• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    gfsk receiver architecture and methodology. The present invention relates to a Gaussian frequency shift (gfsk) receiver which includes a receiver and forward end for receiving a gfsk modulated signal and converting the gfsk modulated received signal to a baseband modulated frequency signal, a channel filter for reducing channel interference which is adjacent to a desired channel of the frequency modulated baseband signal, a demodulator for demodulating the filtered channel baseband modulated signal and for recovering a sequence of symbols, a digital filter to reduce intersymbol (isi) interference from the symbol sequence, a segmenter to produce symbol decisions based on the filtered symbol sequence, and a symbol-to-bit mapper to map symbol decisions to data bits.
    公开号:BR112013002315B1
    申请号:R112013002315-5
    申请日:2011-05-03
    公开日:2021-07-20
    发明作者:Robert E. Rouquette
    申请人:Sensus Usa Inc.；
    IPC主号:
    专利说明:

    Background of the Invention Field of Invention
    [001] This application claims priority to US Application Serial No. 12/847,951, filed July 30, 2010, the entire contents of which are incorporated herein by reference.
    [002] The present description concerns the field of communications and, specifically, the field of data communication, through Frequency Switching (FSK) modulation.
    [003] Gaussian Frequency Switching (GFSK) is a type of bandwidth efficient digital modulation by FSK. Specifically, GFSK modulation uses a pulse-shaping Gaussian filter to reduce the bandwidth of a modulated transmission vehicle. In FSK modulation, a sequence of data symbols with sharp transitions results in a modulated transmission vehicle having frequency discontinuities. Frequency discontinuities result in a broadband transmission vehicle. Smoothing the sharp transitions of the data symbol sequence, however, using a Gaussian pulse-shaping filter, circumvents this problem. The pulse-shaping Gaussian filter removes the higher frequency components in the data symbol sequence which, in turn, allows for a more compact transmission spectrum.
    [004] The compact transmission spectrum facilitated by the GFSK modulation scheme aids wireless communication systems operating in both licensed and unlicensed industrial, scientific, and medical (ISM) bands, reducing spectral bandwidth and spectrum GFSK transmission vehicle out-of-band to meet FCC adjacent channel power rejection requirements. Similar requirements are enforced by international radio spectrum regulatory bodies.
    [005] However, pulse shaping through Gaussian pulse shaping filter induces intersymbol interference (ISI). In fact, it is the formation of pulses by the Gaussian filter that introduces ISI. Consequently, systems designed around the GFSK modulation scheme are designed with low data transfer or higher bit error rate in mind. Conventionally, the ISI associated with the GFSK modulation scheme prohibits data communication at high modulation orders, where multiple bits of data are transmitted per symbol. In an attempt to facilitate GFSK communication systems with greater data transmission, the use of more complex and expensive receiver structures has been proposed. Discussion of Related Prior Art
    [006] As illustrated in Figure 1, a first GFSK 100 system of the related art includes a GFSK 102 transmitter and a GFSK 114 receiver.
    [007] The GFSK transmitter 102 includes a data source 104, a Gaussian filter 106, an FSK modulator 108, a tail-end transmitter 110, and a transmit antenna 112. The Gaussian filter 106 filters a given data symbol sequence of data source 104, and outputs a sequence of pulse-shaped data symbols to the FSK modulator 108. The FSK modulator 108 modulates the vehicle frequency based on the sequence of pulse-shaped data symbols according to an order of selected FSK modulation (ie a number of bits per symbol). The output of the FSK modulator 108 is provided to the rear end of the transmitter 110, where it is upconverted to a transmit frequency and coupled to the transmit antenna 112 for radio frequency (RF) transmission. In this way, the transmit antenna 112 transmits a GFSK modulated transmit vehicle.
    [008] The GFSK receiver 114 includes a receive antenna 116, a receiver front end 118, a channel filter 120, a discriminator 122, a post-detect filter 124, a symbol segmenter 126, and a data collector 128 In operation, the receiving antenna 116 and the front receiver 118 receive a transmitted GFSK modulated signal and downconvert the received GFSK modulated signal to baseband. Channel filter 120 selectively filters the received baseband GFSK modulated signal to reject adjacent channel interference and additive white Gaussian noise (AWGN). Discriminator 122 performs frequency demodulation, providing an output signal that is proportional to the instantaneous frequency of the modulated transmission vehicle and generates a sequence of demodulated symbols. specifically, in the case of a 1 bit/symbol modulation order (ie, 2-GFSK), the discriminator 122 discriminates between two frequencies, f0 + f1 and f0 - f1, where f0 is the unmodulated vehicle frequency. The post-detection filter 124 filters the demodulated symbol sequence produced by the discriminator 122 to reduce the amplified noise by the discriminator 122. The segmenter 126 produces symbol decisions based on the filtered symbol sequence output from the post-detection filter 124, to produce a sequence of symbol decisions, which is provided to data collector 128. At gfsk receiver 114, post-sensing filter 124 is not designed to remove ISI, and segmenter 126 is required to produce symbol decisions at presence of ISI, causing symbol and bit errors to occur.
    [009] In the first GFSK 100 system of the related art, the ISI introduced by the Gaussian filter 106 requires that a modulation scheme of low order modulation (ie, some bits/symbols) be used by the FSK 108 modulator. unacceptable symbol and bit errors will occur at the GFSK receiver 114. Specifically, the ISI introduced by the Gaussian filter 106 causes the "eye" of the demodulated symbol sequence output by the discriminator 122 to close, and thus the segmenter 126 will produce wrong symbol decisions, since the output of discriminator 122 will fail to be consistently above or below the symbol decision threshold(s) of segmenter 126 for certain at given symbol intervals. At higher orders of modulation, it becomes even more difficult for discriminator 122 and segmenter 126 to make correct symbol decisions. In this way, the data transfer rate of the first GFSK system 100 of the related art is limited because of the ISI introduced by the Gaussian filter 106, as only the lowest order modulation schemes can be used without unacceptable levels of symbol errors. The filter channel 120 also contributes to the introduction of ISI in the received signal, further compounding the limitations of the GFSK 100 system.
    [010] As illustrated in Figure 2, a second GFSK 200 system of the related art includes a GFSK 202 transmitter and a GFSK 214 receiver.
    [011] The GFSK 202 transmitter includes a data source 204, a Gaussian filter 206, an FSK 208 modulator, a tail tip transmitter 110, and a transmit antenna 212. The GFSK transmitter 202 operates in the same way as the first transmitter GFSK 102 of the related art.
    [012] The GFSK receiver 214 includes a receive antenna 216, a receiver front end 218, a channel filter 220, a discriminator 222, a maximum likelihood sequence estimator (MLSE) 224, and a data collector 226. In comparison to the first related art GFSK 114 receptor, the second related art GFSK 214 receptor depends on the MLSE 224 estimator to produce symbol decisions in the presence of ISI. That is, the MLSE 224 estimator does not remove the ISI. Instead, MLSE estimator 224 estimates data symbols according to at least one error probability, in the presence of the ISI, and outputs data bits in terms of error probability. For example, the MLSE estimator 224 can use the algorithm to determine a lower error probability symbol decision in an attempt to mitigate the presence of ISI. However, especially at low signal-to-noise ratios (SNR), MLSE estimators cannot adequately mitigate symbol errors due to ISI. Brief Summary of the Invention
    [013] Consequently, an object of this invention is to provide a communications receiver, a communications receiver method, and a computer-readable storage medium storing in it, computer-readable instructions, which present a simple and low-cost approach to high throughput data communication, even in a low SNR environment.
    [014] According to an aspect of the present invention, there is provided a communications receiver including a receiver front end for receiving a modulated signal and converting the modulated signal to a modulated baseband signal, a channel filter to reduce channel interference, which is adjacent to a desired channel of the baseband modulated signal of the baseband modulated signal and producing a filtered channel baseband modulated signal, a demodulator for demodulating the baseband modulated signal of filtered channel and to retrieve a symbol sequence, a digital filter to reduce inter-symbol interference of the symbol sequence, a segmenter to produce symbol decisions based on the filtered symbol sequence, and a symbol-to-bit mapper to map the data bits symbol decisions.
    [015] According to a further aspect of the present invention, there is provided a communications receiver method which includes receiving a modulated signal and converting the modulated signal to a modulated baseband signal, filtering channel interference, which is adjacent to a desired channel for the baseband modulated signal to reduce the channel interference of the baseband modulated signal and to produce a baseband modulated signal filtered channel, demodulating the baseband modulated signal filtered channel to recover a symbol sequence, filtering, by a processor of a data processing apparatus, the symbol sequence to reduce inter-symbol interference of the symbol sequence, producing symbol decisions based on the filtered symbol sequence, and mapping symbol decisions to bits of data.
    [016] According to another aspect of the present invention, there is provided a computer-readable storage medium, storing computer-readable instructions which, when executed by a processor of a communications receiver, direct the processor to perform reception of a signal. modulated and converting the modulated signal to a modulated baseband signal, filtering channel interference, which is adjacent to a desired channel for the modulated baseband signal, so as to reduce the channel interference of the modulated band signal and to produce a baseband modulated signal filtered channel, demodulating the baseband modulated signal filtered channel to recover a sequence of symbols, filtering the symbol sequence to reduce inter-symbol interference (ISI) of the sequence of symbols, producing symbol decisions based on the filtered symbol sequence, and mapping the symbol decisions to data bits. Brief Description of Drawings
    [017] A fuller estimate of the invention and many of the advantages inherent thereto will be readily obtained as it becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings, in which:
    [018] Figure 1 illustrates a first prior art GFSK communication system;
    [019] Figure 2 illustrates a second prior art GFSK communication system;
    [020] Figure 3 is a block diagram of a GFSK communication system;
    [021] Figure 4 is a graph that illustrates the response waveforms of the Gaussian filter filter with various BT products;
    [022] Figure 5 is a waveform diagram illustrating an FSK modulated waveform;
    [023] Figure 6 is a block diagram that illustrates the generation of coefficients of a digital filter;
    [024] Figure 7A is a graph illustrating a demodulated symbol sequence of a received 2-GFSK signal;
    [025] Figure 7B is a graph illustrating the sequence of symbols of Figure 7a after filtering;
    [026] Figure 7C is a graph illustrating a sequence of demodulated symbols from a received 8-GFSK signal;
    [027] Figure 7D is a graph illustrating the sequence of symbols in Figure 7C after filtering;
    [028] Figure 8 is a block diagram illustrating a digital filter;
    [029] Figure 9 is a flowchart illustrating a transmission method;
    [030] Figure 10 is a flowchart illustrating a reception method;
    [031] Figure 11 is a flowchart illustrating a digital filter method, and
    [032] Figure 12 is a schematic block diagram illustrating a modality of data processing apparatus. Detailed Description of the Invention
    [033] The present invention is directed to a communications receiver and a communications receiver method to substantially reduce and largely remove ISI introduced by pulse-shaping Gaussian filters of GFSK transmitters, so that higher FSK modulation requests can be used to increase the data transfer rate.
    [034] The described receiver and receiver method allow the use of Gaussian filters, which more aggressively shape symbol sequences as pulses, compared to conventionally used filters, to substantially mitigate and remove frequency discontinuities and to reduce width of busy transmission bandwidth. Furthermore, the receiver and receiver method described allow larger FSK modulation orders to be used, while higher order modulated symbol sequences are more aggressively shaped as pulses. In this way, the described receiver and receiver method achieve an increase in data transfer while reducing the occupied transmission bandwidth. These and other advantages are achieved by a GFSK receptor including a filter that responds to and removes pulse-molding induced ISI from Gaussian filters of GFSK transmitters.
    [035] Furthermore, the receiver and receiver method described substantially reduce and largely remove ISI introduced by transmitter modulators, receiver demodulators, and receiver channel filters.
    [036] Still, the receiver and receiver method described provide a communication system that incorporates the known advantages of GFSK modulation while reducing the processing steps of conventional GFSK receivers to arrive at a simpler, more efficient and cost effective GFSK receiver .
    [037] The receiver-based ISI removal and receiver method of the present invention are not limited to GFSK communication systems, but can be used to remove ISI and other undesirable communication artifacts from communication systems other than communication systems. GFSK communications, both wired and wireless.
    [038] A modality of a GFSK communications system that provides the above-described advantages will now be described with reference to Figure 3.
    [039] Fig. 3 illustrates a GFSK 300 communication system. The GFSK 300 communication system includes a GFSK 302 transmitter and a GFSK 314 receiver. The GFSK 302 transmitter and GFSK 314 receiver can be combined into a single communication unit as an integrated GFSK transceiver or can be supplied as separate communication units.
    [040] The GFSK transmitter 302 includes a data source 304, a Gaussian filter 306, an FSK 308 modulator, a tail-end transmitter 310, and a transmit antenna 312. The Gaussian filter 306 filters a supplied symbol sequence from of data source 304, and outputs a sequence of symbols in the form of pulses to the modulator FSK 308. The modulator FSK 308 modulates a frequency f0 based on the sequence of symbols in the form of pulses, according to a selected order of FSK modulation m (that is, a number of bits per symbol). The output of the FSK 308 modulator is provided to the tail-end transmitter 310, where it is converted to a transmit frequency and coupled to the transmit antenna 312 for wireless RF transmission. Transmit antenna 312 transmits a GFSK transmit signal modulated at the transmit frequency.
    [041] Arrangements of data source 304 may include a Forward Error Correction (fec) code generator, which adds redundant data to the symbol sequence, so that a GFSK receiver can correct errors without data retransmission using a FEC decoder .
    [042] Referring to Figure 4, the pulse molding characteristics of the Gaussian filter 306 are described. The response of the Gaussian filter 306 is described in terms of its product BT, where B is the -3 db half bandwidth of the filter and T is the symbol period (i.e., rate of 1/symbol rate) of an input symbol. Figure 4 illustrates the impulse response of Gaussian filters of BT = 0.3, 0.5, and 0.8, as wrapped with a rectangular pulse waveform of time period T and unity amplitude. The impulse responses of the Gaussian filters of BT = 0.3, 0.5, and 0.8 are illustrated, respectively, as output response waveforms 402, 404, and 406.
    [043] In Figure 4, the vertical axis or y axis represents the amplitude of the impulse responses, and the horizontal axis or x axis represents the base time of symbol T.
    [044] As illustrated in Figure 4, for lower values of product bt, output response waveforms 406, 404 and 402 become progressively more propagated over symbol time period T, representing increased introduction of ISI between symbols of data. Specifically, any output response waveform, having a duration greater than the symbol period T, corresponds to a Gaussian filter with a BT product which results in the introduction of ISI between symbols when provided with a symbol sequence at your entrance. As illustrated, the ISI between the five symbol periods approximates a BT of approximately 0.3.
    [045] In GFSK communication systems, Gaussian filters with low values of BT products, while introducing significant amounts of ISI, result in more compact and efficient bandwidth modulated transmission vehicles, which is preferable. Specifically, frequency discontinuities in the modulated transmission vehicle, which cause the transmission bandwidth of the modulated transmission vehicle to be undesirably wide, are removed through the use of Gaussian filters in a GFSK communication system. The lower the BT product of a Gaussian filter, the better the removal of frequency discontinuities.
    [046] In preferred embodiments, the use of pulse shaping of Gaussian filters with low BT products such as 0.36 or less is possible because of a GFSK receptor filter described below. Thus, in the preferred embodiment, the Gaussian filter 306 has a BT product of 0.36.
    [047] Referring again to Figure 3, the FSK 308 modulator modulates a vehicle f0 based on a modulation index h and the chosen FSK modulation order m.
    [048] For FSK modulation the selected FSK modulation order m, the modulation index h is defined as: 1. ) h = Δfm/symbol rate
    [049] Where fsymbol rate is the symbol rate and Δfm is the frequency spacing of adjacent symbols. Thus, the modulation index h indicates how much a modulated vehicle varies from its unmodulated frequency, f0. The modulation index h is also related to an amount of bandwidth that an FSK modulated vehicle occupies. A lower modulation index h refers to a lower frequency of occupied bandwidth, and a higher modulation index h refers to a higher frequency of occupied bandwidth. The susceptibility of a GFSK receptor to making wrong symbol decisions increases as the modulation index, h, decreases.
    [050] The FSK 308 modulator of the GFSK 314 receiver can operate at relatively high modulation orders, m, and at relatively low modulation indices, h, compared to those conventionally used. For example, the FSK 308 modulator can operate with modulation orders such as 2-GFSK, 4-GFSK, 8-GFSK, and 16-GFSK, and modalities can use odd modulation orders and modulation orders as high as 256-GFSK or superior. Furthermore, the FSK 308 modulator can operate at modulation indices, h, as low as 1/256.
    [051] However, the modalities are not limited to using FSK modulation, and the FSK 308 modulator can be implemented by a phase shift key modulator (PSK), a Quadrature Amplitude modulator (QAM), or equivalent of the same.
    [052] The 2-GFSK modulation order refers to 1 bit/symbol transmission. Thus, when operating in a 2-GFSK modulation order, the FSK 308 modulator modulates a frequency f0 by shift frequencies +/- f1 to arrive at two modulation symbols, f0 + f1 and f0 - f1. Referring to Figure 5, a sequence of data symbols 500, a non-modulated vehicle 502, and a modulated vehicle 504 are illustrated. Compared to unmodulated vehicle 502, modulated vehicle 504 describes a first FSK symbol 506, at a frequency f0 + f1, and a second FSK symbol 508, at a frequency f0 - f1.
    [053] In Figure 5, the vertical axis or y axis represents amplitude and the horizontal axis or x axis represents time.
    [054] Furthermore, the modulation orders of 4-GFSK, 8-GFSK, and 16-GFSK are based on an extension of 2-GFSK using additional multiples of the f1 frequency shift. That is, for 4-GFSK, which is a modulation order of 2 bits/symbol, the FSK 308 modulator is based on four modulation symbols, f0 + f1, f0 - f1, f0 + 3f1, and f0 - 3f1.
    [055] Referring again to Figure 3, the rear-end transmitter 310 converts the modulated vehicle output from the FSK 308 modulator into a frequency suitable for RF transmission.
    [056] The 310 rear-bridge transmitter includes circuit hardware components necessary to convert the f0 frequency modulated vehicle to the proper frequency for RF transmission. As part of a non-limiting group of hardware components, the rear-end transmitter 310 can include digital to analog converters (DACs), Voltage Controlled Oscillators (VCO), Phase Capture Loops (PLLs), mixers, analog filters , low noise amplifiers (LNAs), and other hardware components recognized as being used to convert a modulated vehicle to a frequency suitable for RF transmission.
    [057] As discussed above, the more compact and efficient bandwidth GFSK modulated transmission vehicles advantageously allow the operation of GFSK transmitters in both the licensed bands and the unlicensed ISM bands, providing a transmission signal that complies with the FCC Adjacent Channel Interference Regulations. In addition, more compact and efficient bandwidth modulated GFSK transmission vehicles advantageously allow the selection of transmission amplifiers, which are simple and cost effective. Thus, because backend transmission 310 transmits a compact bandwidth modulated transmission vehicle, as facilitated by Gaussian filter pulse shaping 306, backend transmission 310 can be designed using a class of amplifiers that are simple , low cost and efficient.
    [058] After conversion by tail-end transmitter 310, a GFSK modulated transmit vehicle signal is transmitted from transmit antenna 312.
    [059] GFSK 302 transmitter modalities can be implemented entirely in hardware as a combination of hardware circuits. Alternatively, the GFSK 302 transmitter can be implemented in a combination of hardware and software. For example, data source 304, Gaussian filter 306, and FSK 308 modulator can be implemented by a processor of a data processing apparatus that executes computer-readable instructions stored on a computer-readable medium as the transmitter of rear tip 310 and transmit antenna 312 can be implemented in hardware as a combination of hardware circuitry.
    [060] Again referring to the GFSK 300 communication system of Figure 3, the GFSK 314 receiver includes a receive antenna 316, a front end of receiver 318, a channel filter 320, an FSK demodulator 322, a digital filter 324, a segmenter 326, a bit-to-symbol mapper 328, and a data collector 330.
    [061] The GFSK receiver 314 receives a GFSK modulated transmit vehicle signal, such as the GFSK modulated transmit vehicle signal transmitted by the GFSK transmitter 302, at antenna 316, and downconverts the received GFSK modulated transmit vehicle signal at the front end of receiver 318, for a baseband modulated frequency signal.
    [062] The front end receiver 318 includes the necessary hardware circuitry components to convert the received GFSK modulated transmission vehicle signal to baseband. As part of a non-limiting group of hardware components, the front end receiver 318 may include analog to digital converters (ADCs), Controlled Voltage Oscillators (VCO), Phase Capture Loops (PLLs), mixers, analog filters, low noise amplifiers (LNAs), and other hardware components recognized as being used to convert a received broadcast vehicle signal.
    [063] The filter channel 320 selectively reduces channel interference adjacent to a desired channel from the baseband frequency modulated signal from the baseband modulated signal and produces a channel filtered baseband modulated signal. In particular, channel filter 320 reduces frequencies other than the frequency of the transmission vehicle. However, as the BT of channel filter 320 decreases, channel filter 320 induces additional ISI in the received signal as it reduces interference from the adjacent channel. Channel filter 320 may have a BT product of 0.75, 0.6, 0.5, or less. Compared to channel filters used in conventional GFSK receivers, channel filter 320 can more aggressively seek adjacent channel interference reduction using a filter with a lower BT product, and the additional ISI can be substantially reduced and large part removed by digital filter 324.
    [064] The frequency demodulator FSK 322 demodulates the channel filtered baseband modulated frequency signal to recover a sequence of symbols. That is, depending on the modulation order, m, used to modulate the GFSK modulated transmission vehicle signal, an FSK 322 frequency demodulator can distinguish between shifting frequency f0 into frequency by multiples (i.e., f1, 3f1, 5f1 , 7f1, etc.) of frequency shift f1. For example, in the case of a 1 bit/symbol modulation order (ie, 2-GFSK), the frequency demodulator 322 discriminates between the two frequencies f0 + f1 and f0 - f1, where f0 is the vehicle frequency, to produce a signal output. In the case of 2-GFSK, f0 + f1 can refer to a demodulated logic "1" and f0 - f1 can be related to a demodulated logic "0". Other modulation orders, such as 4-GFSK, 8-GFSK, and 16-GFSK can be performed based on additional multiples of frequency shift f1 as described above.
    [065] The FSK 322 frequency demodulator can be performed by any frequency demodulator providing an output proportional to the instantaneous frequency at its input. To achieve high fidelity for higher orders of modulation, the preferred embodiment of the FSK 322 frequency demodulator includes a digital signal processor (DSP) frequency demodulator that performs dΦ/dt in a wider bandwidth than the filter. channel 320, where Φ is the instantaneous phase at the input of frequency demodulator FSK 322. Frequency demodulator FSK 322 produces a retrieved symbol sequence.
    [066] The retrieved symbol sequence output from the FSK 322 frequency demodulator is subjected to ISI. The ISI is substantially caused by a Gaussian filter pulse shaping of the transmitter that transmitted the GFSK modulated transmission vehicle signal, such as the 306 Gaussian filter pulse shaping of the GFSK 302 transmitter. frequency frequency FSK322 is subject to ISI, the "eye" of the retrieved sequence of symbols closes. To remove the ISI, the demodulated data output signal is filtered by the digital filter 324 to remove the ISI, before symbol decisions are made by the segmenter 326. In this way, the digital filter 324 opens the "eye" of the sequence of symbols retrieved output from frequency demodulator FSK 322, so that symbol decisions made by segmenter 326 are made with less error, even at high modulation orders, m, and low SNR. Digital filter 324 substantially reduces and largely removes ISI based on a plurality of coefficients. Digital filter 324 also performs filtering similar to filtering performed by post-sensing filter 124 based on the plurality of coefficients.
    [067] A technique for determining the plurality of coefficients is described in association with Figure 6, below, and the structure and operation of digital filter 324 is described in association with Figure 8 below. Since the 324 digital filter removes ISI and also performs post-detect filtering, the GFSK 314 receiver is designed to be at least as simple and as cost effective as conventional GFSK communication systems, but with better performance.
    [068] Again referring to Figure 3, the segmenter 326 produces symbol decisions based on the symbol sequence filtered by the digital filter 324. Based on the filtered symbol sequence, the segmenter 326 is able to distinguish between multiple FSK symbols, without errors . Thus, the use of digital filter 324 allows the FSK 308 modulator of the GFSK 302 transmitter to operate based on modulation orders that are higher than those conventionally used. Thus, the data transmission rate is increased. Data transmission is also increased, due to the removal of the ISI by digital filter 324, because the segmenter 326 is able to make symbol decisions with less error, which results in fewer retransmissions.
    [069] After decisions are produced by segmenter 326, symbol-to-bit mapper 328 maps symbol decisions to data bits. Depending on the modulation order, m, a symbol input for symbol-to-bit mapper 328 can correspond to 1, 2, 4, or more bits of data. Symbol-to-bit mapper 328 may also map input symbols to an odd number of data bits. In addition, embodiments of segmenter 326 and/or symbol-to-bit mapper 328 may include an FEC decoder, which uses redundant data to correct errors without retransmission of the data. The output data bits of symbol-to-bit mapper 328 are provided to data collector 330.
    [070] GFSK 314 receiver modalities can be fully implemented in hardware as a combination of hardware circuits. Alternatively, the GFSK 314 receiver can be implemented in a combination of hardware and software. For example, the receive antenna 316 and the front end of the receiver 318 can be implemented in hardware as a combination of hardware circuits, while the channel filter 320, the FSK demodulator 322, the digital filter 324, the segmenter 326, the symbol to bit mapper 328, and the data collector 330 may be implemented by a processor of a data processing apparatus executing computer readable instructions stored on a computer readable medium.
    [071] Figure 6 is a flowchart illustrating operation 600 to determine a plurality of coefficients. Although Figure 6 is described with reference to determining filter coefficients for digital filter 324 of GFSK receiver 314, operation 600 can be applied to determine coefficients used in digital filters of receivers other than GFSK receiver 314. , the operation illustrated in Figure 6 can be used to determine coefficients to be used in the digital filter method 1100.
    [072] Referring to Figure 6, a rectangular pulse of time symbol T is provided at 602. The rectangular pulse is provided for the Gaussian filter 604. The output response of the Gaussian filter 604 will vary depending on the BT product of the Gaussian filter 604, as discussed above in relation to Figure 4. Gaussian filter 604 can be varied in BT product to produce digital filter coefficients based on a Gaussian filter of a particular BT product. In other words, the digital filter coefficients, determined according to operation 600, will vary depending on at least the BT product of the Gaussian filter 604. In the mode of operation 600 illustrated in Figure 6, the BT product of the Gaussian filter 604 is 0.36 and the output response of Gaussian filter 604 spans over a time period of 5T.
    [073] The output of the 604 Gaussian filter is provided to a 606 frequency modulator, which can be modeled based on the known properties of an FSK modulator, such as the FSK 308 modulator. The 606 frequency modulator can be based on a function modulation transfer rate of the FSK 308 modulator. The digital filter coefficients, determined in accordance with operation 600, will also vary depending on the modulation transfer function of the frequency modulator 606.
    [074] The output of frequency modulator 606 is provided to channel filter 608. Channel filter 608 can be varied in BT product, for example, depending on BT product of channel filter 320 of GFSK 314 receiver. of digital filter, determined in accordance with operation 600, will also vary depending on the BT product of the channel filter 608.
    [075] The output of channel filter 608 is input to frequency demodulator 610, which can be modeled based on known properties of an FSK demodulator, such as FSK demodulator 322. Frequency demodulator 610 can be based on a modulation transfer function of FSK demodulator 322. The digital filter coefficients, determined in accordance with operation 600, will also vary depending on the modulation transfer function of frequency demodulator 610.
    [076] Note that frequency modulator 606, channel filter 608, and frequency demodulator 610 may be omitted from 600 operation, particularly if channel filter 608 is of the linear phase type with a larger BT product or equal to 0.75 and whether the modulation transfer functions of both the frequency modulator 606 and the frequency demodulator 610 are unitary. In the preferred mode of operation 600, the BT of channel filter 608 is 0.75 or less.
    [077] In addition, the trailing tip transmitter and the leading end receiver can be included between the frequency modulator 606, the channel filter 608, and the frequency demodulator 610, so that the trailing tip transmitter responses and the receiver front end of a communication system can be represented in operation 600.
    [078] An output of frequency demodulator 610 is transformed to the frequency domain by a Fast Fourrier Transform (FFT) at 612. The FFT, at 612, generates a frequency domain representation of an aggregation response (including the response which induces ISI) of Gaussian filter 604, frequency modulator 606, channel filter 608, and frequency demodulator 610. The frequency domain representation of the aggregation response is included within a plurality of frequency compartments that are FFT outputs at 612.
    [079] Depending on whether frequency modulator 606, channel filter 608, and frequency demodulator 610 are included in operation 600, the aggregation response may vary accordingly. In addition, the aggregation response may vary depending on whether the transmitting back end and receiving front end are included in operation 600.
    [080] The magnitudes of each of the plurality of FFT output frequency bins 612 are provided to a comparator 614 for comparison with a predetermined value δ. Specifically, a comparison takes place at comparator 614 such that if the magnitude of a frequency bin is less than or equal to δ, the frequency bin is defined to be equal to δ. Otherwise, the frequency compartment is unchanged. After comparator 614, the updated frequency slots are provided as a divider for a first input of a divider 616.
    [081] Comparison on comparator 614 prevents noise gain. In the case of operation 600, the comparison in comparator 614 prevents noise gain by preventing the divisor of divisor 616 from being too close to 0. The value of δ has been recognized as an effective outcome variable for high-frequency roll-off of a digital filter that uses the coefficients determined by operation 600. Specifically, the value of δ used in operation 600 can be varied to achieve a digital filter that has a rolloff response of the desired high frequency, without deteriorating the response of the filter. Nyquist, which is required for ISI removal. In selecting the δ to determine the desired high roll-off frequency of the digital filter, post-detect filtering is achieved by the high roll-off response. Thus, the comparison at comparator 614 and the selection of δ confer the functionality of the post-detection filter 124. In the preferred embodiment of operation 600, the value of δ was determined empirically to be 5x10-3, a real number. Other values of δ can be used, depending on the desired roll-off response, as described above.
    [082] Continuing with Figure 6, a unity impulse is provided at 618 for a Nyquist filter 520. As discussed above, the impulse response of a Nyquist filter, such as the Nyquist filter 620, is 0 for all nT , except for n = 0, as illustrated in Figure 6. In the mode of operation 600 illustrated in Figure 6, the output response of the Nyquist filter 620 extends over a time period of 5T.
    [083] The output of the Nyquist filter 620 is transformed into the frequency domain by an FFT 622. The output of the FFT 622 is a plurality of frequency compartments, which are provided as a dividend to a second input of the divider 616.
    [084] Divisor 616 divides the output of FFT 622 by the output of comparator 614. After divisor 616, a 0 can optionally be added to the output of divisor 616 at the Nyquist frequency in fs/2, to create an odd number of frequency positions, if desired. Having an odd number of frequency positions generates a resultant filter group delay, which is an integer number of samples. The optional insertion of 0 occurs between the divider 616 and an Inverse Fast Fourier Transform (IFFT) 626.
    [085] The output quotient of divider 616 represents a measure of the difference between the aggregation response and the impulse response of the Nyquist filter 620, where the aggregation response is a response of: (1) the Gaussian filter 604, (2) the frequency modulator 606, (3) the channel filter 608, and (4) the frequency demodulator 610. The aggregation response may vary depending on the inclusion or exclusion of frequency modulator 606, the channel filter 608, and frequency demodulator 610. As noted above, frequency demodulator 606, channel filter 608, as well as frequency demodulator 610 can be omitted from operation 600, particularly if channel filter 608 is of the type of linear phase with a BT product greater than or equal to 0.75 and if the modulation transfer functions of both the frequency modulator 606 and the frequency demodulator 610 are one unit.
    [086] By generating digital filter coefficients based on the output difference measure by the divider 616, ISI introduced by the Gaussian filter 604, the frequency modulator 606, the channel filter 608, as well as the frequency demodulator 610 can be substantially removed using a digital filter, including the generated digital filter coefficients.
    [087] The output of divider 616, whether attached with a 0 to the Nyquist frequency fs/2 or not, is provided to the IFFT 626. The IFFT 626 converts the frequency bins output from the divider 616 to an actual output signal from the time domain. The output of IFFT 626 provides digital filter coefficients as illustrated at 628. The real-time domain output of IFFT 626 block includes samples over a time period of 5T or a number of samples over time period 5T plus 1 samples if a 0 is appended to the output of splitter 616.
    [088] Using the digital filter coefficients determined by operation 600, a constant coefficient digital filter to remove ISI can be implemented that converts the response of a communications system that includes a Gaussian filter into one having a Nyquist response, plus the response. of a post-detection filter. By removing the ISI caused by Gaussian filters in a receiver, symbol decisions can be made with fewer errors, even at low SNR and when using high modulation orders.
    [089] According to operation 600, ISI, which is assigned to one or more of: (1) Gaussian filters (2), rear-end (3) transmitter modulators (4) receiver front-end, (5) channel filters, and (6) receiver demodulators can be accounted for and compensated. Operation 600 is not limited to accounting for and offsetting ISI assigned to the above transmitter and receiver components, however, and a person skilled in the art will recognize that ISI assigned to other transmitter and receiver components may be accounted for and offset.
    [090] With reference to the GFSK 300 communications system as an example, ISI assigned to the following components can be counted and compensated based on the digital filter coefficients determined by operation 600: (1) the Gaussian filter 306, (2) the FSK modulator 308, (3) the tail tip transmitter 310, (4) the receiver front end 318, (5) the channel filter 320, and (6) the FSK demodulator (322).
    [091] Figs. 7A, 7B, 7C and 7D illustrate the filtering effect of digital filter 324. In Figs. 7A, 7B, 7C and 7D, the vertical axis or y axis represents amplitude and the horizontal axis or x axis represents time.
    [092] Figure 7A illustrates a sequence of symbols demodulated from a received 2-GFSK signal. As illustrated in Figure 7A, not every symbol achieves full amplitude for 2-GFSK symbol segmentation. Instead, the "eye" of the symbol string is closed due to ISI.
    [093] Figure 7B illustrates the symbol sequence of Figure 7A after filtering by digital filter 324 in accordance with the digital filter coefficients determined by operation 600. As illustrated in Figure 7A, the symbols achieve full 2-GFSK symbol amplitude, and the "eye" of the sequence is not retracted. Thus, Figure 7B illustrates that ISI was removed by digital filter 324, compared to Figure 7A.
    [094] As illustrated in Figure 7C, the collapse of the "eye" of a demodulated symbol sequence of the received 8-GFSK signal is even more pronounced than in Figure 7A. As illustrated in Figure 7C not every symbol approaches its respective level, symbol level for 8-GFSK symbol targeting. Instead, the "eye" of the symbol string is closed due to ISI.
    [095] Figure 7D illustrates a sequence of symbols of Figure 7C after filtering by digital filter 324 in accordance with the digital filter coefficients determined by operation 600. As illustrated in Figure 7D, the symbols reach their respective 8-GFSK symbol levels , and the "eye" of the sequence is not retracted. Thus, Figure 7D illustrates that ISI was removed by digital filter 324, compared to Figure 7C.
    [096] As the "eye" of a demodulated symbol sequence becomes more withdrawn, determining the original logical levels of the symbols becomes more susceptible to symbol decision error. Particularly in higher orders of modulation, such as in Figure 7C, where distinguishing between more than two levels of symbols at a time is based on various thresholds, determining the original logical levels of the symbols is difficult and error-prone. As such, segmenters on conventional GFSK receivers are susceptible to making wrong symbol decisions, especially at low SNR. However, for the GFSK receptor 314 segmenter 326 distinguishing between symbol levels for the symbol sequence illustrated in Figure 7D is not as error prone.
    [097] The structure and operation of digital filter 324 will now be further described with reference to Figure 8.
    [098] Figure 8 illustrates a preferred embodiment of digital filter 324, a constant coefficient 800 finite impulse response (FIR) digital filter. Specifically, the digital FIR filter 800 of Figure 8 includes a string of 802 delay units. a chain of multiplier units 804, and a sum unit 806. The digital FIR filter 800 may also be incorporated by a digital FIR filter, which varies in structure from that illustrated in Figure 8 based on known structures of FIR digital filters.
    [099] The number of delay units in the delay unit chain 802 and the number of multiplier units in the multiplier unit chain 804 are determined based on at least one of, the received symbol symbol rate, the digital filter of sampling frequency fs, the BT product of the pulse shaping of the Gaussian filter 306, and the BT product of the channel filter 320. The sampling frequency of the digital filter, fs, must be greater than or equal to the Nyquist sampling frequency, which is twice the bandwidth of the signal at the output of the FSK 322 demodulator.
    [0100] In the operation of digital filter FIR 800, a symbol sequence, such as the symbol sequence retrieved by the FSK demodulator 322, is introduced into the chain of delay units 802. Each delay unit 802, and corresponding multiplier unit 804, comprises a weighting step that evaluates a symbol of the input symbol sequence by a respective filter coefficient. In Figure 8, [X] represents an input symbol sequence, such as the symbol sequence retrieved by the FSK demodulator 322. Each delay unit in the delay unit chain 802 stores a respective symbol of the input symbol sequence, per a symbol period, and outputs the stored symbol to the next delay unit 802. As illustrated in Figure 8, symbols are marked before, between and after each of the delay units 802 for multiplying the marked symbols by the respective A0-AN filter coefficients in the respective multipliers 804. The A0-AN coefficients are determined as described above in association with Figure 6. The output of each multiplier 804 is provided as input to the sum unit 806. sum 806 sums the outputs of multipliers 804 to produce one symbol of a sequence of output symbols, [Y], per symbol period T. Digital filter FIR 800 removes ISI between symbols in s input symbol sequence [X], and output symbol sequence [Y] is substantially ISI free.
    [0101] Based on the A0-AN filter coefficients, the 324 digital filter and the FIR 800 digital filter incorporate equalization filters that effectively convert the pulse-shaping impulse responses of Gaussian filters to a filter with an impulse response of Nyquist. As illustrated in Figure 6, the impulse response of a Nyquist filter is 0 for all nT except for n = 0 (where n represents an integer). Based on the Nyquist response, ISI can be substantially eliminated from a Nyquist filter-filtered symbol sequence. Digital filter 324 and digital filter FIR 800 are designed to remove residual ISI from a received signal to a part in 1000 (i.e., 0.1%) or less.
    [0102] Furthermore, as described above with respect to Figure 6, the digital filter 324 and the digital filter FIR 800 are further configured to perform post-detection filtering based on the A0-AN filter coefficients. The post-detection filtering performed by the digital filter 324 and the digital filter FIR 800 is similar to that performed by the post-detection filter 124.
    [0103] The FIR 800 digital filter can be fully implemented in hardware as a combination of hardware circuits. Alternatively, the FIR 800 digital filter can be implemented in software by a processor of a data processing apparatus executing computer readable instructions.
    [0104] Figure 9 is a flowchart illustrating a transmission method 900. The transmission method 900 includes pulsing a sequence of symbols at 902, modulation at 904, conversion at 906, and transmission at 908.
    [0105] To perform transmission method 900, a symbol sequence provided by a data source, such as data source 304, is pulse shaped at 902. Pulse shaping at 902 can be performed by a shaping of Gaussian filter pulse, such as Gaussian filter 306, and step 902 induces ISI in the symbol sequence. At 904, a frequency f0 is modulated in accordance with the sequence of symbols in the form of pulses to generate a modulated signal. Modulation at 904 can be implemented by the FSK 308 modulator. The conversion at 906 can be performed by a back end transmitter including a converter and amplifier, such as the rear end transmitter 310. Conversion at 906 converts the modulated signal to a frequency suitable for the RF transmission. RF transmission can be applied on 908 using a suitable transmit antenna.
    [0106] Figure 10 is a flowchart illustrating a receiving method 1000. The receiving method 1000 includes receiving a signal transmitted at 1002, converting the received signal to a baseband modulated signal at 1004, filtering the signal from baseband modulated in order to remove adjacent channel interference at 1006, demodulate the filtered channel modulated baseband signal to recover a symbol sequence at 1008, filtering the symbol sequence at 1010, producing symbol based decisions in the filtered symbol sequence at 1012, and mapping the symbol decisions to the data bits at 1014.
    [0107] To perform the reception method 1000, at 1002, the transmitted signal is received at the antenna, such as antenna 316. Conversion into 1004 is performed through, for example, the front end receiver 318, which converts the received signal for a modulated baseband signal. Removal of adjacent channel interference at 1006 is performed by a channel filter, which can induce ISI in addition to any ISI already present in the received signal. The removal of adjacent channel interference at 1006 can be accomplished through a channel filter, such as channel filter 320. Demodulation of the channel filtered modulated baseband signal at 1008 can be implemented by any suitable demodulator that provides an output proportional to the instantaneous frequency at its input, such as frequency demodulator 322. A retrieved symbol sequence, including ISI among the symbols of the sequence, is produced in 1008.
    [0108] At 1010, the ISI present in the recovered symbol sequence is substantially removed, providing an advantage over conventional techniques. Removal of the ISI at 1010 can be achieved by means of digital filters, such as the digital filter 324, which is described in greater detail above as the FIR filter 800. With the ISI removed by filtration at 1010, accurate symbol decisions can be made , substantially error free, even at low SNR, at 1012. For example, segmenter 326 can produce the symbol decisions at 1012 using a filtered symbol sequence that is substantially free of ISI. At 1014, data bits can be retrieved by mapping the symbol decisions produced at 1012 using symbol to bit mapper 328.
    [0109] Figure 11 is a flowchart illustrating a digital filter method 1100. According to the digital filter method 1100, a sequence of symbols is filtered. Specifically, a symbol set of the symbol sequence is determined at 1102, and each of the symbols of the given symbol set is multiplied by a respective filter coefficient at 1104. Each of the symbols of the given symbol set, multiplied by its respective one filter coefficient, is summed at 1106 to produce a first filtered symbol. The filtered symbol is output at 1108. Then, the determined set of symbols is updated by shifting in step 1110. For example, the determined set of symbols can be updated by a symbol based on a first-in-last symbol delay string -out (first in is last out), such as the 802 delay unit chain. After the update, the determined symbol set is multiplied by filter coefficients in 1104 and again summed in 1106. Thus, one second filtered symbol is output at 1108 based on updating the symbol set determined at 1110. As illustrated in Figure 11, digital filter method 1100 is iterative, and it produces a sequence of filtered output symbols. Multiplication at 1104 can be implemented by multiplier units 804, and addition at 1106 can be implemented by summation unit 806, for example. The respective coefficients of digital filter method 1000 can be the filter coefficients A0-AN.
    [0110] Although they have been described with reference to the GFSK 300 communication system, the 900, 1000, and 1100 methods can be performed using transmitter and receiver hardware circuits understood in the art to be equivalent to those described with respect to the communication system. GFSK 300 communication. In addition, methods 800, 900, and 1000 can be performed using hardware, software, or a combination of hardware and software. For example, the receiving method 1000 may be performed, in whole or in part, by a processor of a data processing apparatus processing in accordance with a set of computer-readable instructions, as described in further detail below, with reference to to Figure 12.
    [0111] According to embodiments implemented using a data processing apparatus executing computer readable instructions, the computer readable instructions are stored in a computer readable storage medium, which, when executed by a processor, configures and drives the processor and/or the processing apparatus for performing functions of the GFSK transmitter 300, the GFSK receiver 314, the operation 600, and the methods 900, 1000, and 1100. Non-limiting examples of the computer readable storage medium include random access memory (RAMs), read-only memories (ROM), optical disks (CD) (DVDs), and magnetic storage media.
    [0112] Figure 12 illustrates an embodiment of data processing apparatus 1200. Data processing apparatus 1200 includes a system bus 1202, a processor 1204, a RAM 1206, a ROM 1208, and an input/output interface 1210. In some embodiments, processor 1204 includes an integrated ADC 1212 and/or an integrated DAC 1214. Alternatively, the ADC 1212 and DAC 1214 may be detached from the processor 1204 and connected to the processor 1204 via the 1202 data bus. and/or the 1210 input/output interface.
    [0113] In operation, computer readable program instructions are loaded from at least one of RAM 1206, ROM 1208, and other storage media (not shown) to processor 1204 for execution. When executed by the 1204 processor, the computer-readable program instructions configure and direct the 1204 processor to perform the characteristics of the GFSK 300 transmitter, GFSK 314 receiver, 600 operation, and 900, 1000, and 1100 methods. Furthermore, to facilitate the implementation of reception method 1000 by the data processing apparatus, a modulated received signal can be converted to a modulated digital signal received using ADC 1212, so that processor 1204 is able to operate on top of a copy. digital of the received modulated signal. Furthermore, along with the processing of transmission method 900 by the data processing apparatus, a modulated digital signal can be converted by DAC 1214 to a modulated analog signal, for transmission as a modulated analog signal.
    [0114] The 1204 processor may include a general purpose Central Processing Unit (CPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), or an Application Specific Integrated Circuit - Application-specific integrated circuit (ASIC).
    [0115] Variations of the communications receiver and the communications receiver method are possible in light of the above description. Thus, the receiver of communications and the method of receiving communications may be practiced other than as specifically described above, based on art-recognized equivalents understood by those skilled in the art.
    权利要求:
    Claims (18)
    [0001]
    1. A Gaussian Frequency Switching communications receiver, GFSK, comprising: a receiver front end (318) for receiving a modulated signal and converting the modulated signal to a modulated baseband signal; a channel filter (320) for reducing channel interference which is adjacent to a desired channel of the modulated baseband signal from the modulated baseband signal and for producing a filtered channel modulated baseband signal. ; a demodulator (322) for demodulating the filtered channel baseband modulated signal and for recovering a sequence of symbols; a digital filter (324) having at least one filter coefficient derived from a Gaussian filter impulse response to reduce inter-symbol interference (ISI) of the symbol sequence; a segmenter (326) for producing symbol decisions based on the filtered symbol sequence, and a symbol-to-bit mapper for mapping symbol decisions to data bits, characterized in that the digital filter (324) comprises a plurality of weighting stages, each weighting stage for weighting a symbol of the symbol sequence by a respective filter coefficient based on a determined impulse response of a Nyquist filter divided by an aggregation response of a determined impulse response of a Gaussian filter (306) and a determined impulse response of the channel filter, and wherein a comparator (614) compares each frequency component of the aggregated response to a predetermined real number, and where a frequency component of the aggregated response is less than that the predetermined real number, replaces the frequency component with the predetermined real number to produce a modified aggregation response, and a divi. sor (616) divides the determined impulse response of the Nyquist filter (620) by the aggregate modified response to determine respective filtering coefficients.
    [0002]
    2. Communications receiver according to claim 1, characterized in that the digital filter (324) comprises a Nyquist equalization filter (620) which reduces ISI from the symbol sequence based on a Nyquist response.
    [0003]
    3. Communications receiver according to claim 1, characterized in that the digital filter (324) comprises a plurality of weighting stages, each weighting stage for weighting a symbol of the symbol sequence by a respective filter coefficient based on an impulse response from a Gaussian filter (306).
    [0004]
    4. A communications receiver according to claim 1, characterized in that the digital filter (324) comprises a plurality of weighting stages, each weighting stage for weighting a symbol of the symbol sequence by a respective filter coefficient based on an aggregate of a determined impulse response from the communications receiver and a determined impulse response from a transmitter (202) of the modulated signal.
    [0005]
    5. Communications receiver according to claim 1, characterized in that the digital filter (324) comprises a plurality of weighting stages, each weighting stage for weighting a symbol of the symbol sequence by a respective filter coefficient based on an aggregate of a determined impulse response from a Gaussian filter (306) and a determined impulse response from the channel filter (320).
    [0006]
    6. A communications receiver according to claim 5, characterized in that: the Gaussian filter (306) has a half-bandwidth to symbol rate (BT) ratio of 0.36 or less, and the filter channel (320) has a half bandwidth to symbol rate (BT) ratio of less than 0.75.
    [0007]
    7. GSFK communications receiver method characterized in that it comprises the steps of: receiving a modulated signal and converting the modulated signal to a modulated baseband signal; filter channel interference, which is adjacent to a desired channel for the baseband modulated signal so as to reduce channel interference from the baseband modulated signal and to produce a baseband modulated channel filtered signal ; demodulating the filtered channel baseband modulated signal to recover a sequence of symbols; filtering, by a processor of a data processing apparatus, the symbol sequence to reduce the inter-symbol interference (ISI) of the symbol sequence; producing symbol decisions based on the filtered symbol sequence, and mapping the symbol decisions to data bits, wherein filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Gaussian filter (306); determining an impulse response from filtering channel interference; aggregating the impulse response from the Gaussian filter (306) and the impulse response from the filter channel interference to produce an aggregation response; transforming the aggregated response into the frequency domain to produce a plurality of frequency bins; replacing the compartments of the plurality of frequency compartments having a magnitude less than a predetermined value with the predetermined value, to produce an aggregated response of the modified frequency domain; determining an impulse response from a Nyquist filter; transforming the Nyquist filter impulse response into the frequency domain to produce a Nyquist frequency domain response; dividing the Nyquist frequency domain response by the aggregated modified frequency domain response to produce a quotient; determining a plurality of filter coefficients based on the quotient, and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0008]
    8. Receiver communications method, according to claim 7, characterized in that filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Nyquist filter; determining a plurality of filter coefficients based on the Nyquist filter impulse response, and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0009]
    9. Receiver communications method according to claim 8, characterized in that filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Gaussian filter (306); determining a plurality of filter coefficients based on the impulse response of the Gaussian filter (306); and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0010]
    10. Receiver communications method according to claim 8, characterized in that filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Gaussian filter (306); determining an impulse response from filtering channel interference; aggregating the impulse response from the Gaussian filter (306) and the impulse response from the filtering channel interference to produce an aggregated response; determining a plurality of filter coefficients based on the aggregated response, and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0011]
    11. Receiver communications method according to claim 8, characterized in that filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Gaussian filter (306); determining an impulse response from filtering channel interference; aggregating the impulse response from the Gaussian filter (306) and the impulse response from the filtering channel interference to produce an aggregated response; determining an impulse response from a Nyquist filter; divide the Nyquist filter impulse response by the aggregate response to produce a quotient; determining a plurality of filter coefficients based on the quotient; and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0012]
    12. Receiver communications method according to claim 11, characterized in that the Gaussian filter (306) has a half-bandwidth to symbol rate (BT) ratio of 0.36 or less, and the interference Filtering channel has a half bandwidth to symbol rate (BT) ratio of less than 0.75.
    [0013]
    13. Receiver communications method according to claim 12, characterized in that the aggregation of the impulse response of the Gaussian filter (306) and the impulse response of the filtering channel interference comprises: determining an impulse response of a frequency modulator; determining an impulse response from the conversion of the modulated signal to a baseband signal, and aggregating the impulse response of the Gaussian filter (306), the impulse response of the filtering channel interference, the impulse response of the frequency modulator , and the impulse response of converting the modulated signal to a baseband signal to produce the aggregated response.
    [0014]
    14. A computer-readable storage medium characterized in that it stores computer-readable instructions which, when executed by a processor of a GSFK communications receiver, cause the processor to perform: receive a modulated signal and convert the modulated signal to a modulated baseband signal; filtering channel interference, which is adjacent to a desired channel from the modulated baseband signal so as to reduce channel interference from the modulated baseband signal and to produce a filtered channel modulated baseband signal; demodulating the filtered channel modulated baseband signal to recover a symbol sequence; filter the symbol sequence to reduce inter-symbol interference (ISI) from the symbol sequence; produce symbol decisions based on the filtered symbol sequence, and map symbol decisions to data bits; wherein filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Gaussian filter (306); determining an impulse response from filtering channel interference; aggregating the impulse response from the Gaussian filter (306) and the impulse response from the filter channel interference to produce an aggregation response; transforming the aggregated response into the frequency domain to produce a plurality of frequency bins; replacing the compartments of the plurality of frequency compartments having a magnitude less than a predetermined value with the predetermined value, to produce an aggregated response of the modified frequency domain; determining an impulse response from a Nyquist filter; transforming the Nyquist filter impulse response into the frequency domain to produce a Nyquist frequency domain response; dividing the Nyquist frequency domain response by the aggregated modified frequency domain response to produce a quotient; determining a plurality of filter coefficients based on the quotient, and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0015]
    A computer-readable storage medium according to claim 14, characterized in that filtering the symbol sequence to reduce ISI comprises: determining an impulse response from a Nyquist filter; determining a plurality of filter coefficients based on the impulse response of the Nyquist filter, and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0016]
    16. A computer-readable storage medium according to claim 15, characterized in that filtering the symbol sequence comprises: determining an impulse response from a Gaussian filter (306); determining a plurality of filter coefficients based on the impulse response of the Gaussian filter (306), and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0017]
    17. A computer-readable storage medium according to claim 16, characterized in that filtering the symbol sequence comprises: determining an impulse response from a Gaussian filter (306); determining an impulse response from filtering channel interference; aggregating the impulse response from the Gaussian filter (306) and the impulse response from the filter channel interference to produce an aggregation response; determining a plurality of filter coefficients based on the aggregated response, and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    [0018]
    18. A computer-readable storage medium according to claim 15, characterized in that filtering the symbol sequence comprises: determining an impulse response from a Nyquist filter; determining an impulse response from a Gaussian filter (306); dividing the Nyquist filter impulse response by the Gaussian filter impulse response (306) to produce a quotient; determining a plurality of filter coefficients based on the quotient; and filtering the symbol sequence to reduce ISI based on the plurality of filter coefficients.
    类似技术:
    公开号 | 公开日 | 专利标题
    BR112013002315B1|2021-07-20|GAUSSIAN FREQUENCY SWITCHING COMMUNICATIONS RECEIVER |, GSFK COMMUNICATIONS RECEIVER METHOD AND COMPUTER-READABLE STORAGE MEDIA
    US9270512B2|2016-02-23|Nonlinearity compensation for reception of OFDM signals
    JP5605897B2|2014-10-15|Receiving apparatus and receiving method
    CN104769875A|2015-07-08|Highly-spectrally-efficient transmission using orthogonal frequency division multiplexing
    EP3306884B1|2019-06-05|Frequency offset estimation for probabilistically shaped qam signal
    CN102946372B|2016-06-15|Ofdm system frequency deviation and time inclined combined estimation method
    SE518137C2|2002-09-03|Pulse formation and equalization in multiple systems with orthogonal frequency division
    CN105874734A|2016-08-17|Transmission device, reception device, and communication system
    JP6214816B2|2017-10-18|Receiver
    KR101995804B1|2019-07-03|Apparatus and method for supporting time-quadrature amplitude modulation in wireless communication system
    US20040042543A1|2004-03-04|Combined equalization for DMT-based modem receiver
    JPH11251904A|1999-09-17|Quantization device and quantization method
    US7711031B2|2010-05-04|Adaptive technique for inter-carrier-interference canceling in OFDM systems
    US8654887B2|2014-02-18|Methods and systems for peak-to-average power reduction without reducing data rate
    CN109167744B|2021-05-14|Phase noise joint estimation method
    CN108781129A|2018-11-09|Log-likelihood calculations circuit, reception device and log-likelihood calculations method
    US10158504B2|2018-12-18|Communication transmission with super-gaussian filtering in receiver
    KR101871362B1|2018-06-27|Apparatus and method for mapping quadrature amplitude modulation symbol
    WO2012051848A1|2012-04-26|Signal detection method and equipment in multi-path channel
    EP2415225A1|2012-02-08|Transmission method and device with symbol time-frequency mapping in sub-channels
    Bröhan2016|FPGA Based Development of an Orthogonal Frequency Division Multiplexing System for Automotive Communication Networks
    Choi et al.2016|Ber Simulation at the Waveform Level
    Clarkson et al.2002|A new joint coding and modulation scheme for channels with clipping
    KR20010064087A|2001-07-09|Quadrature Amplitude Modulation Demodulator for symbol timing recovery with low noise and high speed
    Wang et al.2005|System design and implementation of multiple-symbol encapsulated ofdm
    同族专利:
    公开号 | 公开日
    ES2708727T3|2019-04-10|
    IL224330A|2016-04-21|
    JP5833120B2|2015-12-16|
    BR112013002315A2|2016-05-24|
    CA2806960C|2017-02-28|
    JP2013534385A|2013-09-02|
    EP2599217A1|2013-06-05|
    MX2013001253A|2013-06-24|
    CN103155408B|2015-10-07|
    CN103155408A|2013-06-12|
    CA2806960A1|2012-02-02|
    EP2599217B1|2018-10-31|
    AU2011283184A1|2013-02-07|
    US8625722B2|2014-01-07|
    US20120027132A1|2012-02-02|
    EP2599217A4|2017-08-09|
    WO2012015509A1|2012-02-02|
    AU2011283184B2|2014-07-17|
    RU2013108855A|2014-09-10|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    JP2586169B2|1990-03-06|1997-02-26|日本電気株式会社|Demodulation system|
    US5377229A|1990-12-19|1994-12-27|Motorola, Inc.|Multi-modulation scheme compatible radio|
    JP3371256B2|1991-10-31|2003-01-27|日本電気エンジニアリング株式会社|Automatic equalizer|
    JPH06237277A|1993-02-09|1994-08-23|Nec Corp|Psk carrier signal regenerating device|
    US5361276A|1993-09-13|1994-11-01|At&T Bell Laboratories|All digital maximum likelihood based spread spectrum receiver|
    JPH098854A|1995-06-20|1997-01-10|Matsushita Electric Ind Co Ltd|Multilevel fsk receiver|
    JP3421541B2|1997-06-09|2003-06-30|株式会社鷹山|Signal receiving apparatus and signal receiving method|
    US6025758A|1997-10-30|2000-02-15|Uniden San Diego Research & Development Center, Inc.|Method and apparatus for performing digital data signal modulation|
    US5986691A|1997-12-15|1999-11-16|Rockwell Semiconductor Systems, Inc.|Cable modem optimized for high-speed data transmission from the home to the cable head|
    DE19824408A1|1998-05-30|1999-12-02|Philips Patentverwaltung|Receiver for a digital transmission system|
    US7415066B2|1998-08-10|2008-08-19|Kamilo Feher|Mis-matched modulation-demodulation format selectable filters|
    US6628728B1|1999-04-28|2003-09-30|Cyntrust Communications, Inc.|Nyquist filter and method|
    US7542520B1|1999-12-02|2009-06-02|Qualcomm Incorporated|Apparatus and method for implementing a low complexity digital modulator|
    US6674812B1|1999-12-20|2004-01-06|Agere Systems Inc.|High IF frequencies with a lower frequency logic based FSK modulation selecting a harmonic alias and demodulation using sub-sampling techniques|
    US6438176B1|2000-04-04|2002-08-20|Texas Instruments Incorporated|Digital gaussian frequency shift keying modulator|
    JP4486239B2|2000-09-18|2010-06-23|パナソニック株式会社|Receiving apparatus and timing estimation method|
    US6487240B1|2000-11-10|2002-11-26|Motorola, Inc.|Apparatus for receiving and recovering frequency shift keyed symbols|
    GB0100202D0|2001-01-04|2001-02-14|Koninkl Philips Electronics Nv|Receiver having a variable threshold slicer stage and a method of updating the threshold levels of the slicer stage|
    JP3494370B2|2001-03-23|2004-02-09|松下電器産業株式会社|Waveform equalizer, frequency offset compensation method, program, recording medium, mobile station radio apparatus, base station radio apparatus, and mobile communication system using waveform equalizer|
    TW550911B|2001-06-21|2003-09-01|Realtek Semiconductor Corp|FSK/GFSK signal receiver with the function of digitally compensating frequency offset and the receiving method of the same|
    US20030112896A1|2001-07-11|2003-06-19|Raghavan Sreen A.|Multi-channel communications transceiver|
    US8611408B2|2004-04-09|2013-12-17|Entropic Communications, Inc.|Apparatus for and method of developing equalized values from samples of a signal received from a channel|
    US7529322B2|2005-08-26|2009-05-05|University Of Macau|Two-step channel selection for wireless receiver front-ends|
    FI20085457A0|2008-05-15|2008-05-15|Nokia Corp|Device, method and computer program for demodulation|
    US8259862B2|2009-12-14|2012-09-04|Issc Technologies Corp.|Receivers and symbol decoders thereof|US8665941B1|2012-06-20|2014-03-04|MagnaCom Ltd.|Decision feedback equalizer for highly spectrally efficient communications|
    US8737458B2|2012-06-20|2014-05-27|MagnaCom Ltd.|Highly-spectrally-efficient reception using orthogonal frequency division multiplexing|
    CN103634814A|2012-09-10|2014-03-12|哈尔滨安天科技股份有限公司|Wireless probe circuit device, wireless probe equipment and wireless datagram processing method|
    US10020912B2|2013-03-13|2018-07-10|Sans R&D, Llc|Method and a system for a receiver design in bandwidth constrained communication systems|
    EP3020173B1|2013-07-08|2020-03-11|Hughes Network Systems, LLC|System and method for iterative compensation for linear and nonlinear interference in system employing ftn symbol transmission rates|
    WO2015054437A1|2013-10-08|2015-04-16|Hughes Network Systems|System and method for pre-distortion and iterative compensation for nonlinear distortion in system employing ftn symbol transmission rates|
    US9118519B2|2013-11-01|2015-08-25|MagnaCom Ltd.|Reception of inter-symbol-correlated signals using symbol-by-symbol soft-output demodulator|
    US9130637B2|2014-01-21|2015-09-08|MagnaCom Ltd.|Communication methods and systems for nonlinear multi-user environments|
    US9496900B2|2014-05-06|2016-11-15|MagnaCom Ltd.|Signal acquisition in a multimode environment|
    US8891701B1|2014-06-06|2014-11-18|MagnaCom Ltd.|Nonlinearity compensation for reception of OFDM signals|
    US9246717B2|2014-06-30|2016-01-26|Hughes Network Systems, Llc|Optimized receivers for faster than nyquisttransmission rates in high spectral efficiency satellite systems|
    US9246523B1|2014-08-27|2016-01-26|MagnaCom Ltd.|Transmitter signal shaping|
    US9276619B1|2014-12-08|2016-03-01|MagnaCom Ltd.|Dynamic configuration of modulation and demodulation|
    US9191247B1|2014-12-09|2015-11-17|MagnaCom Ltd.|High-performance sequence estimation system and method of operation|
    EP3272083A4|2015-05-27|2018-12-05|Hewlett-Packard Development Company, L.P.|Recovering independent waveforms from input bitstream data|
    CN105245201B|2015-10-10|2018-02-13|北京中科汉天下电子技术有限公司|A kind of method and Gauss forming filter for exporting Gauss shaping filter result|
    KR101824399B1|2015-12-30|2018-02-01|어보브반도체 주식회사|Bluetooth signal receiving method and device using improved packet detection and symbol timing acquisition|
    US10575267B2|2017-01-05|2020-02-25|Samsung Electronics Co., Ltd|System and method for providing weighted pattern demapper for Bluetooth® low energy long range|
    US11095960B2|2018-03-07|2021-08-17|Michael Edward Klicpera|Water meter and leak detection system having communication with a intelligent central hub listening and speaking apparatus, wireless thermostat and/or home automation system|
    法律状态:
    2018-12-26| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|
    2020-03-10| B15K| Others concerning applications: alteration of classification|Free format text: A CLASSIFICACAO ANTERIOR ERA: H03D 7/16 Ipc: H04L 27/14 (2006.01), H04L 25/03 (2006.01) |
    2020-03-10| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|
    2021-06-08| B09A| Decision: intention to grant [chapter 9.1 patent gazette]|
    2021-07-20| B16A| Patent or certificate of addition of invention granted|Free format text: PRAZO DE VALIDADE: 20 (VINTE) ANOS CONTADOS A PARTIR DE 03/05/2011, OBSERVADAS AS CONDICOES LEGAIS. PATENTE CONCEDIDA CONFORME ADI 5.529/DF, QUE DETERMINA A ALTERACAO DO PRAZO DE CONCESSAO. |
    优先权:
    申请号 | 申请日 | 专利标题
    US12/847,951|2010-07-30|
    US12/847,951|US8625722B2|2010-07-30|2010-07-30|GFSK receiver architecture and methodology|
    PCT/US2011/034974|WO2012015509A1|2010-07-30|2011-05-03|Gfsk receiver architecture and methodology|
    [返回顶部]