Abstract: A block polyphase filter is constructed of a set of filter blocks having different filter functions, and being arranged for parallel processing of portions of an input sequence of signals. Signals of the input sequence are divided among the blocks by a demultiplexer for processing at a clock frequency lower than a clock frequency of the input signal sequence. The filter blocks are arranged in groups, wherein output signals of the blocks in any one group are summed to produce an output signal of the filtered group. Output signals of all of the filter groups are multiplexed to provide an output signal sequence wherein the repetition frequency of the signals may be higher, lower, or equal to the repetition frequency of the input signal sequence depending upon the ratio of the number of filter groups to the number of filter blocks in the set of filter blocks.

Abstract: A circuit 30 for upsampling and upconverting a high rate signal that is divided into M in-phase (I) symbols and M quadrature (Q) symbols. A Nyquist filter 32 upsamples by a factor of k each of the 2M symbols in parallel during one system clock period (CP). The filter 32 has a plurality of 2kM filter components 40, 42, that each provides a continuous output. A plurality of multipliers 50, 52 each upconverts a filter component output with a carrier wave signal 46, 48 that is output from a numerically controlled oscillator 44. A plurality of adders 54 each adds the output of two multipliers 50 to recombine corresponding I and Q samples to output kM samples during a CP. For continuous phase modulation, N parallel bits are input into the filter 32, upsampled in one CP, and accumulated and modulated 82 in parallel in one CP. For analog processing, M (I) and M (Q) symbols are input into an FIR filter 77a, 77b for upsampling, and decimated at a MUX/DAC block 78 for subsequent analog upconversion.

Abstract: A demodulator, suitable for use in a communication system and in a modem, has a block polyphase circuit with circuit blocks for different signal processing functions, particularly filtering, delay, and frequency conversion. The circuit blocks are arranged for parallel processing of different portions of an input sequence of signals. Signals of the input sequence to be filtered are divided among the blocks by a demultiplexer for processing at a clock frequency lower than a clock frequency of the input signal sequence. Signals outputted by groups of the circuit blocks are summed to produce an output signal of the group. Output signals of all of the groups are multiplexed to provide an output signal sequence such that the repetition frequency of the outputted signals may be higher, lower, or equal to that of the input signal sequence. This enables use of programmable circuitry operative at reduced clock rates.

Abstract: A circuit 30 for upsampling and upconverting a high rate signal that is divided into M in phase (I) symbols and M quadrature (Q) symbols. A Nyquist filter 32 upsamples by a factor of k each of the 2M symbols in parallel during one system clock period (CP). The filter 32 has a plurality of 2kM filter components 40, 42, that each provides a continuous output. A plurality of multipliers 50, 52 each upconverts a filter component output with a carrier wave signal 46, 48 that is output from a numerically controlled oscillator 44. A plurality of adders 54 each adds the output of two multipliers 50 to recombine corresponding I and Q samples to output kM samples during a CP. For continuous phase modulation, N parallel bits are input into the filter 32, upsampled in one CP, and accumulated and modulated 82 in parallel in one CP. For analog processing, M (I) and M (Q) symbols are input into an FIR filter 77a, 77b for upsampling, and decimated at a MUX/DAC block 78 for subsequent analog upconversion.

Abstract: A circuit for single or parallel digital fractional interpolation of data samples has a fractional interpolator filter, an oscillator for outputting timing signals to the fractional interpolator filter, and a detector loop with a strobe feedback from the oscillator for outputting a frequency adjustment to the oscillator. Three different approaches are shown to determine the frequency adjustment. One approach is to generate a pulse based on the symbol clock, and measure the differences between the pulse and the strobe and between the strobe and the pulse. The smaller is the frequency adjustment. Another approach is to adjust the strobe period to match the symbol clock period. A third approach is to add an oscillator-driven clock to the symbol clock and integrate the sum over a symbol clock period to generate the frequency adjustment.

Abstract: A circuit and method for correcting phase of a received phase modulated (PM) signal. The method uses k most recently received data bits, which alternate between in-phase I and quadrature Q bits, as an address for a lookup table 60. The lookup table outputs a phase figure 62 derived from a reconstructed waveform. When the most recent k bit is a Q bit, the complement 68 of the phase figure 62 is calculated to yield a phase correction. Otherwise, the phase figure is the phase correction, which is applied to adjust the phase of a delayed version of the received signal. The delayed, phase adjusted signal is then applied to correct the phase of a received signal. The circuit splits an input PM signal in parallel between a matched filter 54 and a delay block 76, 88. The matched filter output provides the input to a register 58 for storing the k data bits. The delay block holds the PM signal until it is input into a loop phase shifter 78 synchronously with the phase correction.

Abstract: A circuit and method for correcting timing of a received phase modulated signal. The method uses k most recently received data bits as an address for a lookup table 60. The lookup table includes reconstructed waveforms from which a timing weighing factor is determined. The received PM from time t1 is delayed, phase adjusted, and multiplied by the timing weighing factor, the product of which is used by a timing adjust block 50 to adjust timing of the PM signal at a time after t1. The circuit inputs a PM signal to a timing adjust block 50. The output is split between a matched filter 54 and a loop phase shifter 78. The matched filter feeds alternating I and Q bits into a register 58 that holds k data bits, which are used as an address for a lookup table 60. The output of the lookup table 60 becomes a timing weighing figure, which is multiplied 74 with an output of the loop phase shifter 78 and then input into the timing adjust block 50 for adjusting timing of a PM signal.