampup high w (see above) 8:3 (not used) data w redundant data which is still written into the fifo. set these bits low. 2:0 txabssymbol data w iq constellation point which is written into the fifo. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 25 d/980/3 read operation bit name active state function address $0x04 7:2 reserved. bit values are not defined. 1:0 upperfiforddata data r reads address access bits 9 and 8 of the fifo data output register, these are placed in bits 1 and 0. address $0x05 7:0 lowerfiforddata data r reads address access bits 7 to 0 of the fifo data output register. reading this location also performs a fifo read operation, thereby moving the next (if any) fifo data location into the fifo data output register. address $0x06 and $0x07 7:0 r reserved. bit values are not defined. for these read operations to be valid, the tx data path must be active ( txen bit of txsetup register set active) and the symbmodubypass bit of the configctrl1 register must also be set active. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 26 d/980/3 address and data format for txdata write access d8 data field [7:0] d9 1 0 0 0 0 address field [6:0] d1 d0 d7 d6 d5 d4 d3 d2 multisymbol bit txrampup bit address and data format for txdata (modulator bypass mode) write access nu data field [7:0] d9 1 0 0 0 0 address field [6:0] d1 d0 nu nu nu nu nu d2 not used txrampup bit address and data format for txdata read access data field [7:0] 0 0 1 0 0 0 0 address field [6:0] d9 d8 r r r r r r d1 d0 d7 d6 d5 d4 d3 d2 1 0 1 0 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 27 d/980/3 rxsetup1 title: first receive set-up control register address: $0x08 function: rw description: receive path set-up and initialisation control bits. bit name active state function 7 rx32bitmode high rw when set active, the rx port operates on 32-bit frames - i data in the msb word, q data in the lsb word. 6 rxsamplesel high rw this bit is used to select which pair of i,q samples is supplied from the possible two when the dataratehi bit in configctrl1 register is in the low mode (inactive). it has no effect when dataratehi is active. 5 rxclkstop high rw when set active causes the rxen bit to also be used to gate the rx data path master clock. when inactive (default state) the rx data path master clock is always supplied. 4 rxen high rw when set active, enables the rx data path, which then starts to process the differential data on the irxp,irxn and qrxp,qrxn pins, outputting results via the rx serial port. this bit also acts as a receive section power enable bit. 3 rxbistactive high rw when set active, enables rx built-in self test (bist) operation. 2 anaadcreset pulse w when this bit is set high, a 4-clock-cycle adc auto reset event is generated. it is not necessary to clear this bit before another adc auto reset event is initiated. r the read state of this bit indicates the logic level last written to this bit. it does not have a functional significance and is only available for test purposes. 1 anaenautoreset low rw when active this bit enables the adc auto reset function. on taking n_reset low, this bit is set active, which is the default operating condition. 0 rxfircoeffreset low rw when set active forces all the rx data path filters to load their default coefficient values. this bit will be set active on taking n_reset low, and therefore needs to be deactivated before default filter coefficients can be overwritten. normal filter operation is unaffected by leaving this bit set. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 28 d/980/3 address and data format for rxsetup1 access data field [7:0] d7 d6 d5 d4 d3 d2 d1 d0 0 0 0 0 0 0 1 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 29 d/980/3 rxsetup2 title: second receive set-up cont rol register address: $0x09 function: rw description: receive i and q vernier control bits. bit name active state function 7:4 qvernierdelay high rw q channel vernier sampling delay, allowing the sampling point to be adjusted to a resolution of 1/16 of the input sample clock rate. 3:0 ivernierdelay high rw i channel vernier sampling delay, allowing the sampling point to be adjusted to a resolution of 1/16 of the input sample clock rate. note: the values are in the format of 4 bit signed 2s-complement integers - the msb being the sign. thus it can be interpreted as adjusting the reference phase by 7/16 of the sample clock period. address and data format for rxsetup2 access data field [7:0] d7 d6 d5 d4 d3 d2 d1 d0 1 0 0 0 0 0 1 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 30 d/980/3 anactrl title: analogue configuration control register address: $0x0a function: rw description: reserved. all bits should be set low. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 31 d/980/3 auxadcctrl title: auxiliary adc data converter control register address: $0x0b function: rw description: this register controls the operation of the four adc channels. these are implemented using a single adc converter which is multiplexed on to each of the adc channels. a conversion cycle consists of performing a conversion for each of the active channels in turn. bit name active state function 7 rw reserved. set this bit low. undefined on read. 6 adcconvertrate high rw this bit changes the adc conversion rate. if this bit is set low, the adc is clocked by mclk/8, yielding a conversion time of 80x mclk periods per adc channel. the maximum sample rate is lower than this. with a single channel selected, the maximum rate is mclk/112 samples/second. setting this bit high will halve the adc clock rate, and hence double the conversion time. 5 adccontconv high rw continuous conversion mode control bit; when inactive, sets the adcs into one-shot conversion mode; when active, the adcs will continuously convert. one-shot conversion mode is initiated by the startconvert bit. in continuous convert mode, the adc will start a new conversion cycle on all active channels after the previous cycle has completed. 4 enableadc4 high rw setting this bit high will enable adc channel 4 for conversion. this bit may be updated at any time, but will only change the active state of the adc channel for the next time it is converted. 3 enableadc3 high rw setting this bit high will enable adc channel 3 for conversion. this bit may be updated at any time, but will only change the active state of the adc channel for the next time it is converted. 2 enableadc2 high rw setting this bit high will enable adc channel 2 for conversion. this bit may be updated at any time, but will only change the active state of the adc channel for the next time it is converted. 1 enableadc1 high rw setting this bit high will enable adc channel 1 for conversion. this bit may be updated at any time, but will only change the active state of the adc channel for the next time it is converted. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 32 d/980/3 0 startconvert high one-shot conversion control bit. only valid when the adcs are set to one-shot conversion mode. w setting this bit high starts the adc data conversion. setting this bit low will stop the conversion. this should only be used for test purposes, because the adc conversion logic will automatically set this bit low when the conversion operation has completed. r this bit can be set high or low by the serial interface, but the adc conversion logic will automatically set it low when the current conversion cycle has completed. address and data format for auxillary adc control access data field [7:0] r d6 d5 d4 d3 d2 d1 d0 1 1 0 0 0 0 1 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 33 d/980/3 auxadcdata title: auxiliary adc data registers address: (eight registers) $0x10 to $0x17 function: r description: these registers enable the user to inspect the conversion value for each of the four auxiliary adcs. there are two read registers per adc, one to obtain the least significant two bits of the data, the other for the most significant eight bits. reading these registers does not affect the adc conversion cycle. reading the msb read register directly reads the adc output and simultaneously causes the two bits in the lsb read register to be written into a holding register. this holding register is read when the lsb read register is read. this mechanism is necessary to allow the user to read msb and lsb data from the same adc conversion cycle. if only the msb read register is read, the converter can be considered as an 8-bit adc. if a 10-bit conversion is required, the msb read register must be read first. bit name active state function address $0x10 7:0 adc1msbdata data r most significant eight bits of the data from the last conversion of auxadc1 . address $0x11 7:2 r reserved. bit values are not defined. 1:0 adc1lsbdata data r least significant two bits of the data from the last conversion of auxadc1 . address $0x12 7:0 adc2msbdata data r most significant eight bits of the data from the last conversion of auxadc2 . address $0x13 7:2 r reserved. bit values are not defined. 1:0 adc2lsbdata data r least significant two bits of the data from the last conversion of auxadc2 . address $0x14 7:0 adc3msbdata data r most significant eight bits of the data from the last conversion of auxadc3 . address $0x15 7:2 r reserved. bit values are not defined. 1:0 adc3lsbdata data r least significant two bits of the data from the last conversion of auxadc3 . address $0x16 7:0 adc4msbdata data r most significant eight bits of the data from the last conversion of auxadc4 . address $0x17 7:2 r reserved. bit values are not defined. 1:0 adc4lsbdata data r least significant two bits of the data from the last conversion of auxadc4 . tetra baseband processor fx980 ? 1997 consumer microcircuits limited 34 d/980/3 address and data format for auxillary adc data access 0 1 data field [9:0] d9 d8 d7 d6 d5 d4 d3 d2 r r r r r r d1 d0 n0 n1 0 0 1 0 n0 n1 0 0 1 0 n1 n0 adc channel 0 0 channel 1 0 1 channel 2 1 0 channel 3 1 1 channel 4 address field [6:0] tetra baseband processor fx980 ? 1997 consumer microcircuits limited 35 d/980/3 ramdacctrl title: ramdac control register address: $0x0c function: rw description: this register controls the operation of dac 1, together with the operation of the memory ( dacsram) which can be used to drive the digital input of dac 1. bit name active state function 7:6 rw reserved. these bits should be set low. undefined on read. 5:3 ramdacrate high rw these three bits set the rate at which the ramdac memory?s dac access address pointer changes. the three bit value ( ramdacrate) causes a change rate of ( 36 x 2 r amdacrate ) khz. see table below. 2 ramdacinc high rw this bit activates the ramdac memory scan operation. setting it active will cause the memory address to increment up to the top (highest) location, conversely setting the bit inactive will cause the memory address to decrement down to the bottom location. if the bit is changed while the memory is being scanned, the current scan will complete before the new state of the ramdacinc bit takes effect. 1 autocycle high rw this bit is only valid if the ramdacactive bit is active. when set active, the auxiliary sram memory will be continually scanned at the rate set by the ramdacrate bits. this enables a symmetrical periodic waveform to be driven out on the auxdac1 pin. the auxiliary sram address cycles from the bottom location up to the top location, and back down to the bottom again. 0 ramdacactive high rw dac 1 input mode bit. when inactive, the auxdacdata registers (offsets 0 and 1) are used as the source for conversion. if this bit is active, the dac is driven from the output of the ramdac memory. ram dac rate select table ramdacctrl[5:3] dac update frequency (khz) 0 0 0 36 0 0 1 72 0 1 0 144 0 1 1 288 1 0 0 576 1 0 1 1152 1 1 0 2304 1 1 1 4608 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 36 d/980/3 address and data format for ramdacctrl access d1 d0 r r d5 d4 d3 d2 data field [7:0] address field [6:0] 0 0 0 0 0 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 37 d/980/3 auxdacdata title: auxiliary dac data registers address: (eight registers) $0x18 to $0x1f function: rw description: there are two input registers for each of the four auxiliary dacs. writing to the auxdac#lsbdata register writes the least significant two bits of dac data. writing to the auxdac#msbdata register writes the most significant eight bits of dac data and then passes all ten bits to the appropriate dac input (only if the ramdacactive bit is set low for dac 1). if the auxdac#msbdata register is written while the auxdac#lsbdata register is left constant, the converter may be treated as an 8-bit dac. bit name active state function address $0x18 7:2 rw reserved. these bits should be set low. undefined on read. 1:0 auxdac1lsbdata data rw writing to this address writes the least significant two bits of the dacdata1 register. these two bits may be read for test purposes. address $0x19 7:0 auxdac1msbdata data rw writing to this address writes the most significant eight bits of the dacdata1 register and updates dac 1. this register may also be read for test purposes. address $0x1a 7:2 rw reserved. these bits should be set low. undefined on read. 1:0 auxdac2lsbdata data rw writing to this address writes the least significant two bits of the dacdata2 register. these two bits may be read for test purposes. address $0x1b 7:0 auxdac2msbdata data rw writing to this address writes the most significant eight bits of the dacdata2 register and updates dac 2. this register may also be read for test purposes. address $0x1c 7:2 rw reserved. these bits should be set low. undefined on read. 1:0 auxdac3lsbdata data rw writing to this address writes the least significant two bits of the dacdata3 register. these two bits may be read for test purposes. address $0x1d 7:0 auxdac3msbdata data rw writing to this address writes the most significant eight bits of the dacdata3 register and updates dac 3. this register may also be read for test purposes. address $0x1e 7:2 rw reserved. these bits should be set low. undefined on read. 1:0 auxdac4lsbdata data rw writing to this address writes the least significant two bits of the dacdata4 register. these two bits may be read for test purposes. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 38 d/980/3 address $0x1f 7:0 auxdac4msbdata data rw writing to this address writes the most significant eight bits of the dacdata4 register and updates dac 4. this register may also be read for test purposes. address and data format for auxillary dac data access 0 1 data field [9:0] address field [6:0] d9 d8 d7 d6 d5 d4 d3 d2 r r r r r r d1 d0 0 0 1 1 n1 n0 0 0 1 1 n1 n0 n1 n0 channel selected 0 0 channel 1 0 1 channel 2 1 0 channel 3 1 1 channel 4 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 39 d/980/3 loopbackctrl title: loopback control register address: $0x0d function: rw description: this register is only used for test purposes. for normal operation all these bits should be inactive. bit name active state function 7:6 rw reserved. these bits should be set low. undefined on read. 5 firreset high rw when active, this bit holds all fir filters in reset, by resetting the fir address pointers. this by itself does not reset the data register rams. a separate access is provided to disable the complete tx or rx data path. taking n_reset low will also reset the fir filter coefficients to their default values. 4 digloopback high rw when set active this bit enables the digital loopback feature. this connects the output of the rx data path 49 -tap filter to the input of the tx data path 49-tap digital filter, thereby allowing an analogue signal presented at the rx inputs to be filtered by a raised cosine filter and monitored at the tx outputs as an analogue signal. 3 analoopback high rw when set active this bit enables the analogue loopback feature. this connects the output of the tx data path dac to the input of rx data path adc, thus passing transmit constellation data through a raised cosine filter and allowing the resultant data samples to be monitored digitally at the rx output. 2 rxdpaccesssel high rw when set active this bit disables the rx data path sample clock, thereby enabling the data path access register to directly update the output of the rx data path operator. 1 txdpaccesssel high rw when set active this bit disables the tx data path sample clock, thereby enabling the data path access register to directly update the input to the 15-tap digital filter without the data being overridden by subsequent sample clocks 0 txtorxdatapath high rw when set active this bit connects the tx (i,q) dac input to the serial receive port ( rx ). this enables the output of the transmit 15-tap filter to be observed in real time. data is taken from the i and q channels on alternate 144khz sample clocks. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 40 d/980/3 address and data format for loopbackctrl access d1 d0 r r d5 d4 d3 d2 data field [7:0] address field [6:0] 0 1 0 0 0 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 41 d/980/3 txerrorstatus title: transmit error status register. address: $0x0e function: r description: this register is the tx data path error status register. the txirqactive bit is set active when one of the other bits in this register is the source of an interrupt event. all these error conditions are caused by transitory events, therefore the error condition is latched (marked with an ?l?). reading this status register causes all latched bits to be set inactive, unless an error event is currently pending. setting any bit of this register high will cause an interrupt to be generated (n_irq will be set low) if the source of the interrupt has not been masked in the corresponding mask register. bit name active state function 7 r reserved. bit value is not defined. 6 txdatapathqof high rl data path gain, phase and offset (gpo) adjustment-unit: q channel overflow error status bit. 5 txdatapathiof high rl data path gain, phase and offset (gpo) adjustment-unit: i channel overflow error status bit. 4 tx15tapqof high rl 15-tap q filter data accumulator overflow error status bit. 3 tx15tapiof high rl 15-tap i filter data accumulator overflow error status bit. 2 tx49tapof high rl 49-tap i and q filter data accumulator overflow error status bit. 1 tx79tapof high rl 79-tap i and q filter data accumulator overflow error status bit. 0 txirqactive high rl this bit is set high if there is an active interrupt caused by one of the status bits in this register. address and data format for txerrorstatus access d1 d0 r d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 1 0 0 0 0 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 42 d/980/3 txerrstatmask title: transmit error status interrupt mask register address: $0x0f function: rw description: masks interrupts in the txerrorstatus register. on taking n_reset low, these bits are set active, so masking out all possible interrupt sources. each bit which is taken inactive will allow its associated status bit, when active, to generate an interrupt. bit name active state function 7 data rw reserved for manufacturer?s test purposes. this bit should be set low. 6 n_txdatapathqof_mask low rw gpo q channel error interrupt mask bit. 5 n_txdatapathiof_mask low rw gpo i channel error interrupt mask bit. 4 n_tx15tapqof_mask low rw 15-tap q filter error interrupt mask bit. 3 n_tx15tapiof_mask low rw 15-tap i filter error interrupt mask bit. 2 n_tx49tapof_mask low rw 49-tap i and q filter error interrupt mask bit. 1 n_tx79tapof_mask low rw 79-tap i and q filter error interrupt mask bit. 0 data reserved for manufacturer?s test purposes. this bit should be set low. address and data format for txerrstatmask access d1 r r d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 1 1 0 0 0 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 43 d/980/3 rxerrorstatus title: receive error status register. address: $0x20 function: r description: this register is the rx data path error status register. the rxirqactive bit is set active when one of the other bits in this register is the source of an interrupt event. all these error conditions are caused by transitory events, therefore the error condition is latched (marked with an ?l?). reading this status register causes all latched bits to be set inactive unless an error event is currently pending. setting any bit of this register high will cause an interrupt to be generated (n_irq will be set low) if the source of the interrupt has not been masked in the corresponding mask register. bit name active state function 7 rxdatapathqof high rl data path gain, phase and offset (gpo) adjustment unit: q channel overflow error status bit. 6 rxdatapathiof high rl data path gain, phase and offset (gpo) adjustment unit: i channel overflow error status bit. 5 adcqof high rl adc q channel overflow error due to excessive input amplitude. 4 adciof high rl adc i channel overflow error due to excessive input amplitude. 3 rx63tapof high rl 63-tap i and q filter data accumulator overflow error status bit. 2 rx49tapof high rl 49-tap i and q filter data accumulator overflow error status bit. 1 evensamplephase high rl when this status bit is active, the associated interrupt may be used to re-synchronise the rx data if for any reason data synchronisation is lost. if the corresponding mask bit is set inactive, an interrupt will be generated on the next q -phase data in the rx output register. the next falling edge of sclk with rxfs high indicates the lsb of the q channel data. the mask bit should be disabled after this to prevent continuous q-phase interrupts. 0 rxirqactive high rl this bit is set high if there is an active interrupt caused by one of the status bits in this register. address and data format for rxerrorstatus access d1 d0 d7 d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 0 0 0 1 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 44 d/980/3 rxerrorstatmask title: receive error status interrupt mask register. address: $0x21 function: rw description: masks interrupts in the rxerrorstatus register. on taking n_reset low, these bits are set active, so masking out all possible interrupt sources. each bit which is taken inactive will allow its associated status bit, when active, to generate an interrupt. bit name active state function 7 n_rxdatapathqof_mask low rw gpo q channel error interrupt mask bit. 6 n_rxdatapathiof_mask low rw gpo i channel error interrupt mask bit. 5 n_adcqof_mask low rw adc q channel error interrupt mask bit. 4 n_adciof_mask low rw adc i channel error interrupt mask bit. 3 n_rx63tapof_mask low rw 63-tap i and q filter error interrupt mask bit. 2 n_rx49tapof_mask low rw 49-tap i and q filter error interrupt mask bit. 1 evensamplephase_mask low rw rx data q-phase interrupt mask bit. 0 data rw reserved for manufacturer?s test purposes. this bit should be set low. address and data format for rxstatmask access d1 r d7 d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 0 1 0 1 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 45 d/980/3 txfifostatus title: transmit data fifo status register address: $0x22 function: r description: this register is the tx data fifo status register. the txirqactive bit is set active when one of the other bits in this register is the source of an interrupt event. some of these status conditions are caused by transitory events, therefore their state is latched (marked with an ?l?). the bits marked with a parenthesised ?l? are only latched in their interrupt generation state if their associated mask bit is inactive. reading this status register causes all latched bits to be set inactive, unless an error event is currently pending. setting any bit of this register high will cause an interrupt to be generated (n_irq will be set low) if the source of the interrupt has not been masked in the corresponding mask register. bit name active state function 7 txpathen high/ low r(l) when active (high) this bit shows that the tx data path is currently active. this enables the user to confirm that ramp down has completed. for interrupt generation purposes, a logic low on this bit will be considered as active. 6 fifounderread high rl error status bit. when active indicates a read from the fifo occurred while the fifo was empty. 5 fifooverwrite high rl error status bit. when active indicates a write to the fifo occurred while the fifo was full. 4 fifofull high/ low r(l) most significant fifo length status bit. when active (high) this bit also indicates the fifo is full. for interrupt generation purposes, a logic low on this bit will be considered as active. 3:2 fifolength (low) r(l) these two bits contain the pointer to the next free fifo address and indicate the following status: 00 - indicates fifo is empty 01 - one location used 10 - two locations used 11 - three locations used for interrupt generation purposes, a logic low on either of these bits will be considered as active. 1 fifoempty high r(l) when active indicates the fifo is empty. 0 fifoirqactive high rl this bit is set high if there is an active interrupt caused by one of the status bits in this register. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 46 d/980/3 address and data format for txfifostatus access d1 d0 d7 d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 1 0 0 1 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 47 d/980/3 txfifostatmask title: transmit data fifo status interrupt mask register address: $0x23 function: rw description: masks interrupts in the txfifostatus register. on taking n_reset low, these bits are set active, so masking out all possible interrupt sources. each inactive bit will allow its associated status bit to generate an interrupt. in the case of the status bits marked in the txfifostatus register with a parenthesised ?l?, taking the mask bit inactive will enable the latching mechanism. bit name active state function 7 n_txpathen_mask low rw tx data path active interrupt mask bit. 6 n_fifounderread_mask low rw fifo underflow interrupt mask bit. 5 n_fifooverwrite_mask low rw fifo overflow interrupt mask bit. 4 n_fifofull_mask low rw fifo full interrupt mask bit. 3 n_fifolength1_mask low rw fifo length status (msb) interrupt mask bit. 2 n_fifolength0_mask low rw fifo length status (lsb) interrupt mask bit. 1 n_fifoempty_mask low rw fifo empty interrupt mask bit. 0 data rw reserved for manufacturer?s test purposes. this bit should be set low. address and data format for txfifostatmask access d1 r d7 d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 1 1 0 1 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 48 d/980/3 coefframdata title: i/o access addresses for the five coefficient memories. address: $0x24 to $0x2d (mapped over 10 locations) function: rw description: each coefficient ram has both msb and lsb address ports assigned for read/write access. there are three transmit (tx) fir filters with read/write coefficients and two receive (rx) filters, with coefficient sizes of 12 and 16 bits respectively. access to the coefficient memory is valid when the coefframioen bit is active. asserting the coefframioen will reset the coefficient address pointer to the first location (a1). the msb port should be accessed first, as accessing the lsb port will move the coefficient address pointer to the next coefficient location (a[n+1 ]) (refer to description of coefframiordinc bit for details). subsequent accesses to the lsb port of the coefficient address will increment the coefficient address pointer . as all filters are symmetrical and ? odd ? , only n + 1 2 locations can be programmed, where n is the filter tap length. performing an i/o access after the last coefficient address pointer is not valid, and may corrupt existing coefficients. only one fir filter coefficient ram may be accessed at a time. if further memories are to be accessed then the coefframioen must first be deactivated, and then activated again, allowing the next fir filter coefficient ram to be incrementally accessed. bit name active state function address $0x24 7:0 tx15tapcoefflsb data rw transmit 15-tap filter lsb coefficient data port. post -increment the coefficient address pointer. address $0x25 7:4 rw reserved. set these bits high. undefined on read. 3:0 tx15tapcoeffmsb data rw transmit 15-tap filter msb coefficient data port. address $0x26 7:0 tx49tapcoefflsb data rw transmit 49-tap filter lsb coefficient data port. post-increment the coefficient address pointer. address $0x27 7:4 rw reserved. set these bits high. undefined on read. 3:0 tx49tapcoeffmsb data rw transmit 49-tap filter msb coefficient data port. address $0x28 7:0 tx79tapcoefflsb data rw transmit 79-tap filter lsb coefficient data port. post -increment the coefficient address pointer. address $0x29 7:4 rw reserved. set these bits high. undefined on read. 3:0 tx79tapcoeffmsb data rw transmit 79-tap filter msb coefficient data port. address $0x2a 7:0 rx49tapcoefflsb data rw receive 49-tap filter lsb coefficient data port. post -increment the coefficient address pointer. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 49 d/980/3 address $0x2b 7:0 rx49tapcoeffmsb data rw receive 49-tap filter msb coefficient data port. address $0x2c 7:0 rx63tapcoefflsb data rw receive 63-tap filter lsb coefficient data port. post -increment the coefficient address pointer. address $0x2d 7:0 rx63tapcoeffmsb data rw receive 63-tap filter msb coefficient data port. address and data format for 15-tap tx fir coefficient ram io access d11 d10 d9 d8 r r r r d7 d6 d5 d4 d3 d2 d1 d0 coefficient data field [11:0] address field [6:1] (coefficient pointer)++ 0 coefficient pointer 1 0 1 0 a0 0 1 0 address and data format for 49-tap tx fir coefficient ram io access d11 d10 d9 d8 r r r r d7 d6 d5 d4 d3 d2 d1 d0 coefficient data field [11:0] address field [6:1] (coefficient pointer)++ 0 coefficient pointer 1 0 1 1 a0 0 1 0 address and data format for 79-tap tx fir coefficient ram io access d11 d10 d9 d8 r r r r d7 d6 d5 d4 d3 d2 d1 d0 coefficient data field [11:0] address field [6:1] (coefficient pointer)++ 0 coefficient pointer 1 1 0 0 a0 0 1 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 50 d/980/3 address and data format for 49-tap rx fir coefficient ram io access d11 d10 d9 d8 d15 d14 d13 d12 d7 d6 d5 d4 d3 d2 d1 d0 coefficient data field [15:0] address field [6:1] (coefficient pointer)++ 0 coefficient pointer 1 1 0 1 a0 0 1 0 address and data format for 63-tap rx fir coefficient ram io access d11 d10 d9 d8 d15 d14 d13 d12 d7 d6 d5 d4 d3 d2 d1 d0 coefficient data field [15:0] address field [6:1] (coefficient pointer)++ 0 coefficient pointer 1 1 1 0 a0 0 1 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 51 d/980/3 sramdata title: i/o access address for the auxiliary dac1 memories. address: $0x70 to $0x73 (mapped over 4 locations) function: rw description: these four address locations allow access to the 64 x 10 bit sram. the contents of this ram can be pre-loaded with a table of values which can be automatically sent to auxiliary dac1 in either a single cycle or continuous mode, see ramdacctrl for details. therefore the ram can be used in conjunction with dac1 to enable user defined profile power ramping of an external rf power transmitter stage. the ram contents are addressed incrementally by first taking the sramioen bit active. while this bit is inactive the sram address pointer is held reset. the physical address applied to the ram is formed from the 4-bit sram address pointer and the two lsb bits from the i/o access address (a1,a0). therefore four locations in the ram can be accessed by directly addressing $0x70 to $0x73. however, accessing location $0x73 post-increments (by a block of four addresses) the sram address pointer , thus moving the pointer to the next ram location block. the 10-bit data word is split between ? odd ? and ? even ? locations with the msb byte in ? odd ? addresses (a0 = 1) and 2 lsb?s in ? even ? addresses. the sramiordinc bit determines whether a read or a write operation will increment the sram address pointer . all 16 locations are accessed incrementally, further accesses to this port while the sramioen bit is active are not valid and may cause data loss. bit name active state function address $0x70 7:2 rw reserved. set these bits low. undefined on read. 1:0 sramlsbport0 data rw access port for the lsb register. address $0x71 7:0 srammsbport0 data rw access port for the msb register address $0x72 7:2 rw reserved. set these bits low. undefined on read. 1:0 sramlsbport1 data rw access port for the lsb register. address $0x73 7:0 srammsbport3 data rw access port for the msb register. post -increment sram address pointer. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 52 d/980/3 address and data format for sram data i/o access d9 d8 d7 d6 d5 d4 d3 d2 r r r r r r d1 d0 sram address pointer 0 0 sram address pointer 0 1 sram address pointer 1 0 ( sram address pointer)++ 1 1 data field [9:0] address field [6:0] d9 d8 d7 d6 d5 d4 d3 d2 r r r r r r d1 d0 a1 a0 1 1 1 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 53 d/980/3 txrampupinc title: transmit ramp up increment registers. address: $0x4c to $0x4d (mapped over 2 locations) function: rw description: the value in this register sets the scale of the tx amplitude gain increments which occur over each sample clock period, thus determining the tx amplitude ramp up time period. the value is always positive. the ramp up rate, in terms of the number of symbols, is given by the formula: n 64 n symbols inc = where: n symbols is the ramp time in terms of number of symbols. n inc is the value in the register. bit name active state function address $0x4c 7:0 rampupinclsb data rw least significant 8 bits of the ramp up increment register. address $0x4d 7:1 rw reserved. set these bits low. undefined on read. 0 rampupincmsb data rw most significant bit of the ramp up increment register. address and data format for txrampupinc access d1 d0 d7 d6 d5 d 4 d3 d2 r r r r r r r d8 0 data field [8:0] 1 address field [6:0] 1 0 1 1 0 0 1 0 1 1 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 54 d/980/3 txrampdndec title: transmit ramp down decrement registers. address: $0x4e to $0x4f (mapped over 2 locations) function: rw description: the value in this register sets the scale of the tx amplitude gain decrements which occur over each sample clock period, thus determining the tx amplitude ramp down time period. the value is always positive. the ramp down rate, in terms of the number of symbols, is given by the formula: n 64 n symbols inc = where: n symbols is the ramp time in terms of number of symbols. n inc is the value in the register. bit name active state function address $0x4e 7:0 rampdninclsb data rw least significant 8 bits of the ramp down increment register. address $0x4f 7:1 rw reserved. set these bits low. undefined on read. 0 rampdnincmsb data rw most significant bit of the ramp down increment register. address and data format for txrampdndec access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r r r d8 0 data field [8:0] 1 address field [6:0] 1 0 1 1 1 0 1 0 1 1 1 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 55 d/980/3 txiqgainmult title: transmit i and q channel gain multiplier registers address: $0x42, $0x43 , $0x48 and $0x49 (4 locations) function: rw description: a 2s-complement multiplication is performed on the magnitude of the tx data path signal and the result is then re-normalised to the system?s dynamic range: thus the function may be considered as a digital attenuator. this register sets the multiplier, the result being given by the formula: d d g 2 out in val 11 = ? ? where: d in d out g val is the signal input, is the signal output, is the value in the register. bit name active state function address $0x42 7:0 txigainlsb data rw least significant 8 bits of the txigain register ( g val ). address $0x43 7:3 rw reserved. set these bits low. undefined on read. 2:0 txigainmsb data rw most significant 3 bits of the txigain register ( g val ). address $0x48 7:0 txqgainlsb data rw least significant 8 bits of the txqgain register ( g val ). address $0x49 7:3 rw reserved. set these bits low. undefined on read. 2:0 txqgainmsb data rw most significant 3 bits of the txqgain register ( g val ). tetra baseband processor fx980 ? 1997 consumer microcircuits limited 56 d/980/3 address and data format for txigain access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r d10 d9 d8 0 data field [10:0] 1 address field [6:0] 1 0 0 0 1 0 1 0 0 0 1 0 address and data format for txqgain access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r d10 d9 d8 0 data field [10:0] 1 address field [6:0] 1 0 1 0 0 0 1 0 1 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 57 d/980/3 txiqoffset title: transmit i and q channel offset correction register address: $0x44, $0x45, $0x4a, and $0x4b (4 locations) function: rw description: this register controls the tx data path signal offset. this offset is a 2s-complement value ( n offset ), which is applied to the tx signal after the gain multiplier ( g val ), but before the dac. the offset applied is at the discretion of the user. inappropriate values may cause arithmetic overflow in the subsequent operator sections. the result is given by the formula: d d + n 2 out in offset 15 = ? ? where: d in d out n offset is the signal input, is the signal output, is the 2s-complement value in the register. bit name active state function address $0x44 7:0 txioffsetlsb data rw least significant 8 bits of the txioffset register ( n offset ). address $0x45 7:4 rw reserved. set these bits low. undefined on read. 3:0 txioffsetmsb data rw most significant 4 bits of the txioffset register ( n offset ). address $0x4a 7:0 txqoffsetlsb data rw least significant 8 bits of the txqoffset register ( n offset ). address $0x4b 7:4 rw reserved. set these bits low. undefined on read. 3:0 txqoffsetmsb data rw most significant 4 bits of the txqoffset register ( n offset ). tetra baseband processor fx980 ? 1997 consumer microcircuits limited 58 d/980/3 address and data format for txioffset access d1 d0 d7 d6 d5 d4 d3 d2 r r r r d11 d10 d9 d8 0 data field [11:0] 1 address field [6:0] 1 0 0 1 0 0 1 0 0 1 0 0 address and data format for txqoffset access d1 d0 d7 d6 d5 d4 d3 d2 r r r r d11 d10 d9 d8 0 data field [11:0] 1 address field [6:0] 1 0 1 0 1 0 1 0 1 0 1 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 59 d/980/3 txphase title: transmit i and q channel phase correction register address: $0x40, $0x41, $0x46, $0x47 (4 locations) function: rw description: this register controls the tx data path i and q channel phase compensation. the phase may be adjusted by 7.1 with respect to the input data signal phase. as each channel has separate phase adjustments the maximum differential phase compensation that can be achieved is 14.2 . the phase adjustment value written to this register is a 2s-complement value ( n phase ). the amount of phase adjustment applied is given by the formula: f tan n 2 - 1 phase 11 = ? ? where: f n phase is the phase adjustment, is the value in the register and has a range of -256 to +255. note: although each channel is separately adjustable with its own compensation value, the effect of phase adjustment is only detectable by measuring the phase angle between i and q channels. it should be noted that the n phase value has the effect of lagging the i channel for positive values of n phase (conversely, leading the phase for negative values) and leading the q channel for positive values of n phase (conversely, lagging the phase for negative values). for example, putting the value 10 (decimal) into both txiphase and txqphase would produce a differential phase on i and q of: 90 o - 2(tan -1 (4.88x10 -3 )) = 89.44 o bit name active state function address $0x40 7:0 txiphaselsb data rw least significant 8 bits of the txiphase register ( n phase ). address $0x41 7:1 rw reserved. set these bits low. undefined on read. 0 txiphasemsb data rw most significant bit of the txiphase register (sign bit). address $0x46 7:0 txqphaselsb data rw least significant 8 bits of the txqphase register ( n phase ). address $0x47 7:1 rw reserved. set these bits low. undefined on read. 0 txqphasemsb data rw most significant bit of the txqphase register (sign bit). tetra baseband processor fx980 ? 1997 consumer microcircuits limited 60 d/980/3 address and data format for txiphase access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r r r d8 0 data field [9:0] 1 address field [6:0] 1 0 0 0 0 0 1 0 0 0 0 0 address and data format for txqphase access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r r r d8 0 data field [9:0] 1 address field [6:0] 1 0 0 1 1 0 1 0 0 1 1 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 61 d/980/3 txdataaccess title: tx data path access point. address: $0x50 to $0x53 (mapped over 4 locations) function: rw description: this register block allows direct access to the tx data path values just after the gain, phase and offset adjustment block. both read and write operations are permitted. a read operation reads the signal values on the i and q channels. a write operation will write data to the data path just before the 15-tap filter. to prevent normal tx data overwriting this value the txdpaccesssel bit in the loopbackctrl register should be set active. the msb read data register is buffered to enable access to a discrete sample value (if this register was not buffered, data from different sample periods could be in the msb and lsb registers). therefore the lsb register must be read first for correct operation. bit name active state function address $0x50 7:0 txdpidatalsb data rw least significant 8 bits of the txdpidata register. this register must be read before its associated msb register. address $0x51 7:4 rw reserved. set these bits low. undefined on read. 3:0 txdpidatamsb data rw most significant 2 bits of the txdpidata register. address $0x52 7:0 txdpqdatalsb data rw least significant 8 bits of the txdpqdata register. this register must be read before its associated msb register. address $0x53 7:4 rw reserved. set these bits low. undefined on read. 3:0 txdpqdatamsb data rw most significant 2 bits of the txdpqdata register. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 62 d/980/3 address and data format for txdpidata access d1 d0 d7 d6 d5 d4 d3 d2 r r r r d11 d10 d9 d8 0 data field [11:0] 1 address field [6:0] 1 1 0 0 0 0 1 1 0 0 0 0 address and data format for txdpqdata access d1 d0 d7 d6 d5 d4 d3 d2 r r r r d11 d10 d9 d8 0 data field [11:0] 1 address field [6:0] 1 1 0 0 1 0 1 1 0 0 1 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 63 d/980/3 rxiqgainmult title: receive i and q channel gain multiplier register address: $0x30, $0x31, $0x34 and $0x35 (4 locations) function: rw description: a 2s-complement multiplication is performed on the magnitude of the rx data path signal and the result is then re-normalised to the system?s dynamic range: thus the function may be considered as a digital attenuator. this multiplication is applied to the rx signal after the adc decimation filter, but before offset adjustment and the 63-tap and 49-tap fir filters. this register sets the multiplier, the result being given by the formula: d d g 2 out in val 15 = ? ? where: d in d out g val is the signal input, is the signal output, is the value in the register. bit name active state function address $0x30 7:0 rxigainlsb data rw least significant 8 bits of the rxigain register ( g val ). address $0x31 7:3 rw reserved. set these bits low. undefined on read. 2:0 rxigainmsb data rw most significant 3 bits of the rxigain register ( g val ). address $0x34 7:0 rxqgainlsb data rw least significant 8 bits of the rxqgain register ( g val ). address $0x35 7:3 rw reserved. set these bits low. undefined on read. 2:0 rxqgainmsb data rw most significant 3 bits of the rxqgain register ( g val ). tetra baseband processor fx980 ? 1997 consumer microcircuits limited 64 d/980/3 address and data format for rxigain access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r d10 d9 d8 0 data field [10:0] 1 address field [6:0] 0 1 0 0 0 1 0 1 0 0 0 1 address and data format for rxqgain access d1 d0 d7 d6 d5 d4 d3 d2 r r r r r d10 d9 d8 0 data field [10:0] 1 address field [6:0] 0 1 0 1 0 1 0 1 0 1 0 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 65 d/980/3 rxiqoffset title: receive i and q channel offset correction register address: $0x32, $0x33, $0x36, and $0x37 (4 locatio ns) function: rw description: this register controls the rx data path signal offset. this offset is a 2s-complement value ( n offset ), which is applied to the rx signal after the gain multiplier ( g val ), but before the 63 -tap and 49-tap fir filters. the offset applied is at the discretion of the user. inappropriate values may cause arithmetic overflow in the subsequent operator sections. the result is given by the formula: d d + n 2 out in offset 15 = ? ? where: d in d out n offset is the signal input, is the signal output, is the 2s-complement value in the register. bit name active state function address $0x32 7:0 rxioffsetlsb data rw least significant 8 bits of the rxioffset register ( n offset ). address $0x33 7:4 rw reserved. set these bits low. undefined on read. 3:0 rxioffsetmsb data rw most significant 4 bits of the rxioffset register ( n offset ). address $0x36 7:0 rxqoffsetlsb data rw least significant 8 bits of the rxqoffset register ( n offset ). address $0x37 7:4 rw reserved. set these bits low. undefined on read. 3:0 rxqoffsetmsb data rw most significant 4 bits of the rxqoffset register ( n offset ). tetra baseband processor fx980 ? 1997 consumer microcircuits limited 66 d/980/3 address and data format for rxioffset access d1 d0 d7 d6 d5 d4 d3 d2 r r r r d11 d10 d9 d8 0 data field [11:0] 1 address field [6:0] 0 1 0 0 1 1 0 1 0 0 1 1 address and data format for rxqoffset access d1 d0 d7 d6 d5 d4 d3 d2 r r r r d11 d10 d9 d8 0 data field [11:0] 1 address field [6:0] 0 1 0 1 1 1 0 1 0 1 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 67 d/980/3 rxdataaccess title: rx data path access point. address: $0x38 to $0x3b (mapped over 4 locations) function: rw description: this register block allows direct access to the rx data path values just after the 59-tap (rx anti- alias) filter. both read and write operations are permitted. a read operation reads the signal values on the i and q channels. a write operation will write data to the rx data path operator output. to prevent normal rx data overwriting this value the rxdpaccesssel bit in the loopbackctrl register should be set active. the msb read data register is buffered to enable access of a discrete sample value (if this register was not buffered, data from different sample periods could be in the msb and lsb registers). therefore the lsb register must be read first for correct operation. bit name active state function address $0x38 7:0 rxdpidatalsb data rw least significant 8 bits of the rxdpidata register. this register must be read before its associated msb register. address $0x39 7:0 rxdpidatamsb data rw most significant 8 bits of the rxdpidata register. address $0x3a 7:0 rxdpqdatalsb data rw least significant 8 bits of the rxdpqdata register. this register must be read before its associated msb register. address $0x3b 7:0 rxdpqdatamsb data rw most significant 8 bits of the rxdpqdata register. address and data format for rxdpidata access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 0 1 1 0 0 1 0 1 1 0 0 1 address and data format for rxdpqdata access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 0 1 1 0 1 1 0 1 1 0 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 68 d/980/3 bistcontrol title: built in self test control register address: $0x62 function: rw description: this register block allows control of bist operations. bit name active state function 7 testcompleteack high/ low rw this bit is set by the user and cleared by the bist controller when a bist cycle has been completed. 6 n_rampdelayen low rw allow ramp control signal delay. this delay is required for normal operations, by matching the fir filter delays. for bist operations it can be disabled thus reducing bist test time. 5 bistdataratehi high rw selects bist data rate = 2.34 mhz default rate (low) = 1.44 khz 4 bisten high rw enables bist operations. 3 continuousbist high rw selects continuous bist mode. default (low) selects single cycle mode. 2 enrxdigitalfeedback high rw selects rx digital loop feedback for 49-tap tx fir input data. default (low) selects normal tx data. 1 en49tlqdata high rw selects bist data for 49-tap tx fir filter input. default (low) selects normal data. 0 ensymtestdata high rw selects bist data for 79-tap fir filter input. default (low) selects normal data. address and data format for bistcontrol access d1 d0 d7 d6 d5 d4 d3 d2 data field [7:0] address field [6:0] 1 0 1 1 0 0 0 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 69 d/980/3 bistprsg title: built in self test pseudo random sequence generator address: $0x60 to $0x61 (2 locations) function: rw description: this register block allows control of bist operations. this 16-bit number controls the length of the bist data sequence. it is the initial value (or seed) written to the pseudo-random sequence generation logic. the length of the bist data sequence is a function of the feedback logic equation and this initial value. the feedback function is fixed so run lengths are therefore controlled by this value. which values to apply to give specific run lengths can be determined from a look-up table. this table may be provided on request. bit name active state function address $0x60 7:0 bistprsglsb data rw least significant 8 bits of the bistprsg register. this register must be read before its associated msb register. address $0x61 7:0 bistprsgmsb data rw most significant 8 bits of the bistprsg register. this register must be read after its associated lsb register. address and data format for bistprsg access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 1 0 0 0 0 1 1 0 0 0 0 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 70 d/980/3 bistcrcregisters title: built in self test cyclic redundancy code checking registers address: $0x64 to $0x6d (10 locations) function: rw description: this register block allows bist crc checksums to be read. bit name active state function address $0x64 7:0 79tapi_crclsb data rw transmit i channel 79-tap filter lsb register. address $0x65 7:0 79tapi_crcmsb data rw transmit i channel 79-tap filter msb register. address $0x66 7:0 79tapq_crclsb data rw transmit q channel 79-tap filter lsb register. address $0x67 7:0 79tapq_crcmsb data rw transmit q channel 79-tap filter msb register address $0x68 7:0 sdm_crclsb data rw transmit sdm dac lsb register. address $0x69 7:0 sdm_crcmsb data rw transmit sdm dac msb register. address $0x6a 7:0 rxi_crclsb data rw receive i channel lsb register. address $0x6b 7:0 rxq_crclsb data rw receive i channel msb register. address $0x6c 7:0 rxq_crclsb data rw receive q channel lsb register. address $0x6d 7:0 rxq_crcmsb data rw receive q channel msb register. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 71 d/980/3 address and data format for 79-tap i channel crc reg access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 1 0 0 1 0 1 1 0 0 1 0 1 address and data format for 79-tap q channel crc reg access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 1 0 0 1 1 1 1 0 0 1 1 1 address and data format for sdm crc reg access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 1 0 1 0 0 1 1 0 1 0 0 1 address and data format for rx i channel crc reg access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 1 0 1 0 1 1 1 0 1 0 1 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 72 d/980/3 address and data format for rx q channel crc reg access d1 d0 d7 d6 d5 d4 d3 d2 d15 d14 d13 d12 d11 d10 d9 d8 0 data field [15:0] 1 address field [6:0] 1 0 1 1 0 1 1 0 1 1 0 1 tetra baseband processor fx980 ? 1997 consumer microcircuits limited 73 d/980/3 1.6 application notes tbd 1.6.1 general 1.6.2 transmitter 1.6.3 receiver 1.6.4 timing tetra baseband processor fx980 ? 1997 consumer microcircuits limited 74 d/980/3 1.7 performance specification 1.7.1 electrical performance absolute maximum ratings exceeding these maximum ratings can result in damage to the device. min. max. units supply v dd - v ss -0.3 7.0 v v cc1 - v ss1 -0.3 7.0 v v cc2 - v ss2 -0.3 7.0 v v cc3 - v ssb -0.3 7.0 v v dd1 - v ssa -0.3 7.0 v voltage on any pin to v ss -0.3 v dd + 0.3 v voltage on any pin to v ss1 -0.3 v cc1 + 0.3 v voltage on any pin to v ss2 -0.3 v cc2 + 0.3 v voltage on any pin to v ssa -0.3 v dd1 + 0.3 v voltage on any pin to v ssb -0.3 v cc3 + 0.3 v current into or out of v dd , v cc1 , v cc2 , v cc3 , -30 +30 ma v dd1 , v ss ,v ss1 , v ss2 , v ssb and v ssa current into or out of any other pin -20 +20 ma voltage differential between power supplies (v dd , v cc1 , v cc2 , v cc3 and v dd1 ) 0 0.3 v (v ss , v ss1 , v ss2 , v ssb and v ssa ) 0 50 mv l6 package min. max. units total allowable power dissipation at tamb = 25c 800 mw ... derating 13 mw/c storage temperature -55 +125 c operating temperature -40 +85 c # package min. max. units total allowable power dissipation at tamb = 25c 550 mw ... derating 9 mw/c storage temperature -55 +125 c operating temperature -40 +85 c operating limits correct operation of the device outside these limits is not implied. notes min. max. units supply v dd - v ss 4.5 5.5 v v cc1 - v ss1 4.5 5.5 v v cc2 - v ss2 4.5 5.5 v v cc3 - v ssb 4.5 5.5 v v dd1 - v ssa 4.5 5.5 v operating temperature -40 +85 c mclk frequency tbd tbd mhz tetra baseband processor fx980 ? 1997 consumer microcircuits limited 75 d/980/3 operating characteristics for the following conditions unless otherwise specified: mclk frequency = 9.216mhz, symbol rate = 18k bits/sec, (v dd - v ss ) = (v cc1 - v ss1 ) = (v cc2 - v ss2 ) = (v cc3 - v ssb ) = (v dd1 - v ssa ) = 3.0v to 3.6v for 3.3v parameters, 4.5v to 5.5v, for 5.0v parameters. tamb = - 40c to +85c. at 5v, bias/ctrl = 0 and at 3.3v bias/ctrl = 1 in powerdownctrl register will optimise the analogue performance in the tx and rx sections. it is assumed that all powersave and clock stop bits are set, where appropriate. notes min. typ. max. units 5v dc parameters (mclk not toggled) i dd (tx powersaved) 1 20 ma i dd (rx powersaved) 1 20 ma i dd ( aux powersaved) 1 34 ma i dd (all powersaved) 1 <0.05 ma i dd (not powersaved) 1 36 ma 5v a c parameters (mclk at 9.216mhz) i dd (tx powersaved) 1 40 ma i dd (rx powersaved) 1 35 ma i dd ( aux powersaved) 1 62 ma i dd (all powersaved) 1 12 ma i dd (not powersaved) 1 64 ma 3.3v dc parameters (mclk not toggled) i dd (tx powersaved) 1 18 ma i dd (rx powersaved) 1 18 ma i dd ( aux powersaved) 1 31 ma i dd (all powersaved) 1 <0.05 ma i dd (not powersaved) 1 32 ma 3.3v a c parameters (mclk at 9.216mhz) i dd (tx powersaved) 1 31 ma i dd (rx powersaved) 1 33 ma i dd ( aux powersaved) 1 48 ma i dd (all powersaved) 1 7.5 ma i dd (not powersaved) 1 49 ma mclk input 'high' pulse width 3 40 ns 'low' pulse width 3 40 ns input impedance (at 100hz) 10 m w notes: 1. not including any current drawn from the device pins by external circuitry. 3. timing for an external input to the mclk pin. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 76 d/980/3 transmit parameters parameter typ units conditions/comments input bit rate no. of channels modulation type rrc roll-off coefficient (a) |h(f)| 0 - 5.85khz |h(f)| @ 9khz |h(f)| @ 10.05khz |h(f)| @ 12.15khz max spurii @ 16khz @ 25khz @ 50khz @ 75khz fir filter sampling rate dac output update rate dac resolution integral accuracy differential accuracy offset gain matching, i to q gain matching, (i or q) to ideal tx phase matching, i to q storage time active power vector error (rms., typical) vector error (rms., max) vector error (peak, typical) vector error (peak, max) i,q output level vcc = 5.0v vcc = 3.3v 36 2 p /4 dqpsk 0.35 0 0.3 -3 0.3 -6 1 < -30 -60 -68 -78 -80 144 2.304 12 < 0.5 < 0.25 < 25 < 0.3db < 0.3 < 0.5 < 18 <105 0.017 0.025 0.045 0.07 2.5 1.65 kbps db db db db dbc dbc dbc dbc khz mhz bits lsb lsb mv db db degrees symbols mw v v 2 bits/symbol i and q 0 db corresponds to 1v pk-pk relative to maximum passband signal level without adjustment without adjustment normalised, 0 - 9khz after adjustment, 0 - 9khz 3.3v, rx, aux powered down vector errors measured with ideal if and rf sections after gain and offset adjustment, and specified as a fraction of the nominal vector value. peak to peak, differential at maximum gain notes: all parameters refer to the entire tx baseband i and q channels, unless otherwise indicated. a gain multiplier function allows independent proportional control of each channel. the multiplier is a 12 -bit word for each channel, input via the serial interface, representing a value from 0 to 1. this multiplication is applied to the signals from the fir filters. offset adjustment for each channel is available by loading a 12-bit word into the transmit offset register via the serial interface. tetra baseband processor fx980 ? 1997 consumer microcircuits limited 77 d/980/3 receive parameters parameter typ units conditions/comments input impedance differential input voltage range vcc=5.0v vcc=3.3v 3rd order intercept with internal anti-alias filter disabled:- anti -alias requirements @ 130khz @ 2.3mhz adc sampling rate adc resolution integral accuracy differential accuracy rrc roll-off coefficient (a) |h(f)| 0 - 5.85khz |h(f)| @ 9khz |h(f)| @ 10.05khz |h(f)| @ 12.15khz |h(f)| @ 16khz |h(f)| @ 25khz |h(f)| @ 50khz |h(f)| @ 75khz fir filter sampling rate output rate (16 bit words per channel) - selectable offset gain matching, i to q phase matching, i to q storage time active power with internal anti-alias filter enabled:- anti -alias requirements @ 130khz @ 2.3mhz active power <10 > 100 2.8 1.8 tbd <-30 <-120 2.304 16 < 1 < 1 0.35 0 0.2 -3 0.2 -6 1 < -30 < -70 < -70 < -80 < -90 2.304 144 144 or 72 < 10 < 0.1 < 0.5 < 15 <100 <-30 <110 pf k w v pk-pk ( typ) db db mhz bits lsb lsb db db db db db db db db mhz khz khz khz mv db degrees symbols mw db mw capacitive load to v ss1 or v ss2 source should be < 1000 w note this means 0.7v or 0.45v on each input of the differential pair. w.r.t. max. input level w.r.t. max. input level 0 db corresponds to 1v pk-pk decimation sections rrc sections output via the serial interface at 4.608 mhz or 2.304mhz without adjustment without adjustment, 0 -10khz 0 - 10khz 3.3v power supply use network shown in figure 2 w.r.t. max. input level tetra baseband processor fx980 ? 1997 consumer microcircuits limited 78 d/980/3 notes: offset adjustment for each channel is available by loading a 16-bit word into the receive offset register via the serial interface. optimally, anti-alias filtering should be carried out as much as possible prior to any agc function before the receive inputs. this allows the agc to act on a reduced bandwidth signal and thereby improve the relative magnitude of the wanted part. the device has been designed to reduce the complexity of any external anti- alias filter as much as possible and a 4-pole butterworth with a -3db point at about 60khz should be adequate. the internal anti-alias filter, if used, cannot provide the required 110db attenuation at 2.3mhz and must be supplemented by external filtering. the most simple supplementary system may be a one- or two-pole filter before the agc and an rc network after the agc with a -3db point on each filter of about 200khz. auxiliary circuit parameters parameter typ units conditions/comments dacs resolution settling time to 0.5 lsb output resistance integral non-linearity differential non-linearity zero error (offset) power (all dacs operating) minimum resistive load rms output noise voltage in 30khz bandwidth adc and multiplexed inputs maximum input source impedance resolution maximum input signal "linear rate of change" for < 1 bit error conversion time integral non-linearity differential non-linearity zero error (offset) a-d clock frequency input capacitance power 10 <10 <250 <4 <1 20 <10 5 10 25 10 0.27 12 <2 <1 20 mclk/8 <5 <3 bits m sec w bits bit mv mw k w m v k w bits mv/ m s m sec bits bit mv (hz) pf mw worst case large signal transition guaranteed monotonic gives < 1 bit error no missing codes tetra baseband processor fx980 ? 1997 consumer microcircuits limited 79 d/980/3 1.7.1. electrical performance timing diagrams the following timings are provisional: 1.7.1.1 serial ports timing parameter marker min max units mclk to sclk out - low to high t cslh 15 50 ns mclk to sclk out - high to low t cshl 10 35 ns cmddat set-up to falling edge of sclk t sis 35 ns cmdfs set-up to falling edge of sclk t sis 35 ns cmddat hold from fall edge of sclk t sih 0 ns cmdfs hold from fall edge of sclk t sih 0 ns rxdat propagation from rising edge of sclk t sop 5 ns rxfs propagation from rising edge of sclk t sop 5 ns cmdrddat propagation from rising edge of sclk t sop 5 ns cmdrdfs propagation from rising edge of sclk t sop 5 ns rxdat hold from rising edge of sclk t soh -5 ns rxfs hold from rising edge of sclk t soh -5 ns cmdrddat hold from rising edge of sclk t soh -5 ns cmdrdfs hold from rising edge of sclk t soh -5 ns ** cmd port in bi-dir mode ** cmddat propagation from rising edge of sclk t sop 7 ns cmddat hold from rising edge of sclk t soh -7 ns figure 4 serial port interfaces - timing parameters tetra baseband processor fx980 ? 1997 consumer microcircuits limited 80 d/980/3 figure 5a basic serial port signals tetra baseband processor fx980 ? 1997 consumer microcircuits limited 81 d/980/3 figure 5b command write operation tetra baseband processor fx980 ? 1997 consumer microcircuits limited 82 d/980/3 figure 5c bi-dir command read operation tetra baseband processor fx980 ? 1997 consumer microcircuits limited 83 d/980/3 figure 5d non bi-dir command read operation tetra baseband processor fx980 ? 1997 consumer microcircuits limited 84 d/980/3 figure 5e rx data serial port read operation tetra baseband processor fx980 ? 1997 consumer microcircuits limited 85 d/980/3 1.7.2 packaging tetra baseband processor fx980 handling precautions: this product includes input protection, however, precautions should be taken to prevent device damage from electro-static discharge. cml does not assume any responsibility for the use of any circuitry described. no ipr or circuit patent licences are implied. cml reserves the right at any time without notice to change the said circuitry and this product specification. cml has a policy of testing every product shipped using calibrated test equipment to ensure compliance with this product specification. specific testing of all circuit parameters is not necessarily performed. consumer microcircuits limited 1 wheaton road telephone: +44 1376 513833 witham - essex telefax: +44 1376 518247 cm8 3td - england e-mail: sales@cmlmicro.co.uk http://www.cmlmicro.co.uk |
Price & Availability of FX980L6
 |
|
|
|
|
All Rights Reserved © IC-ON-LINE 2003 - 2022 |
| [Add Bookmark] [Contact Us] [Link exchange] [Privacy policy] |
|
Mirror Sites : [www.datasheet.hk]
[www.maxim4u.com] [www.ic-on-line.cn]
[www.ic-on-line.com] [www.ic-on-line.net]
[www.alldatasheet.com.cn]
[www.gdcy.com]
[www.gdcy.net] |