CN103346860A - Decoding unit of wireless transmission system and decoding method thereof - Google Patents

Abstract

The invention discloses a decoding unit of a wireless transmission system and a decoding method of the decoding unit of the wireless transmission system. The decoding unit of the wireless transmission system mainly comprises a Hamming distance calculating module, an add-compare-select module, a surviving route storage module, a minimum value selecting module and a backtracking module. According to the decoding unit of the wireless transmission system and the decoding method of the decoding unit of the wireless transmission system, a set of designing and achieving scheme suitable for convolutional coding of a high-definition video wireless transmission system is developed independently for key problems of timeliness and reliability of coding and decoding and usage of hardware resources and the like. The decoding unit and the decoding method are applied to multiple modern communication fields such as multi-carrier wireless communication, portable type wireless video communication, a handheld type transmission terminal and a high-efficient RF communication, and can effectively and reliably solve the problems that error code correction is generated after massage source codes are transmitted through an information channel.

Description

A kind of decoding unit of wireless transmitting system and interpretation method thereof

Technical field

The present invention relates to utilize Coded Orthogonal Frequency Division Multiplexing (COFDM) to carry out correcting in the HD video wireless transmitting system of data processing technique code translator and the method for error code, it is adapted to convolutional encoding and the Viterbi decoding of HD video wireless transmitting system.

Background technology

In recent years, along with GPRS, CDMA, 3G(TD-SCDMA, WCDMA, CDMA2000), the rise of cordless communication network such as OFDM, wireless video transmission system is with its maneuverability, visual pattern, advantage such as easy to operate, be widely used in every field such as army individual-soldier operation, public security remote monitoring evidence obtaining, TV programme relay, the monitoring of mine oilfield safety, the environmental monitoring of underground garage supermarket, cell environment monitoring, and become that people realize at a distance, a kind of important means of on-site supervision on a large scale.The COFDM technology is strong with its transmission rate height, antijamming capability, and this makes realizes that at " in the high-speed motion " and " non-looking under the poker spare " high-quality realtime graphic and transfer of data become possibility.

Aspect the radio communication encoding and decoding technique, channel coding/decoding technology based on convolutional encoding gains great popularity because of advantages such as its code efficiency height, good in anti-interference performance, therefore convolutional encoding and decoding technique are paid close attention to widely, are that domestic and international experts and scholars study one of focus.But in wireless communication procedure, noise jamming in the channel can cause certain error rate of receiving terminal, this will influence the accuracy of wireless data transmission, particularly in the algorithm implementation procedure of convolutional encoding and Viterbi decoding, source code produces code symbols by encoder, after propagating through noisy channel, the error code phenomenon takes place, thereby can cause decoding error, this problem is perplexing the designer at present always.

Summary of the invention

Technical problem to be solved by this invention is to realize a kind of constraint length at sign indicating number hour, and decoding efficiency is higher, and the faster structure of speed relatively simply is easy to the Convolutional Decoder Assembly realized.

To achieve these goals, the technical solution used in the present invention is: a kind of decoding unit of wireless transmitting system, the Hamming distance computing module calculates the Hamming distance between the coding possible under input data and the current encoder state, and the Hamming distance signal that calculates is delivered to ACS module and path memory module, the ACS module is delivered to minimum value with the accumulation distance signal and selects module, and relatively drawn by it and lessly in distance signal to be delivered to the path apart from the path as the survivor path value and to recall module, described path is recalled module survivor path value and is delivered to path memory module storage, outwards exports according to the Hamming distance signal of survivor path value and the input of path memory module simultaneously and draws decode results.

A kind of interpretation method of wireless transmitting system:

Step 1, reception signal;

Step 2, according to the length of the corresponding status switch in path and receiving sequence calculating path branch;

The length that step 3, the path branches that adds up obtain and be path;

Step 4, relatively to obtain the shortest path length in arbitrary moment be survivor path;

Step 5, the corresponding input symbols of survivor path are decode results just.

Further, source code V _kBy encoder X _k, produce code symbols, be designated as Y _k, after the noisy channel propagation, the reception signal is Z _k

The length of path branches in the described step 2:

$\mathrm{\λ}\left({\mathrm{\ξ}}_k\right)\stackrel{\mathrm{\Δ}}{=}-\ln P\left(X_{k+1}\right|X_k)-\ln P\left(Z_k\right|{\mathrm{\ξ}}_k)$

Wherein, ξ _kExpression is from X _kTo X _K+1State transitions

In the described step 3 path be-lnP (X, Z),

Wherein X is and the corresponding status switch in this path, and Z is receiving sequence,

$P(X,Z)=P\left(X\right)P\left(Z\right|X)=\underset{k=0}{\overset{K-1}{\mathrm{\Π}}}P\left(X_{k+1}\right|X_k)\underset{k=0}{\overset{K-1}{\mathrm{\Π}}}P\left(Z_k\right|X_{k+1},X_k)$

Then $-\ln P(X,Z)=\underset{k=0}{\overset{K-1}{\mathrm{\Σ}}}\mathrm{\λ}\left({\mathrm{\ξ}}_k\right)$ 。

The present invention be directed to encoding and decoding key issue independent developments such as real-time, reliability and hardware resource use the design implementation deciphered of the one cover convolutional encoding that is adapted to the HD video wireless transmitting system and Viterbi, this decoding unit and interpretation method are applicable to a plurality of fields of modern communications, fields such as multi-carrier wireless communication, portable mobile wireless video communication, hand-hold type transmission terminal and efficient RF communication application for example reliably, has effectively solved source code produces error code correction behind the dissemination channel of making an uproar a difficult problem.

Description of drawings

Below the content of every width of cloth accompanying drawing expression in the specification of the present invention and the mark among the figure are made brief description:

Fig. 1 is Viterbi decoding algorithm schematic diagram;

Fig. 2 Viterbi decoding unit is realized block diagram;

Fig. 3 Hamming distance computing module theory diagram;

Fig. 4 ACS module principle block diagram;

Fig. 5 recalls in the path module principle block diagram

Fig. 6 is Viterbi decoding simulation result figure.

Embodiment

The present invention is based on the Viterbi decoding technique of the HD video wireless transmitting system of COFDM; COFDM(Code Orthogonal Frequency Division Multiplexing) be Coded Orthogonal Frequency Division Multiplexing (COFDM), it is a kind of novel data processing technique, rely on its high availability of frequency spectrum, high transmission rate and strong anti-multipath interference performance, caused that at the wireless image transmission field people pay close attention to widely; Viterbi deciphers a kind of decoding algorithm of convolution code, and the error correcting capability that it is very strong has application in a lot of systems, and this has wherein just comprised wireless video transmission system.This achievement is devoted to solve the decoding difficult problem of the convolution code of ofdm system, is based on one of the core key technology of the high-resolution video transfer system of COFDM, is applicable to foundation and enforcement based on the high-resolution video transfer system of COFDM.

Source code V _kBy encoder X _k, produce code symbols, be designated as Y _k, after the noisy channel propagation, the reception signal is Z _kThe work that Viterbi decoding unit and interpretation method thereof are done is exactly to utilize decoding algorithm, from Z _kJudge V _k, recover information source information.

If the shift register in the convolution coder has N position, X so _kState will have 2 ^NKind.Whenever a source code enters encoder, the state of shift register changes once, and export one with the unique corresponding codes value of this state.Viterbi decoding is to select most possible one in the path of all possible state conversion of mark, and the corresponding source code of this paths is exactly decode results.

According to above-mentioned principle, receiving Z _kAfter, at first, according to the length of the corresponding status switch in path and receiving sequence calculating path branch, the length of path branches is: $\mathrm{\λ}\left({\mathrm{\ξ}}_k\right)\stackrel{\mathrm{\Δ}}{=}-\ln P\left(X_{k+1}\right|X_k)-\ln P\left(Z_k\right|{\mathrm{\ξ}}_k)$

Wherein, ξ _kExpression is from X _kTo X _K+1State transitions;

The definition path be-lnP (X, Z), wherein X is and the corresponding status switch in this path, Z is receiving sequence,

$P(X,Z)=P\left(X\right)P\left(Z\right|X)=\underset{k=0}{\overset{K-1}{\mathrm{\Π}}}P\left(X_{k+1}\right|X_k)\underset{k=0}{\overset{K-1}{\mathrm{\Π}}}P\left(Z_k\right|X_{k+1},X_k)$

Then overall path can be expressed as $-\ln P(X,Z)=\underset{k=0}{\overset{K-1}{\mathrm{\Σ}}}\mathrm{\λ}\left({\mathrm{\ξ}}_k\right)$ ，

Be path branches length and.Therefore, the shortest path in arbitrary moment is exactly each individual path length and minimum path, is called survivor path.And for each moment k, all have 2 in the grid chart ^NBar survivor path, every paths arrive k+1 state X constantly _k

The corresponding input symbols of survivor path that relatively obtains arbitrary moment just is decode results.

Be example with (2,1,2) convolution code, as shown in Figure 1, S ₀-S ₃4 kinds of possibility states of expression encoder for convolution codes, to 7T, each state has two branches from time point 0T, points to by the source code V that newly enters encoder _kThe state that bring (0 or 1) changes, and for example current state is S ₀, if input V _k=0, then next moment state still is S ₀, output encoder Y _k=00.Otherwise if be input as 1, state changes over S ₁, output encoder Y _k=11. output encoders with individual path compare with the coding that receives, and have just obtained Hamming distance, compose and give this section individual path as path.The digitized representation Hamming distance on next door, path among Fig. 1, numeral in the circle arrive this state shortest path length and, suppose the coded sequence Z={00 that receives, 01,10,00,00,00,00}, calculate successively each Hamming distance constantly and this state of arrival shortest path and, select the paths of lacking most constantly at 7T, the corresponding input symbols in this path is decode results just.

According to above-mentioned algorithm principle, utilize programmable logic device FPGA to make up hardware and implement block diagram as shown in Figure 2, the decoding unit of wireless transmitting system adds than choosing (ACS) module mainly by the Hamming distance computing module, survivor path memory module, minimum value are selected module and are recalled the module formation.

Be described below with regard to each module and function conjunction with figs. thereof below:

As shown in Figure 3, the Hamming distance computing module mainly is responsible for calculating the Hamming distance between the coding possible under input data and the current encoder state.Under a coder state, two kinds of possible codings outputs are arranged, so when the input data are 2, corresponding 4 Hamming distances.These 4 Hamming distances are given ACS module, and each ACS example is selected two corresponding distances according to the coder state of handling, and these two distances are that the state diagram according to encoder obtains.

As shown in Figure 4, the ACS module mainly is responsible for calculating the previous moment accumulation distance of current encoder state and the Hamming distance sum of current this state of arrival, produce new accumulation distance, and arrive the path accumulation distance to two of each state and compare, select the path of less distance as survivor path, preserve this simultaneously apart from sue for peace reference range relatively constantly as next.

The path memory module mainly is responsible for depositing the survivor path value that the ACS module produces, and the path recalls module survivor path value, and module is recalled in Hamming distance signal to the path of path memory module input simultaneously.

Minimum value selects module mainly to be responsible for when decoding time reaches decoding depth all survivor paths being compared, and selects a minimum paths.

Module mainly responsible final state and corresponding survivor path value of each moment according to minimal path are recalled in the path, determine the back tracking point of previous stage, finish up to all recalling, and give fullpath and survival value for change.At first select module to obtain final state by minimum value, the highest order of each state is exactly the survival value, it is exactly the state of previous moment that low five survivor path values with previous moment are reformulated 6, and the like can obtain all survival values, the survival value sequence be exactly decoding sequence the inverted sequence sequence.As shown in Figure 5, K+1 state constantly is 000000, and highest order is 0, and then K+1 survival value constantly is 0, low five of K+1 state constantly is that 00000, K survivor path value constantly is 1, and six of composition is 000001, then K state constantly is 000001, and highest order is 1, and the survival value is 1.

The simulation result of Viterbi decoding as shown in Figure 6 is input as (11,10,00,10,01,10,10,10,10,10,11,10,01,00,11,00,10,10,11,00 ...) decode results is (1,0,1,1,0,1,0,0,1,1,1,1,1,1,1,0,0,1,1,1 ...) consistent with standard results, show the feasibility that this Viterbi deciphers.

Claims (3)

1. the decoding unit of a wireless transmitting system, it is characterized in that: the Hamming distance computing module calculates the Hamming distance between the coding possible under input data and the current encoder state, and the Hamming distance signal that calculates is delivered to ACS module and path memory module, the ACS module is delivered to minimum value with the accumulation distance signal and selects module, and relatively drawn by it and lessly in distance signal to be delivered to the path apart from the path as the survivor path value and to recall module, described path is recalled module survivor path value and is delivered to path memory module storage, outwards exports according to the Hamming distance signal of survivor path value and the input of path memory module simultaneously and draws decode results.

2. the interpretation method of a wireless transmitting system is characterized in that:

Step 1, reception signal;

Step 2, according to the length of the corresponding status switch in path and receiving sequence calculating path branch;

The length that step 3, the path branches that adds up obtain and be path;

Step 4, relatively to obtain the shortest path length in arbitrary moment be survivor path;

Step 5, the corresponding input symbols of survivor path are decode results just.

3. interpretation method according to claim 2 is characterized in that:

Source code V _kBy encoder X _k, produce code symbols, be designated as Y _k, after the noisy channel propagation, the reception signal is Z _k

The length of path branches in the described step 2:

$\mathrm{\λ}\left({\mathrm{\ξ}}_k\right)\stackrel{\mathrm{\Δ}}{=}-\ln P\left(X_{k+1}\right|X_k)-\ln P\left(Z_k\right|{\mathrm{\ξ}}_k)$

Wherein, ξ _kExpression is from X _kTo X _K+1State transitions

In the described step 3 path be-lnP (X, Z),

Wherein X is and the corresponding status switch in this path, and Z is receiving sequence,

$P(X,Z)=P\left(X\right)P\left(Z\right|X)=\underset{k=0}{\overset{K-1}{\mathrm{\Π}}}P\left(X_{k+1}\right|X_k)\underset{k=0}{\overset{K-1}{\mathrm{\Π}}}P\left(Z_k\right|X_{k+1},X_k)$

Then $-\ln P(X,Z)=\underset{k=0}{\overset{K-1}{\mathrm{\Σ}}}\mathrm{\λ}\left({\mathrm{\ξ}}_k\right)$ 。

Images