From the above the inference is that high free distance codes with less multiplicity will have the maximum error bound at less range and hence the performance will be good. Study on turbo codes specifies that block interleaver based turbo codes have low free distance as well as with high multiplicity.
This affects the performance of overall turbo codes to a great extent particularly average to higher SNRs. One of the effective solutions to lessen the error floor effect is utilization of a suitable interleaver compatible with the structure of constituent RSC encoders. In this case, input bit streams, which produce low weights for the first RSC code are permuted by an interleaver in a way that prohibits generation of low weights for the second RSC code to increase free distance value of the turbo code.
It has been accepted that the best performance is achieved by random interleavers [5]. But the implementation of such an interleaver where the interleaving takes place on a random basis and the de-interleaving should also takes place in the same random basis is very difficult. In addition to implement and use same random structure in interleaving and de-interleaving these have to be synchronized to some extent for which the input stream of bits may be required to be stored.
This is really not anticipated in some cases, where the length of the bit stream is so large [6]. The block interleavers incur more memory so as to maintain the shifting order. But non-block interleavers are designed with less memory and also have self-synchronization property with their de- interleavers so that the complexity of non-block interleavers is less.
To utilize the above features of non- block interleavers in turbo codes, these interleavers need to operate as block interleavers. This can be simply accomplished if few bits are injected at the end of every data block driving the interleaver memories to the known state. In addition to utilizing advantages of non-block interleavers, this process eases the coding and decoding analysis. To reduce the number of the bits, few optimizations may be done to the interleaver arrangement. A number of studies were undertaken on the importance of interleaver on the turbo code performance [7]-[9] and a number of interleavers are presented in the literature with diverse features and complexities [10] — [13].
The interleaver which is basically introduced to scramble the order of information bits takes enormous amount of time in cases. Sri Manu Prasad and Dr. Venkateswarlu the complexity is less then may be the time for execution is less and error probability will be high.
The otherwise cases will be either worst or ultimate. In the case of block interleaver also when the execution time is less then the capacity of converting burst errors into simple errors is less and consequently the error probability is high.
In this paper a novel scheme was presented which aims to reduce the time complexity of the interleaving process to great extent while maintaining a similar error performance. The rest of the paper is organized as follows. The proposed hybrid interleaver was presented in section III. Experimental results are presented in section IV.
Concluding remarks are given in section V. Block Interleaver Block interleaver as mentioned earlier rearranges the input bit stream. Block interleavers can be classified into mainly three categories. They are Matrix interleaver, Random interleaver and Algebraic interleaver. From the expression it is obvious that if the number of rows is large than that of columns the time complexity is very less. But in this case the capability of the interleaver in converting the burst error to simple error is limited to few bits.
When the number of rows is less than that of columns then the time complexity is high and at the same time the length of the burst that can converted to simple error is high.
The bits input to the interleaver are denoted by i1, i2, i3,. The bit sequence i1, i2, i3,. The output is pruned by deleting dummy bits that were padded to the input of the rectangular matrix before intra-row and inter row permutations, i.
Now let us take another look at block interleavers. Consider matrix interleaver for the time and memory complexity analysis. Assume the incoming flow of data bits are framed as MXN array.
The values of M and N depend on the expected size of burst error. The memory requirement at the interleaver stage will be MN bits assuming bits as reference. Similar memory requirement is needed at the de-interleaver as well.
Hence a total of 4MN bits of memory is required for MN bits of data. The time complexity is crucial thing in fast communications. Some of the practical installations omit the interleaving and de-interleaving in order to reduce the latency introduced by the interleaver pair. The other interleaver considered is 3GPP interleaver which is basically the improved matrix based block interleaver suggested by Third Generation Partnership Project.
The time and memory constraints are close to that of standard block interleaver. When the number input frames is high then the complexity of both the interleaving schemes becomes similar because in 3GPP interleaver in addition to standard matrix writing and reading there are some additional operations which are somehow scalar operations. Now if the total bits in input frame are split into subgroups and if the individual groups are processed using the normal interleavers the time complexity can be reduced.
Let the input frame is split into 4 groups. But when the total input bits are split into four groups and applied to separate interleavers, the four groups of input bits are concentrated in that region only. That means the first quarter of bits is placed again in the first quarter only, similarly the remaining quarters. Hence a second stage of interleaving is proposed in which these quarters are interleaved based on a direct assignment without using another interleaver.
Hence this scheme can be regarded as a two stage interleaving scheme. In this paper block interleavers are used and any other interleavers and any other combinations can also be considered. The structure of the proposed interleaver is shown in figure 1. Capacity of two-relay diamond networks with rate-limited links to the relays and a binary adder multiple access channel. Song, H. Combined constrained code and LDPC code for long-haul fiber-optic communication systems.
Du, Q. Security enhancement for wireless multimedia communications by fountain code. Google Scholar. Song, J. Can gray code improve the performance of distributed video coding? IEEE Access , 4 , — Shojafar, M. Improving channel assignment in multi-radio wireless mesh networks with learning automata.
Wirel Pers Commun , 82 1 , 61— Turbo block codes for the binary adder channel. Peterson, W. Error-correcting codes. Cambridge: MIT Press. Wolf, J. Efficient maximum likelihood decoding of linear block codes using a trellis.
Morelos-Zaragoza, R. The art of error correcting coding. Pyndiah, R. Near-optimum decoding of product codes: block turbo codes. Haesik, K. Low complexity iterative decoding of product codes using a generalized array code form of the Nordstrom-Robinson code. Cover, T. Elements of information theory 2nd edition p. Hoboken: Wiley-interscience. Hagenauer, J. Iterative decoding of binary block and convolutional codes. Franz, V. Concatenated decoding with a reduced-search BCJR algorithm. Sabeti, L.
In: IEEE 62nd vehicular technology conference, Benchimol, I. Low complexity trellis representations of convolutional codes via sectionalization of the minimal trellis. Telecommun Syst , 59 4 , — Download references.
You can also search for this author in PubMed Google Scholar. Correspondence to M. Reprints and Permissions. Turbo decoding of simple product codes in a two user binary adder channel employing the Bahl—Cocke—Jelinek—Raviv algorithm.
Telecommun Syst 68, — Download citation. Published : 07 November Issue Date : July Anyone you share the following link with will be able to read this content:. Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative. Skip to main content. Search SpringerLink Search. References 1. Accessed 2 May 2. Article Google Scholar 4. Article Google Scholar 5. Article Google Scholar 6. Article Google Scholar 7.
Article Google Scholar 8.
0コメント