Digital Signal Processing

2512 Submissions

[2] viXra:2512.0073 [pdf] submitted on 2025-12-15 04:56:09

Computing the Singular Value Decomposition (SVD) with Fixed Point CORDIC Operations with Application to MIMO-OFDM

Authors: Sasan Ardalan
Comments: 33 Pages.

In this paper, the computation of the Singular Value Decomposition (SVD) of complex matrices will be presented using fixed point arithmetic. The application of CORDIC operations for fixed point implementations of the SVD of complex matrices will be introduced. SVD plays a major role in Closed Loop MIMO OFDM systems. The impact of fixed point implementation of SVD in a Closed Loop MIMO-OFDM system is examined. The ratio of Maximum to Minimum Singular Value (MMSVR) is computed for both fixed point (CORDIC) and floating point operations (using the LAPACK library). The fixed point implementation closely tracks the floating point results over fading channel models. It is shown that for highly ill-conditioned sub carriers the fixed point implementation deviates from the floating point MMSVR. This leads to noise enhancement and degradation of performance. By adding transmit diversity in Closed Loop MIMO-OFDM the MMSVR can be reduced and performance substantially enhanced for the fixed point implementation. It is also shown how SVD can be used in Open Loop MIMO-OFDM systems. This paper is an important introduction to the algorithms implemented in the GitHub repository for MIMO-OFDM:https://github.com/silicondsp/mimo-ofdm-release
Category: Digital Signal Processing

[1] viXra:2512.0062 [pdf] submitted on 2025-12-15 02:02:49

NIR Spectroscopy based Non-Invasive Blood Glucose Concentration Measurement using BP Algorithm

Authors: Hyok Chol Song, Chol Jin Oh, Chol Hyon Sim, Chol Min Won
Comments: 10 Pages.

Diabetes mellitus poses a significant challenge in clinical settings, necessitating frequent blood glucose measurements for insulin dosage determination. Conventional invasive methods, such as finger pricking, carry risks of infection and skin callusing. Non-invasive monitoring techniques offer a promising alternative for patients with hyperglycemia or hypoglycemia, enabling regular self-monitoring and advancing diabetes research. However, the accuracy and universality of most current non-invasive methods for measuring blood glucose concentration (BGC) remain inadequate. Achieving clinical credibility requires the elimination of individual discrepancies (IDs) in measurements. This study focuses on enhancing monitoring accuracy to a clinically acceptable level by mitigating the effects of IDs. We conducted a detailed analysis of factors influencing Near-Infrared spectroscopy measurements to reduce prediction error. An artificial neural network with a backpropagation algorithm was employed to predict BGC from the acquired spectral data. Experimental results confirm that our proposed BGC prediction model effectively leverages IDs, achieving performance that meets clinical standards.
Category: Digital Signal Processing