البحث — المستودع الرقمي العراقي

المستخلص: في هذا العمل, تم اقتراح خوارزميتين لتحسين اداء معيد الترميز التكراري التجميعي. الاول انجز باضافة رموز ارشاد بين سيل البيانات التي تدخل الى المرمز. في هذه الطريقة, رموزالارشاد ترمز مع البيانات المدخلة (رموز الارشاد المرمزه او الداخليه) مقارنة بالطريقة الاع | In this work, two algorithms were suggested in order to improve the performance of systematic RA decoding. The first one was accomplished by the insertion of pilot symbols between the data stream that entering the encoder. In this way the pilot symbols are encoded along with the input data (encoded or internal pilots) in contrast to the usual way where pilots are inserted after encoding (unencoded or external pilots). The positions where pilots should be inserted are chosen in such a way that to improve the minimum Hamming distance and/or to reduce the error coefficients of the code. Moreover, the known pilots have higher reliability than data and can significantly improve the initial decoding.The second proposed algorithm includes the utilization of the inserted pilots to estimate scaling (correction) factors. Two - dimensional (2D) correction system was suggested in order to enhance the performance of traditional Minimum - Sum (MS) decoding of regular RA codes. By calculating the mean square difference (MSD) between the values of received pilots and the a - posteriori data of bit and check node related to them, which are created by the MS decoder, an adaptive method can be achieved for getting the correction factors.Since encoded pilots in systematic RA code are embedded in data, it is possible to puncture all or part of them to increase the code rate. The pilots still have their effect on increasing the minimum Hamming distance.Extensive simulation tests reveal that the suggested quasi - optimal algorithms supply error performance be able to compare as Sum - Product (SP) decoding whilst maintain less complexity. Practical experiments are carried out by utilizing the capability of data acquisition toolbox supported by MATLAB to provide a hardware interface with the computer sound card. Here, the sound card plays the role of ADC and DAC converter in the physical layer. A different graphical user interfaces (GUI) are designed for the transmitting and receiving terminals to facilitate the selection of different code options. Signals are transmitted over wireless channel using a 2.4 GHz band indoor audio tranceiver. According to the aquired results, the modified coding system outperforms the original one when it is tested over AWGN and fading channels. An improvement may be realized about 1.5 dB and 1.3 dB at BER=?10?^( - 5) for punctured and unpunctured systematic (50,35,3,7) code with adaptive scaling, with three encoded pilots over AWGN channel and about 3.4 dB and 3.3 dB at BER=?10?^( - 4) for punctured and unpunctured systematic (50,35,3,7) code with constant scaling, with two encoded pilots over fading channel.