CASPIAN JOURNAL

MANAGEMENT AND HIGH TECHNOLOGIES

Development of algorithm optimization parameters multithreshold decoder self-orthogonal codes the component wise descent method

Read Shevlyakov Dmitriy A. Development of algorithm optimization parameters multithreshold decoder self-orthogonal codes the component wise descent method // Caspian journal : management and high technologies. — 2015. — №3. — pp. 75-85.

Shevlyakov Dmitriy A. - post-graduate student, Ryazan State Radio Engineering University, 59/1 Gagarin St., Ryazan, 390005, Russian Federation, d.a.shevlyakov@gmail.com

In modern communication systems to correct any transmission errors using error-correcting coding methods/decoding. The article gives a brief description of one such method - multithreshold decoder (MTD). The problems associated with increasing the efficiency of the MPD through the use of multi-dimensional optimization of the parameters of the method. Based on the fact that optimization should be carried out by several groups of parameters was chosen method of coordinate descent. Its algorithm is applied to this problem is described in detail in the article. It is shown that the developed algorithm allows to make the correct settings and thus for this class of problems save time and computing resources by 15 decimal places. The results MTD work with optimized parameters that allow the conclusion that the best of his work area is shifted by 0.6 dB in the direction of the channel capacity compared to non-optimized for this. Thus, the results obtained by this algorithm in a channel with additive white Gaussian noise, let adjust MTD (without complication) to provide the best results. It is possible to recommend the MTD for use in modern communication systems in order to improve its quality and noise immunity.

Key words: цифровая связь, системы передачи цифровой информации, помехоустойчивое кодирование, достоверность передачи данных, энергетический выигрыш кодирования, самоортогональные коды, многопороговые декодеры, компьютерное моделирование, методы многомерной оптимиза