Th-Uwb Receiver Based on Two PDFS Approximation in Multiuser Systems

Abstract

Abstract-As the distribution of the multiuser interference plus noise in time-hopping ultrawide bandwidth (TH-UWB) systems can not be reliably approximated by Gaussian probability density function (PDF), the bit error rate (BER) performance of the conventional matched filter receiver in TH-UWB multiuser systems degrades. In this paper we study a novel TH-UWB receiver based on two PDFs approximating the distribution of the multiuser interference plus noise. Firstly we investigate the difference between the distribution of the multiuser interference plus noise when the received pulse is collided by interfering pulses and that when it is not. Then two PDFs are developed to approximate the distribution of multiuser interference plus noise in these two cases respectively instead of using one PDF for both cases as done in other research works. Based on these two PDFs, a new detection scheme of TH-UWB receiver is proposed. The results show that BER performance of the proposed receiver is improved by 50% or more as compared to the conventional matched filter receiver, blinking receiver, Gaussian mixture receiver and p-order receiver using simulations.

A Maximum Entropy inspired model for the convolutional noise PDF

“…i are the real and imaginary parts of the Lagrange multipliers related to the convolutional noise pdf respectively derived in Section 3.3: (16) and from [1]:…”

“…In our simulation we use the 16QAM input case, a modulation with ±{1, 3} levels for in-phase and quadrature components. For the 16QAM case the Lagrange multipliers λ s 2 , λ s 4 , λ s 2 and λ s 4 are taken according to (16) and (17) …”

“…In the literature we may find several works [12][13][14][15][16] dealing with pdfs applicable for the non-Gaussian case but encompasses also the Gaussian model. The multiple-user interference (MUI) in time hopped impulse-radio ultrawide bandwidth (UWB) systems is impulse-like and poorly approximated by a Gaussian distribution [12].…”

“…Several alternative distributions for approximating the MUI process and the MUI-plus-noise process in UWB systems are motivated and compared in [12]. In [16], the approach of using two pdfs is adopted to approximate the distribution of the multiuser interference plus noise. Specifically, a generalized Gaussian pdf is used in [16] to approximate the distribution of the total disturbance when the received pulse is collided and a Gaussian pdf is applied to approximate the distribution of the total disturbance when the received pulse is not collided.…”

“…In [16], the approach of using two pdfs is adopted to approximate the distribution of the multiuser interference plus noise. Specifically, a generalized Gaussian pdf is used in [16] to approximate the distribution of the total disturbance when the received pulse is collided and a Gaussian pdf is applied to approximate the distribution of the total disturbance when the received pulse is not collided. In [14], a new detector is proposed for amplify-and-forward (AF) relaying system when communicating with the assistance of relays where the probability density function is estimated with the help of the kernel density technique and a generalized Gaussian kernel is proposed that provides more flexibility and encompasses Gaussian and uniform kernels as special cases.…”