[Sobana, 3(1): January, 2014]

ISSN: 2277-9655 Impact Factor: 1.852

IJESRT
INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY

Design and Implementation of Map Decoding for DB Convolutional Turbo Decoder

*1,2

S.Sobana*1, M.Kasthuri2 Associate Professor, Department of ECE, PSNA College of Engineering & Technology, Dindigul, India sobanaa@gmail.com Abstract

This paper proposed an approximate Gaussian density evolution based MAP Decoder for high performance wireless communication. This paper considers a three node (source (S), relay (R) and destination (D)) wireless cooperative communication system, which can be considered as a building block for larger wireless networks. This proposed work adaptively changes the code and Modulation format according to the channel conditions, to improve performance without adding more network cooperation overhead. In proposed architecture, the higher throughput, less power consumption and less area are achieved. The architecture is implemented using spartan3E family and XC3S500E device in Xilinx 9.2i.The proposed system is written in VHDL language and synthesized in Xilinx 9.2i and stimulated using Modelsim 5.7. In 200MHz operation the Coprocessor consumes 79mW in static and 96mW in dynamic in the total summation of 175mW. Keywords: MAP Decoder, SB/DB Decoding, Likelihood Ratio, Interleaver, CTC

I. Introduction
With rapid growth of multimedia services, the Hence, a CTC decoding accelerator which works in the convolutional turbo code (CTC) has been widely adopted multi standard platforms is desired to achieve the smooth as one of forward error correcting (FEC) schemes of migration for the multiple wireless applications. The wireless standards to have a reliable transmission over CTC decoder needs the hardware implementation for noisy channels. Single-binary (SB) CTC, proposed in MAP decoding in an important issues of memory 1993 [2], has been the well-known FEC code that can organization. In particular, the power reduction of state achieve high data rates and coding gains close to the metrics cache (SMC) is critical for MAP decoders. With Shannon limit. The SB CTC code has been adopted in the SB CTC decoding, the power consumption of the the FEC schemes of wideband code division multiple SMC is reduced. Some researchers have been proposed access (WCDMA) [3], high-speed downlink packet reduction in power consumption. The reverse access (HSDPA) [3], and long term evolution (LTE) [3]. computations significantly reduce SMC power In 1999, non-binary CTC [4] was introduced to achieve consumption with reversion flag caches and reversion superior performance than the SB CTC. In recent years, checkers. However, the reversion checker and reversion double-binary (DB) CTC has been adopted in advanced flag cache prolong the critical path or decoding cycles. In wireless communication standards, such as worldwide addition, the computational complexities of the reverse interoperability for microwave access (WiMAX) [5]. computations are increased dramatically when the Some CTC decoders have been implemented as reverse computations are extended from the SB to the application-specific integrated circuits (ASICs), such as DB MAP. the HSDPA CTC decoder [6] and WiMAX CTC decoder [6]. Recently, high-end portable/mobile devices become II. Literature Survey prevalent in wireless markets. There are large growing Cooperative and relay communication has emergence and demands for an inexpensive solution to received much attention recently [3][6] due to the access the ubiquitous wireless services. Meanwhile, these potential performance improvement. At the network wireless standards, such as 3GPP and WiMAX level, relays can extend wireless network coverage if a standards, adopted CTC schemes with different coding direct source-to destination link does not exist [7]. At the parameters and different throughput rates [9], [10]. To physical level, the use of relays can achieve the diversity deal with the accelerated evolution of these standards, gains offered by multiple antenna space-time systems the multi standard platforms which can seamlessly work through using several relay nodes, possibly with only one across the multiple standards were proposed in [8]. http: // www.ijesrt.com(C)International Journal of Engineering Sciences & Research Technology [1-5]

[Sobana, 3(1): January, 2014]

ISSN: 2277-9655 Impact Factor: 1.852

20 times higher (@ 10 iterations) than the Turbo-decoder antenna per physical node [5]. The latter is the primary throughput. focus of this paper. This paper considers a three node (source (S), relay (R) and destination (D)) wireless cooperative communication system, which can be III. Proposed System considered as a building block for larger wireless We model the relative geometry of the source, networks. The basic cooperative schemes that have been relay and destination nodes as follows. The distances previously considered include Amplify-and-Forward between nodes (S-D, S-R and R-D) are normalized (AF) and Decode-and-Forward (DF) [3]. Here we against the source to destination (S-D) distance, specified consider a form of DF. It has been shown in [8] that DF as dSD = 1. The other two distances, dSR and dRD, are schemes, in which the relay nodes always forward the then always less than or equal to 1. Transmission is decoded results without indicating their reliability, tend organized in a per-packet fashion. The transmission of to not achieve the desired diversity gain due to erroneous each packet is divided into two stages, namely the decoding at the relay. The MAP decoders have to correct broadcast and relay stages in a time division format. the random error. Some selective DF was proposed in [8] During the broadcast stage, the source encodes a block of to solve this problem, where the relay keeps silent when information bits using a recursive systematic errors are detected in its decoded data. Note this usually convolutional (RSC) code and broadcasts it to both the requires the use of a cyclic redundancy check (CRC) destination and relay. No CRC is employed. The relay code embedded in the source-to-relay (S-R) link. Among decodes the broadcast message, interleaves the decoded the coded DF schemes, those using distributed Turbo bits and re-encodes them using the same or a different codes (DTCs) have been shown to offer good RSC code. It forwards the encoded packet to the performance [9], [10]. In [9], a DTC was proposed using destination during the relaying stage (while the source DF, where the two component codes are separately remains silent). The destination delays decoding until the generated at the source and relay. It, however, assumes end of the relay stage. The corresponding received an ideal S-R link. The DTC schemes proposed in [11], signals at the relay and destination for a single packet [12] showed that end to- end performance can be Here we briefly summarize the Turbo decoding improved significantly without using a CRC, by algorithm used at the destination [2]. After receiving the appropriately weighting the relayed information which encoded packet from the relay, the destination uses a may contain errors. To address this problem, a simple Turbo algorithm to decode the coded packets received non-selective DF scheme was proposed for DTCs in [2]. during the broadcast and relay stages. Each component In [2], the relay decodes the received signals without code is decoded using a BCJR decoding algorithm. performing a CRC check. A Turbo encoder has a simple The packet forwarded from the relay then has architecture, generating parity information together with much less impact on the iterative decoding of the input data by two constituent encoders. Each of the two transmitted packet and in making final data decisions at encoders encodes the input data and scrambles input data the destination. Hence, error propagation is mitigated. through an interleaver. A Turbo decoder performs The resulting cooperative Turbo decoder is shown in Fig. iterative decoding to enhance the error correction 1. Following iterative decoding, final decoding decisions capability. are made based on the sum of the extrinsic information A Turbo decoder consists of two soft input/soft variables from the component decoders for each output (SISO) decoders, each of which takes a-priori information bit. Consequently, the relay can use different information computed from the other SISO decoder as an modulations and codes from those used during the input for iterative decoding Turbo codes are extensively broadcast stage. used in current communications standards and have a promising outlook for future generations. The advantages of software defined radio, especially dynamic reconfiguration, make it very attractive in this multistandard scenario. The main drawback of turbo codes is the complex decoder structure which entails a power and area consuming VLSI implementation. Among all algorithms that can compute turbo decoding, the maximum a posteriori (MAP) algorithm [3] provides the best performance at low signal to noise levels. MAP Fig. 1 Proposed Interleaver Architecture decoders are superior with respect to communications performance and for that reason preferred in advanced implementations. The MAP decoder throughput is up to http: // www.ijesrt.com(C)International Journal of Engineering Sciences & Research Technology [1-5]

[Sobana, 3(1): January, 2014]

ISSN: 2277-9655 Impact Factor: 1.852

out for the scheme of [2]. However, it may be easily extended to similar schemes such as that of [16]. All the simulation results presented in this section are obtained by averaging over 200 packets, each of which is formed by encoding 20000 information bits. For simplicity, we consider only BPSK with bit 1 transmitted as +1 and 0 transmitted as 1. A. Channel without Interleaver The codeword dddd is altered in three bits, so it cannot be decoded (decoding failure) or might be decoded into a wrong codeword (false decoding). This of the two happens depends on the error correcting code applied shown in

Fig. 2 MAP decoder with LLR Unit

Direct Code: It is received directly from the transmitter section with low BER. Relayed Code: This code is faded by hop to hop transmission in channel with high BER. Using both direct & relayed code commonly reduces the bit error rate. While using MAP decoder we can reduce only random errors. If we want to reduce burst errors move on to interleaver & LLT Unit shown in Fig. 1 and Fig. 2. Iterative decoding can be considered as a nonlinear dynamic feedback system [14], which is very difficult to rigorously analyze. As a result, semi-analytical methods such as EXIT charts [19] and the Gaussian approximation method [13] have been developed to investigate its behavior. These methods rely on a combination of analytical modeling and simulation. They study the input-output behavior of only one component channel decoder. This allows the iterative decoding process to be analyzed using a simple non-iterative method to provide insight into the overall decoding process. The DTC scheme studied in this paper is complicated by the fact that one of the component code words may be obtained from encoding a string of bits in which a small percentage of the bits differ from the information bits contained in the other component codeword due to the errors at the relay. Thus, although we use the approximate Gaussian density evolution method [17] here, the process has to be modified in order to analyze the scheme of [2]. We first introduce some general concepts of density evolution and apply them to the three node cooperative network when the relay has error-freedecoding. We then present the specific density evolution analysis for the cooperative scheme of [2], when the relay may make decoding errors. The channels in the simulations are the channels directly connected to the destination (S-D or R-D), since the quality of the S-R link is reflected in either the relay BER or the parameter for a given R-D channel SNR. In this paper, the analysis is primarily carried

Fig. 3 Transmission of Bits without interleaver

B. Channel with Interleaver In each of the code words aaaa, eeee, ffff, gggg, only one bit is altered, so our one-bit-error-correctingcode can decode everything correctly shown in Fig. 4.

Fig. 4 Transmission of Bits with interleaver

IV. Simulation Results

In proposed architecture, the higher throughput, less power consumption and less area are achieved shown in Fig 5,6 and 7 .The architecture is implemented using spartan3E family and XC3S500E device in Xilinx 9.2i.The proposed system is written in VHDL language and synthesized in Xilinx 9.2i and stimulated using Modelsim 5.7. Dynamic power is defined as amount of power consumed by switching activities of FF, where as static power is power consumed by leakage current. In 200MHz operation the Coprocessor consumes 79mW in static and 96mW in dynamic in the total summation of 175mW.

http: // www.ijesrt.com(C)International Journal of Engineering Sciences & Research Technology [1-5]

[Sobana, 3(1): January, 2014]

ISSN: 2277-9655 Impact Factor: 1.852 V. Conclusion

Based on the simulation results, which show that a satisfactory of BER performance can be achieved. In 200MHz operation the Coprocessor consumes 79mW in static and 96mW in dynamic in the total summation of 175mW. When compared to previous results SB mode is common. But slightly differs in DB mode. This paper proposes a new decoding structure with interleaver employing the power consumption. In the proposed MAP decoder architecture, the less power consumption and memory area is achieved with DB decoder technique. The proposed architecture is easy to implement

VI. References
[1] Rui Lin, Philippa A. Martin, and Desmond P. Taylor, Approximate Gaussian Density Evolution Based Analysis of Distributed and Adaptive Turbo Codes, in IEEE Transactions on Communications, Vol. 60, No. 8, August 2012. [2] R. Lin, P. A. Martin, and D. P. Taylor, Cooperative signalling with soft information combining, J. Electr. Comput. Eng., Hindawi, vol. 2010, Article ID 530190, 2010. [3] J. N. Laneman, D. N. C. Tse, and G. W. Wornell, Cooperative diversity in wireless networks: efficient protocols and outage behavior, IEEE Trans. Inf. Theory, vol. 50, pp. 30623080, Dec. 2004. [4] M. Janani, A. Hedayat, T. E. Hunter, and A. Nosatinia, Coded cooperation in wireless communications: spacetime transmission and iterative decoding, IEEE Trans. Signal Process., vol. 52, pp. 362370, Feb. 2004. [5] R. U. Nabar, H. Bolcskei, and F.W. Kneubuhler, Fading relay channels: performance limits and space-time signal design, IEEE J. Sel. Areas Commun., vol. 22, pp. 10991109, Aug. 2004. [6] S. Borade, L. Zheng, and R. Gallager, Amplify-and-forward in wireless relay networks: rate, diversity and network size, IEEE Trans. Inf. Theory, vol. 53, pp. 3302 3318, Oct. 2007. [7] Xiao, T. E. Fuja, and J. D. J. Costello, An analysis of mobile relaying for coverage extension, in Proc. 2008 ISIT, pp. 2262 2266. [8] J. N. Laneman, Cooperative diversity in wireless networks: algorithm and architectures, Ph.D. thesis, MIT, 2002. [9] B. Zhao and M. C. Valenti, Distributed turbo coded diversity for relay channel, IEE

Fig. 5 RTL View of proposed architecture

Fig. 6 Technology Schematic View

http: // www.ijesrt.com(C)International Journal of Engineering Sciences & Research Technology [1-5]

[Sobana, 3(1): January, 2014]

ISSN: 2277-9655 Impact Factor: 1.852

Electron. Lett., vol. 39, pp. 786787, May 2003. [10] Z. Zhang and T. M. Duman, Capacityapproaching turbo coding for half-duplex relaying, IEEE Trans. Commun., vol. 55, pp. 18951906, Oct. 2007. [11] Y. Li, B. Vucetic, T. F. Wong, and M. Dohler, Distributed turbo coding with soft information relaying in multihop relay networks, IEEE J. Sel. Areas Commun., vol. 24, pp. 20402050, Nov. 2006. [12] H. H. Sneessens and L. Vandendorpe, Soft decode and forward improves cooperative communications, in Proc. 2008 IEEE 3G. [13] H. E. Gamal and A. R. Hammons, Analyzing the Turbo decoder using the Gaussian approximation, IEEE Trans. Inf. Theory, vol. 47, pp. 671 686, Feb. 2001. [14] D. Divsalar, S. Dolinar, and F. Pollara, Iterative Turbo decoder analysis based on density evolution, IEEE J. Sel. Areas Commun., vol. 19, pp. 891907, May 2001. [15] T. Richardson, A. Shokrollahi, and R. Urbanke, Design of capacityapproaching irregular lowdensity parity check codes, IEEE Trans. Inf. Theroy, vol. 47, pp. 619 637, Feb. 2001. [16] H. H. Sneessens, J. Louveaux, and L. Vandendorpe, Turbo-coded decode-andforward strategy resilient to relay errors, in Proc. 2008 IEEE ICASSP. [17] T. Richardson and R. Urbanke, The capacity of low density parity check codes under message passing decoding, IEEE Trans. Inf. Theroy, vol. 47, pp. 599618, Feb. 2001. [18] S. Lin and J. D. J. Costello, Error Control Coding, 2nd edition. Pearson Prentice Hall, 2004. [19] T. Wang, A. Cano, G. B. Giannakis, and J. N. Laneman, High performance cooperative demodulation with decode-and-forward relays, IEEE Trans. Commun., vol. 55, pp. 14271438, July 2007.

http: // www.ijesrt.com(C)International Journal of Engineering Sciences & Research Technology [1-5]