KR100510699B1 - Receiver OOB QPSK - Google Patents

Abstract

The present invention relates to a receiver, and more particularly to an OOB QPSK receiver for open cable systems. As described above, the OOB QPSK receiver for an open cable system according to the present invention includes an A / D converter and a phase separator for separating data output from the A / D converter into I and Q signals, and recovering the I and Q signals and carrier waves. A complex multiplier for multiplying the loop signal output to the baseband, a resampler unit for resampling the baseband signal, a matched filter unit for performing matching filtering behind the resampler to complete baseband phase shaping, and the completion A channel equalizer for splitting the baseband phase shaping at a symbol rate, multiplying a carrier frequency offset for carrier recovery behind the channel equalizer, and a carrier recoverer and a phase derotator for compensating for a misaligned phase ) And a timing recovery unit.

Description

OPS GP receiver Receiver {Receiver OOB QPSK}

The present invention relates to a receiver, and more particularly to an OOB QPSK receiver for open cable systems.

Recently, digitalization of broadcast media is rapidly progressing worldwide, and digital broadcasting of terrestrial broadcasting and satellite broadcasting has started in Korea. In Korea, where more than 60% of total TV receiving households are receiving broadcasts through wired networks, digitalization of cable broadcasting is a little late compared to satellites and terrestrial waves. It is becoming.

The Ministry of Information and Communication has been working on the digitalization of cable broadcasting since 1999, and in 2001, the US-based Open Cable standard was selected as the domestic digital cable broadcasting standard. In the US, cable broadcasting accounts for the largest portion of the overall broadcast market, but the equipment market, including associated headend equipment and set-top boxes, is almost exclusively owned by two companies (Motorola, Scientific Atlanta), It can be said to have. In order to eliminate this closedness, since 2005, the part related to CA is separated from set-top box.

The open cable standard consists of a subscriber station (Host) with a restricted reception area and a conditional access module (POD) that includes a conditional access part to meet FCC regulations. I am trying to.

Open cable standards require the following features:

First, they support FCC requirements, which include segregation of CAs, purchase of subscriber terminals in competitive markets, and consistency with existing digital cable broadcasting systems.

Second, performance parameters must be supported to allow acceptable network operation and audio and video quality.

Third, the conditional access function by the removable conditional access module should be supported.

Fourth, out-of-band signaling should be supported, and specific bands are used for EPG-related service information, entitlement management messages (EMM), and data services.

Fifth, it is to support the output function of HDTV compressed signal stream by IEEE 1394 connection.

Sixth, support copying to meet the Motion Picture Association of America (MPAA) requirements for advanced programs.

In addition, it must support download functions and application software.

In addition, there are three types of matching standards, which include cable network matching, subscriber terminal and conditional access module matching, and subscriber terminal and external device matching. Cable network matching is a signal transmitted and received from a subscriber terminal through a transmission line. It is prescribed.

The subscriber station and the conditional access module matching define a matching signal between the subscriber station and the conditional access module, and the subscriber station and external device matching define a matching signal between the subscriber station and an external device such as a TV and a VCR.

The cable network matching mostly accepted the "Digital Cable Network Interface Standard" (SCTE 40 2001), which is a SCTE standard. This matching defines the signal between the subscriber station and the CATV network by dividing it into physical layer characteristics, transport layer characteristics, services, and related protocol stacks.

The Open Cable standard allows the transmission of service / system information, entitlement management messages (EMM), data, etc., of scrambled digital channels through out-of-band channels, and SCTE 40 (2001) uses DVS167 and DVS178. It is specified to use one of two methods.

The out-of-band transmission must be bidirectional, and both downlink and uplink use quadrature phase shift keying (QPSK) modulation. The transmission speed supports 1.544 Mbps, 2.408 Mbps, 3.088 Mbps for downlink, 0.256 Mbps, 1.544 Mbps, 3.088 Mbps for uplink. The frequency band used is 70 ~ 130MHz for the downlink and 5 ~ 42MHz for the uplink. The RF channel bandwidth is 1.0 / 1.5 / 2.0 MHz for the downside and 0.192 / 1.0 / 2.0 MHz for the upside.

1 shows a configuration of an open cable system. As shown in FIG. 1, an MPEG-2 stream is sent to a signal in a band, and demodulation is performed at a receiving side (Open Cable STB, Cable Ready DTV) with 64/256 QAM. Other control signals can be transmitted and received in both directions outside the band. Therefore, a chip that can receive and transmit out-of-band QPSK is required for open cable system operation. In particular, the ability to accurately demodulate control signals is essential.

2 is a diagram illustrating an OOB QPSK receiver for a general open cable system.

Referring to FIG. 2, a signal lowered to 44 MHz is output from an analog front end before the A / D converter 1, and the A / D converter 1 receives the signal to sample data at a sampling rate of 14 MHz. The timing error loop is calculated from the sampled signal and the matched filter output to form a timing recovery loop to compensate for the final timing error in the resampler 2. The timing error compensated data is divided into I and Q signals. The Hilbert transform filter is used to generate the Q signal. The I and Q signals are multiplied by the short carrier loop signal output to lower the baseband to a 2MHz signal, a 44MHz residual image. Here, the output of the carrier recovery unit is an output of a number controlled oscillator (NCO), which is a sum of 2 MHz and a carrier frequency offset calculated by the carrier recovery unit.

Matching filtering is performed on these outputs to complete the baseband pulse shape and the carrier recovery loop is constructed using this data. In this case, a loop for tracking the carrier recovery offset and a loop for tracking the static phase offset are configured together. This is lowered to the symbol rate to perform channel equalization.

Through this process, the final symbol data is obtained, and from this data, the differential encoding is released from the transmitter, and the DRX signal is given to the Point Of Depoloyment (POD) at a clock twice as fast as the symbol. It is then given to the POD with CRX, which is a clock twice as fast as the previously mentioned symbol.

Such a passband architecture has a benefit in the compensation of the carrier recovery offset due to the characteristics of the carrier recovery loop having a short group delay, but requires a high order and degrades the performance of the resampler that compensates for the timing error.

Fully tracking the timing error and performing the remaining demodulation increases the stability of the overall system. Under this structure, vulnerabilities are revealed in this area. Thus, the system-wide SNR characteristics will be degraded.

Accordingly, an object of the present invention is to provide an OOB QPSK receiver that performs timing recovery in baseband in view of the above-mentioned problems of the prior art.

       According to an aspect of the present invention for achieving the above object, an A / D converter, a phase separation unit for separating the data output from the A / D converter into I, Q signals, the I, Q signal and the carrier wave A complex multiplier for multiplying the recovery loop signal output to the baseband, a resampler unit for resampling the baseband signal, a matched filter unit for performing matching filtering behind the resampler, and completing baseband phase shaping; A channel equalizer for dividing the completed baseband phase shaping at a symbol rate, a carrier frequency offset multiplied by a carrier frequency offset for carrier recovery behind the channel equalizer, and a carrier recovery unit and a phase derotator (compensating for a distorted phase) A derotator) and a timing recovery unit.

       Preferably, the resampler unit is a value obtained by adding a timing offset calculated as an output of the timing recovery unit to a ratio of a sampling rate and a fixed clock rate for each QPSK mode.

       The feed forward input of the channel equalizer is used as the output of the matched filter, and the feed back input uses the phase derotator output.

       In addition, an error for updating the coefficient of the channel equalizer is calculated by the derotator output.

       The feed forward input of the channel equalizer uses the output of the matched filter unit, and the feed back input uses data obtained by slicing the output of the channel equalizer, and the coefficient of the channel equalizer. The error to update uses the output of the channel equalizer.

        In addition, the derotator output is used as a feed forward input of the channel equalizer, and the output of the channel equalizer is sliced as a feed back input.

Hereinafter, a configuration and an operation according to an exemplary embodiment of the present invention will be described with reference to the accompanying drawings.

3 is a diagram illustrating a configuration of an OOB QPSK receiver according to an embodiment of the present invention. Phase separation separating an A / D converter 11 and data output from the A / D converter 11 into I and Q signals is shown in FIG. 12, a complex multiplier 13 multiplying the I and Q signals by the carrier recovery loop signal output to the baseband, a resampler 14 for resampling the baseband signal, and the resampler ( 14) a matching filter unit 15 for performing matching filtering to complete baseband phase shaping, a channel equalizer 16 for dividing the completed baseband phase shaping at a symbol rate, and the channel equalization. A carrier recovery unit for multiplying the carrier frequency offset for carrier recovery behind the unit 16 and compensating for a misaligned phase is composed of a phase derotator 17 and a timing recovery unit 19.

Referring to the operation of the OOB QPSK receiver according to the present invention as follows.

A signal lowered to 44 MHz is output from the analog front end before the A / D converter 11, and the data is received and sampled at a sampling rate of 14 MHz. The phase separator 12 divides the signal into I and Q signals. This I, Q signal is multiplied by the signal output of the carrier recovery unit 17 started behind the channel equalizer 16 to lower the 2 MHz signal to the baseband. The resampling 14 is performed on the baseband signal, which is also a timing offset calculated by the loop output of the timing recovery unit 19, and a sampling rate and peak seed for each QPSK mode (0.772Msps / 1.024Msps / 1.544Msps). The interpolation time of the finite impulse response filter is determined by adding the ratio of the fixed clock rate so that the output is clocked eight times faster than the actual symbol rate. Matched filtering is performed on this output value to complete baseband pulse shaping, and the resultant is lowered to a symbol rate to perform equalizing. Carrier recovery begins behind channel equalizer 16, multiplies carrier frequency offset behind phase splitter, and removes a short loop to eliminate residual carrier phase jitter. There is a phase derotator 17 that generates and compensates for the skewed phase behind the channel equalizer 16. This process completes the demodulation and provides the POD with the final data in I, Q order in the differential decoder. In this structure, the channel equalizer 16 is shown in detail. The channel equalizer 16 is a decision feedback equalizer (DFE), and the feed forward input is used as the output of the matched filter 12 and the feed back is applied. The input uses a phase derotator 17 output. An error for updating the coefficient of the channel equalizer 16 is calculated by the output of the derotator 17. Under this structure, there is a delay to the output of the derotator 17, which results in several pipelines for updating the coefficients, but this delay is equal to channel equalization. Without significantly affecting the performance of the equalizer, the OOB QPSK system has a very low symbol rate, resulting in minimal delay and optimized performance.

4 is a second embodiment of the OOB QPSK receiver according to the present invention. The basic structure is similar to that of FIG. 3, but there is a difference in the channel equalizer 27. The feed forward input of the channel equalizer 26 uses the output of the matched filter unit 22 as in the first embodiment of FIG. 3, but the feed back input outputs the channel equalizer 26. Use sliced data. The error of updating the coefficients of the channel equalizer 26 is obtained by using the output of the channel equalizer 26 instead of the output of the derotator 27. In this case, the pipeline delay for updating the coefficients is reduced, but the error resulting from the difference between the data that is not compensated for the static phase and the constellation point starts with a default value. The channel equalizer 27 serves as a burden on the system for a channel that does not have Inter Symbol Interference (ISI), thereby lowering the convergence speed of the system.

   FIG. 5 shows a third embodiment of an OOB QPSK receiver in accordance with the present invention using a derotator 36 output as a feed forward input and a channel equalizer 37 as a feed back input. Slice the output of and use it. The error of the channel equalizer 36 is obtained from the output of the channel equalizer 36. It tracks carrier frequency offset and static phase offset by performing carrier recovery with undivided data, which is a burden on carrier recovery loop 36 for channels with ISI. It is a factor that degrades performance. Instead, the group delay of the carrier recovery loop 36 is reduced.

6 is a graph showing the frequency response of the resampler in the pass band, Figure 7 is a graph showing the frequency response of the resampler in the baseband.

Comparing the frequency response characteristics of the resampler in FIGS. 6 to 7, it can be seen that the pass band satisfies a certain performance only when the third order is 8 taps. This shows sufficient performance. In addition, when the second order is used in the actual resampler implementation, there is room for further reduction of hardware since sharing of coefficients is possible.

As described above, the present invention has an advantage in that the burden of hardware is relatively reduced along with the performance improvement of the resampler while changing from the existing passband structure to the baseband structure.

Those skilled in the art will appreciate that various changes and modifications can be made without departing from the spirit of the present invention.

Therefore, the technical scope of the present invention should not be limited to the contents described in the examples, but should be defined by the claims.

1 is a block diagram showing the configuration of an open cable system

2 is a block diagram showing the configuration of a conventional OOB QPSK receiver

3 is a first embodiment showing the configuration of an OOB QPSK receiver according to the present invention;

4 is a second embodiment showing the configuration of an OOB QPSK receiver according to the present invention;

5 is a third embodiment showing the configuration of an OOB QPSK receiver according to the present invention;

6 is a graph showing the frequency response characteristics of the resampler in the pass band

7 is a graph showing the frequency response characteristics of the resampler in the baseband

Claims (7)

        An A / D converter;         A phase separator for separating data output from the A / D converter into I and Q signals;         A complex multiplier multiplying the I and Q signals by a carrier recovery loop signal output to a base band;         A resampler unit for resampling the baseband signal;         A matching filter unit performing matching filtering behind the resampler to complete baseband phase shaping;         A channel equalizer for dividing the completed base band phase shaping at a symbol rate;         A carrier recovery unit and a phase derotator for multiplying a carrier frequency offset for carrier recovery behind the channel equalizer and compensating for a misaligned phase;         OOB QPSK receiver characterized in that it comprises a timing recovery unit.         The method of claim 1,         And the resampler unit adds a timing offset calculated as an output of the timing recovery unit, and a ratio of a sampling rate and a fixed clock rate for each QPSK mode.         The method of claim 1,         OOB QPSK receiver, characterized in that the feed forward input of the channel equalizer is used as the output of the matched filter, and the feed back input is the phase derotator output.         The method of claim 3, wherein         The error for updating the coefficients of the channel equalizer is calculated by the derotator output, characterized in that the OOB QPSK receiver.         The method of claim 1,         The feed forward input of the channel equalizer uses the output of the matched filter unit, and the feed back input uses data slicing the channel equalizer output. The method of claim 5, The error updating the coefficients of the channel equalizer uses an output of the channel equalizer.          The method of claim 1,          An OOB QPSK receiver, characterized in that the derotator output is used as a feed forward input of the channel equalizer and the output of the channel equalizer is sliced as a feed back input