CN108923887B - Soft decision decoder structure of multi-system partial response CPM signal - Google Patents
Soft decision decoder structure of multi-system partial response CPM signal Download PDFInfo
- Publication number
- CN108923887B CN108923887B CN201810671401.8A CN201810671401A CN108923887B CN 108923887 B CN108923887 B CN 108923887B CN 201810671401 A CN201810671401 A CN 201810671401A CN 108923887 B CN108923887 B CN 108923887B
- Authority
- CN
- China
- Prior art keywords
- state
- module
- reliability matrix
- path
- serial number
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0045—Arrangements at the receiver end
- H04L1/0054—Maximum-likelihood or sequential decoding, e.g. Viterbi, Fano, ZJ algorithms
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0045—Arrangements at the receiver end
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0061—Error detection codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
- H04L1/0056—Systems characterized by the type of code used
- H04L1/0064—Concatenated codes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/18—Phase-modulated carrier systems, i.e. using phase-shift keying
- H04L27/22—Demodulator circuits; Receiver circuits
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Error Detection And Correction (AREA)
Abstract
The invention provides a soft-decision decoder for a multi-system partial response CPM signal. The method comprises the following steps: the device comprises an adding and comparing module, a reliability matrix updating unit, a state selecting module, a first delay module, a second delay module and a soft decision output module. The branch measurement and modulation index synchronous signals of all paths are input into an addition and comparison module, the addition and comparison module calculates the survivor path serial number and the competition path serial number of each state and the measurement difference (path measurement difference for short) of the survivor path and the competition path and outputs the measurement difference to a reliability matrix updating unit, and the accumulated measurement of each state is calculated and output to a state selection module. The state picking module finds the state with the largest cumulative metric among all the states. The soft decision output module calculates the likelihood ratio (multi-system likelihood information) of all the symbols and outputs the likelihood ratio. The decoder can be suitable for soft decision decoding of a multilevel CPM signal; is suitable for high-speed decoding.
Description
Technical Field
The invention relates to the technical field of wireless communication and remote measurement and control, and provides a soft-decision decoder structure for multilevel partial response CPM (Continuous Phase Modulation) signals.
Background
The CPM signal has the characteristic of continuous phase, has great advantages in power efficiency and bandwidth efficiency compared with other modulation modes, and has wide application prospects in the fields of satellite communication and remote measurement and control. The CPM signal can be divided into a full-response CPM signal and a partial-response CPM signal according to whether a partial response length L of the CPM signal is greater than 1, the partial-response CPM signal having a memory, and a waveform in a current symbol period is related not only to a symbol in the current symbol period but also to symbols in the previous symbol periods. Partial response CPM can achieve better bandwidth utilization. The CPM signal is processed into a hard decision value (i.e., 0 or 1) in a general demodulation process, during which a large amount of intermediate information is lost. Better error resilience can be achieved if the likelihood information of each decision value generated in the demodulation process assists channel decoding. SOVA (Soft-output Viterbi) is a Soft-Decision Algorithm that Outputs likelihood information based on an output Viterbi hard-Decision value (Hagenauer J, Hoeher P.A Viterbi Algorithm with Soft-Decision Outputs and bits Applications [ C ]// Global electronic communications Conference,1989, and inhibition. communications Technology for the 1990s and beyond. GLOBECOM'89.IEEE. IEEE Xplore,1989: 1680-. At present, most of soft decision decoders based on SOVA algorithm are inherited and developed by Two-Step SOVA structure (Joeressen O J, Meyr H.A 40Mbit/s soft-output Viterbi decoder [ J ]. IEEE Journal of Solid-State Circuits,1995,30(7): 812-. As shown in fig. 1, the structure comprises an add-compare-select module, a survivor path storage module, a metric difference quantization, normalization and delay module, a path comparison module, a likelihood value update module, a delay module and a selection module. The branch metrics of all paths are input into a comparison and addition module, the comparison and addition module compares the metric values of the candidate paths of all the states and selects a survival path and a competition path of all the states, the accumulated metrics of all the states are updated, the serial numbers of the survival path and the competition path of each state are sent to a survival path storage module, the serial numbers of the survival path and the competition path of each state after delay are sent to a path comparison module, and the metric difference of the survival path and the competition path of each state is sent to a metric difference quantization, normalization and delay module; and the survivor path storage module backtracks according to the sent state transition flag vector, sends the symbols corresponding to the survivor path and the competition path obtained by backtracking to the path comparison module, and sends the serial number of the state view passed by the survivor path to the metric difference quantization, normalization and delay module. The metric difference quantization, normalization and delay module selects the metric difference corresponding to the state passed by the survivor path according to the state serial number passed by the survivor path, and sends the metric difference to the likelihood value updating module; the path comparison module outputs a hard decision value on one hand, finds out the positions of unequal corresponding symbols on the survivor path and the competition path on the other hand, and sends the positions to the likelihood value updating module; and the likelihood value updating module selects the measurement difference of the positions with unequal corresponding symbols on the survivor path and the competition path according to the measurement difference of the positions with unequal corresponding symbols on the survivor path and the competition path and the measurement difference corresponding to the state of the survivor path, and takes the minimum value as likelihood information to be output. The hard decision and likelihood information are combined to be a soft decision. The algorithm based on the structure is not an optimal SOVA algorithm, the performance cannot be close to that of a Max-Log-MAP algorithm (an approximate algorithm of a maximum posterior probability algorithm in a logarithmic domain) under an ideal condition, and the performance is worse when the structure is improved and under a multi-system application scene. The invention provides a soft-decision decoder structure of a multilevel partial response CPM signal, which is based on a soft-output Viterbi algorithm (Leibohui, Malayan, Wangshui, Thankui, Gaoqian) of a multilevel partial response Multi-h CPM, and the performance of the soft-output Viterbi algorithm [ J ] of the multilevel partial response Multi-h CPM, the report of terahertz science and electronic information 20148.16(2): 277-.
Disclosure of Invention
The invention aims to provide a soft-decision decoder structure of a multilevel partial response CPM signal, which can be suitable for soft-decision decoding of the multilevel CPM signal; the performance of the realized algorithm is similar to the Max-Log-MAP algorithm under the ideal condition; each symbol requires less processing clock and is suitable for high-speed decoding.
As shown in fig. 2, a soft decision decoder for a multilevel partial response CPM signal is specifically characterized by comprising: the device comprises an adding and comparing module, a reliability matrix updating unit, a state selecting module, a first delay module, a second delay module and a soft decision output module. The branch metrics and modulation index synchronization signals of all paths (which are not needed in a single-index CPM application scene) are input into an addition-comparison selection module, the addition-comparison selection module executes addition-comparison selection operation of the conventional SOVA decoder, calculates the survivor path serial number and the competition path serial number of each state, and the metric difference (path metric difference for short) between the survivor path and the competition path, outputs the metric difference to a reliability matrix updating unit, calculates the cumulative metric of each state, and outputs the cumulative metric to a state selection module. The modulation index synchronous signals are simultaneously input to the first delay module, and the first delay module outputs the modulation index synchronous signals after delay to the reliability matrix updating unit. The reliability matrix updating unit is used for updating the reliability matrix of each state and outputting the first column of the reliability matrix of each state to the soft decision output module. The state selecting module finds out the state with the maximum accumulated metric in all the states and outputs the serial number of the state to the second delay module. The second delay module delays the serial number of the state with the maximum accumulative metric and outputs the delayed serial number of the state with the maximum accumulative metric to the soft decision output module. The soft decision output module calculates the likelihood ratio (multi-system likelihood information) of all the symbols and outputs the likelihood ratio.
Further, the reliability matrix updating unit includes: the device comprises a reliability matrix mapping module and a plurality of reliability matrix updating modules, wherein the number of the reliability matrix updating modules corresponds to the number of states of CPM signals. At the current moment, the input of the reliability matrix mapping module is the survivor path serial number and the competition path serial number of each state at the current moment; the reliability matrix mapping module backtracks the survivor path and the competition path of each state according to the survivor path sequence number and the competition path sequence number of each state, finds the state passed by the survivor path of each state at the previous moment and the state passed by the competition path at the previous moment, outputs the reliability matrix of the previous moment corresponding to the state passed by the survivor path at the previous moment of each state to the reliability matrix updating module corresponding to the state, and outputs the reliability matrix of the previous moment corresponding to the state passed by the competition path at the previous moment to the reliability matrix updating module corresponding to the state. Meanwhile, each reliability matrix updating module inputs the path metric difference of the corresponding state of the module obtained in the comparison and selection module. And the reliability matrix updating module corresponding to each state calculates and updates the reliability matrix corresponding to the state at the current moment, and outputs the first row of the reliability matrix to the soft decision output module.
The invention has the beneficial effects that: a brand-new SOVA decoder structure is provided, a reliability matrix representing the reliability of the multilevel system is continuously updated under the structure, and the likelihood information of the multilevel system can be output; the performance of the realized algorithm can approach the Max-Log-MAP algorithm under the ideal condition; the reliability matrix updating realized by the reliability matrix updating unit can be completed in 1 clock, and is suitable for high-speed realization.
Drawings
FIG. 1 is a schematic diagram of a prior art SOVA decoder;
fig. 2 is a schematic diagram of a soft-decision decoder for a multilevel partial response CPM signal according to the present invention;
FIG. 3 is a state transition trellis for a multilevel partial response CPM signal;
fig. 4 is a schematic diagram of a reliability matrix updating unit in a soft-decision decoder for a multilevel partial response CPM signal according to the present invention;
FIG. 5 is a block diagram of a concatenated LDPC multi-modulation index CPM system to which the present invention is applied;
FIG. 6 is a diagram of error performance of a concatenated LDPC multi-modulation index CPM system using the present invention;
table 1 shows the resource occupation achieved by the present invention on an FPGA;
Detailed Description
In order to make the present invention easier to understand, the following describes the present invention in further detail by taking a soft decision decoder with a state number of 8 and a system number of M-4 and a partial response CPM as an example implemented on an FPGA (Field Programmable Gate Array) in conjunction with the attached drawings.
State transition network of partial response CPM with state number of 8 and carry number M of 4The grid is shown in figure 3. Wherein S0～S7The CPM signal corresponds to 8 states. The lines represent state transition paths, where the solid lines represent transition paths of symbol 0, the denser dashed lines represent transition paths of symbol 1, the sparser dashed lines represent transition paths of symbol 2, and the dotted lines represent transition paths of symbol 3. (CPM state transition trellis is well known in the CPM art, see document Anderson J B, Aulin T, Sundberg C E.digital Phase Modulation [ J ]].Applications of Communications Theory,1986:412-412.)
The comparing and selecting module of the invention has similar structure and function with the comparing and selecting module of the existing SOVA decoder. The difference is that each state of the 4-ary CPM signal has 4 candidate paths (refer to fig. 3), whereas each state of the binary CPM signal has only two candidate paths. Therefore, the difference between the present invention and the conventional SOVA decoder is that the conventional SOVA decoder only needs to select the path with the larger cumulative metric from the two candidate paths as the survivor path and the other path as the competing path, whereas the present invention needs to select the path with the largest cumulative metric from the M candidate paths, regard it as the survivor path, select the path with the second largest cumulative metric, and regard it as the competing path.
The modulation index synchronous signals are simultaneously input into the first delay module. The first delay module delays the modulation index synchronous signal, so that the output delayed modulation index synchronous signal is aligned with the output of the addition, comparison and selection module in time, and the output is output to the reliability matrix updating module. The processing time of the add-compare-select module is usually 3 or 4 clocks, so the delay time of the first delay module is 3 or 4 clocks.
As shown in fig. 4, the reliability matrix updating unit includes: the reliability matrix updating module comprises a reliability matrix mapping module and a plurality of reliability matrix updating modules. At the current moment, the input of the reliability matrix mapping module is the survivor path serial number and the competition path serial number of each state at the current moment; the reliability matrix mapping module backtracks the survivor path and the competition path of each state according to the survivor path serial number and the competition path serial number of each state, and finds the state passed by the survivor path of each state at the previous moment and the state passed by the competition path at the previous moment; and then outputting the reliability matrix of the previous moment corresponding to the state passed by the survivor path of each state at the previous moment to the reliability matrix updating module corresponding to the state, and outputting the reliability matrix of the previous moment corresponding to the state passed by the competing path at the previous moment to the reliability matrix updating module corresponding to the state. Meanwhile, each reliability matrix updating module inputs the path metric difference of the corresponding state of the module obtained in the comparison and selection module. And the reliability matrix updating module corresponding to each state calculates and updates the reliability matrix corresponding to the state at the current moment, and outputs the first row of the reliability matrix to the soft decision output module.
The corresponding reliability matrix in each state is a 4 row delta column matrix. The value of delta is larger, the precision of the reliability matrix is higher, and the value of delta is smaller than 30 generally. To facilitate the description of the update process of the elements in the reliability matrix, the state at time k S is usedi(0. ltoreq. i.ltoreq.7, i is an integer) is expressed as
State S at time k +1iThe reliability matrix of the survivor path at the corresponding k time in the state that the k time passes is expressed as
State S at time k +1iThe reliability matrix of the corresponding k time in the state that the k time passes by the competition path is expressed as
Wherein j and mu are integers, j is more than or equal to 0 and less than or equal to delta-1, and mu is more than or equal to 0 and less than or equal to 3.
The element initialization rule in the reliability matrix is as follows:
at time 0, state SiLast column of the corresponding reliability matrix in
The middle element is assigned according to the following formula:
wherein v is the entry state SiThe symbol corresponding to the candidate path of (4).
State SiThe other elements in (i.e., the other elements in the reliability matrix except the last column of elements) are initialized to + ∞ (plus infinity), and in practice + ∞canbe replaced by twice the maximum value of the path metric difference, e.g., when the path metric difference is 5 bits quantized, the maximum value of the path metric difference is 31, and + ∞isreplaced by 63.
The rule for updating the reliability matrix in the reliability matrix updating module is as follows:
at the moment k +1, the last column in the reliability matrix is still valued according to the initialized rule, and the other element updating rules are as follows:
wherein l is more than or equal to 0 and less than or equal to delta-2, min represents a small operation, diffk+1(Si) Represents the state S at the time k +1iThe path metric difference of (2).
At time k, after the reliability matrix is updated, state SiThe corresponding reliability matrix update module updates the first column element of the reliability matrix corresponding to the state, i.e. the element
And outputting the signal to a soft decision output module.
The state selecting module is used for finding out the state with the maximum accumulated metric in all the states and outputting the serial number of the state to the second delay module.
The second delay module delays the serial number of the state with the maximum accumulated metric, so that the serial number is aligned with the first column of the corresponding reliability matrix in each state output by the reliability matrix updating module in time, and the serial number is output to the soft decision output module.
And c is the serial number of the state with the maximum accumulative measurement of the input soft decision output module. The soft decision output module picks out the state ScAnd calculating and outputting likelihood ratios of all symbols in a first column (representing the probability of each symbol appearing at the moment of k-delta +1, wherein the probability is higher when the value is smaller). Likelihood ratio of symbol mu at time k-delta +1
The calculation method comprises the following steps:
wherein μ ═ 0,1,2, 3.
The likelihood ratios of the symbols may then be translated into likelihood ratios of the bits according to actual needs.
The invention is applied to a multi-modulation index CPM system of cascade LDPC (Low Density Parity Check Codes), the system structure is shown in FIG. 5, a sending end carries out LDPC coding on an information sequence to be sent, then carries out CPM modulation, and finally sends out an obtained intermediate frequency CPM signal through a channel. After digital down-conversion and synchronization, the receiving end inputs the decoder provided by the invention to perform soft decision decoding, and the output of the decoder is converted into a bit likelihood ratio to perform LDPC decoding, so that a transmitted information sequence can be obtained. CPM modulation index h ═ 5,6]The code length of the LDPC code is 8176, the code rate is 7/8, and δ is 20. The error code performance is shown in FIG. 6, the curve with five stars represents the performance of the decoder (FPGA implementation) provided by the invention, the curve with diamonds represents the performance of the Max-Log-MAP algorithm (MTLAB simulation), and the graph shows that the error code rate is 10-5In this case, the Eb/No (energy per bit to noise power spectral density) differs by only 0.2 dB. The method is realized on XC7VX690T (FPGA development board), the transmission bit rate is 60M bit/s, the processing clock is 120MHz, and the highest clock is120MHz, the main resource occupation situation is shown in Table 1, and it can be seen from the table that the soft decision decoder provided by the invention occupies less resources and is suitable for high-speed transmission.
TABLE 1
Claims (1)
1. A soft-decision decoder for a multilevel partial response CPM signal, CPM being continuous phase modulation, comprising: the system comprises an adding and comparing module, a reliability matrix updating unit, a state selecting module, a first delay module, a second delay module and a soft decision output module; the branch measurement and modulation index synchronous signals of all paths are input into an adding and comparing module, the adding and comparing module calculates the survivor path serial number and the competition path serial number of each state and the measurement difference of the survivor path and the competition path and outputs the measurement difference to a reliability matrix updating unit; the adding and comparing module calculates the accumulated measurement of each state and outputs the accumulated measurement to the state selecting module; the modulation index synchronous signals are simultaneously input to a first delay module, and the first delay module outputs the modulation index synchronous signals after delay to a reliability matrix updating unit; the reliability matrix updating unit is used for updating the reliability matrix of each state and outputting the first column of the reliability matrix of each state to the soft decision output module; the state selecting module finds out the state with the maximum accumulative measurement in all the states and outputs the serial number of the state to the second delay module; the second delay module delays the serial number of the state with the maximum accumulative metric and outputs the delayed serial number of the state with the maximum accumulative metric to the soft decision output module; the soft decision output module calculates the likelihood ratio of all the symbols and outputs the likelihood ratio;
the reliability matrix updating unit includes: the reliability matrix mapping module and the reliability matrix updating modules are arranged, and the number of the reliability matrix updating modules corresponds to the state number of the CPM signals; at the current moment, the input of the reliability matrix mapping module is the survivor path serial number and the competition path serial number of each state at the current moment; the reliability matrix mapping module backtracks the survivor path and the competition path of each state according to the survivor path serial number and the competition path serial number of each state, finds the state passed by the survivor path of each state at the previous moment and the state passed by the competition path at the previous moment, outputs the reliability matrix of the previous moment corresponding to the state passed by the survivor path at the previous moment of each state to the reliability matrix updating module corresponding to the state, and outputs the reliability matrix of the previous moment corresponding to the state passed by the competition path at the previous moment to the reliability matrix updating module corresponding to the state; meanwhile, each reliability matrix updating module inputs and adds the path metric difference of the corresponding state of the module obtained in the comparing and selecting module; and the reliability matrix updating module corresponding to each state calculates and updates the reliability matrix corresponding to the state at the current moment, and outputs the first row of the reliability matrix to the soft decision output module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810671401.8A CN108923887B (en) | 2018-06-26 | 2018-06-26 | Soft decision decoder structure of multi-system partial response CPM signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810671401.8A CN108923887B (en) | 2018-06-26 | 2018-06-26 | Soft decision decoder structure of multi-system partial response CPM signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108923887A CN108923887A (en) | 2018-11-30 |
| CN108923887B true CN108923887B (en) | 2021-01-15 |
Family
ID=64422606
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810671401.8A Active CN108923887B (en) | 2018-06-26 | 2018-06-26 | Soft decision decoder structure of multi-system partial response CPM signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108923887B (en) |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3280834B2 (en) * | 1995-09-04 | 2002-05-13 | 沖電気工業株式会社 | Signal judging device and receiving device in coded communication system, signal judging method, and channel state estimating method |
| US7676736B2 (en) * | 2006-09-13 | 2010-03-09 | Harris Corporation | Programmable continuous phase modulation (CPM) decoder and associated methods |
| CN102148681B (en) * | 2011-05-19 | 2013-08-07 | 北京邮电大学 | System and method for realizing iterative timing synchronization of continuous phase modulation signal |
| CN102523076B (en) * | 2012-01-04 | 2014-05-14 | 西安电子科技大学 | Universal and configurable high-speed Turbo code decoding method |
| CN106487392B (en) * | 2015-08-24 | 2019-11-08 | 北京航空航天大学 | Down-sampled interpretation method and device |
-
2018
- 2018-06-26 CN CN201810671401.8A patent/CN108923887B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN108923887A (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kliewer et al. | Iterative joint source-channel decoding of variable-length codes using residual source redundancy | |
| KR19980079114A (en) | Method and apparatus for decoding trellis code data | |
| WO2006018605A1 (en) | Soft decision enhancement | |
| Suganya et al. | RTL design and VLSI implementation of an efficient convolutional encoder and adaptive Viterbi decoder | |
| US20070183489A1 (en) | Apparatus for decoding a signal and method thereof and a trellis coded modulation decoder and method thereof | |
| CN108923887B (en) | Soft decision decoder structure of multi-system partial response CPM signal | |
| US20040052318A1 (en) | Viterbi decoder | |
| US10177876B2 (en) | Sequence detector | |
| Schuh et al. | Reduced complexity super-trellis decoding for convolutionally encoded transmission over ISI-channels | |
| Sánchez et al. | High speed low complexity radix-16 max-log-MAP SISO decoder | |
| US7917565B2 (en) | High-speed radix-4 butterfly module and method of performing Viterbi decoding using the same | |
| Viraktamath et al. | Impact of code rate on the performance of Viterbi decoder in AWGN channel | |
| US7170947B2 (en) | Data receiver | |
| Moritz et al. | Turbo decoding using the sectionalized minimal trellis of the constituent code: performance-complexity trade-off | |
| Basavaraj et al. | FPGA implementation of high speed convolutional encoder and viterbi decoder for software defined radio | |
| Mahran et al. | Modified adaptive Viterbi algorithm for convolutional codes decoding | |
| Samy et al. | Low complexity iterative decoding of Reed–Solomon convolutional concatenated codes | |
| del Barco et al. | FPGA implementation of high-speed parallel maximum a posteriori (MAP) decoders | |
| Osman et al. | Performance of multilevel turbo codes with group partitioning over satellite channels | |
| Véstias et al. | Sliding block viterbi decoders in FPGA | |
| Marín et al. | Hardware architecture of MAP algorithm for turbo codes implemented in a FPGA | |
| Swathi et al. | FPGA implementation of viterbi decoder for software defined radio applications | |
| Hosseini et al. | Hardware-efficient implementation of the SOVA for SOQPSK-TG | |
| Pham et al. | Soft detection of CPM in multipath fading | |
| KR100210405B1 (en) | Method and apparatus for tracebacking survivor path of trellis-coded data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |