Asma Maalej

Data acquisition test platform for non uniformly controlled ADC

Non Uniform Sampling (NUS) was presented as an emerging solution to reduce aliases for ADC (Analog-to-Digital Converter) in Software Defined Radio receiver. In this paper, practical implementations of NUS, called TQ-RS (Time Quantized Random Sampling) are presented. A test setup for non uniformly controlled data acquisition system is detailed. Experimental results are presented and discussed.

Pseudo-random Direct Sampler for frequency down-conversion in a multi-standard receiver

For software defined radio, non uniform sampling (NUS) is a process promising IF down-conversion with alias reduction. Applied to multi-standard receiver, this technique relaxes system design constraints. This paper presents a new design strategy for clock generation in NUS based multi-standard radio receiver supporting GSM, UMTS and Wi-Fi standards. Sampling frequency choice is discussed in order to get only one shared anti-aliasing filter suitable for all considered standards. The proposed non uniform clock generator, the pseudo-random direct sampler (PDS), promises a direct driving of the sampler circuit leading to low-power oriented design regarding classic RF sampling receiver.

Non uniform sampling for power consumption reduction in SDR receiver baseband stage

In this paper, authors point out the effect of non uniform sampling (NUS) on the power consumption in SDR receiver baseband stage. This feature is drawn by theoretical formulas regarding spectrum shape and power estimation of different baseband components. An example is drawn to focus on NUS ability to reduce anti-aliasing filter (AAF) consumption by 25%.

Towards time-quantized random sampling for multistandard receiver baseband stage

In this paper, the ability of time-quantized random sampling (TQ-RS) to reduce aliases is discussed. Authors present simulation results of power spectrum density of a time-quantized random sampled signal. The oversampling ratio impact on alias reduction is outlined. TQ-RS application is proposed for baseband stage design of software defined radio receiver supporting the GSM, the UMTS and the IEEE802.11a standards. Obtained design results are compared to design results considering the uniform sampling. This comparison shows that the anti-aliasing filter order has been reduced from 4 to 3 and the maximal gain of automatic gain control is decreases with at least 12 dB. A test setup for data acquisition system with TQ-RS is detailed. Validation results confirm the targeted performances in terms of signal-to-noise ratio (SNR) of TQ-RS acquisition. The SNR is equal to 53.13 dB in TQ-RS compared to 54 dB in uniform sampling.

Test setup and spurious replicas identification in time-quantized pseudorandom sampling-based ADC in SDR multistandard receiver

Software defined radio (SDR) multistandard receiver is an enabling technology for radio evolution. The time-quantized pseudorandom sampling (TQ-PRS) is an advantageous processing for SDR multistandard receiver design. The TQ-PRS allows aliases attenuation leading to a relaxed receiver baseband stage. In this paper, the authors focus on experimental proof of concept of the TQ-PRS-based analog-to-digital converter (ADC). The spectrum analysis of the test results shows spurious replicas which are not present in simulation results. The presented investigations show that the spurious replicas are coming from a pseudorandom delay with the same periodicity as TQ-PRS clock. A solution based on sharing receiver channel selection filter for spurious attenuation is proposed.

Pseudo-random oscillator design for multistandard RF Non Uniformly Sampling receiver

Non Uniform Sampling technique applied to Software Defined Radio theme seems to be a suitable solution to avoid aliases and reduce their drawbacks on other terminal components especially filters. The proposed work presents a solution to apply Non Uniform Sampling in a radio receiver. This paper deals with designing and implementing a non uniform generator that manages the sampling step. The proposed generator has an architecture close to the Direct Digital Synthesizer one.

Pseudorandom Direct Sampler for Non-Uniform Sub-sampling Architecture in a Multistandard Receiver

In this paper, a Non-Uniform Sampling (NUS) technique for down-conversion stage in a multistandard radio receiver is proposed. For both narrowband and wideband standard processing, NUS promises relaxing system design constraints, decreasing the sampling frequency as well as reducing power consumption. A non-uniform clock generator, called Pseudorandom Direct Sampler (PDS), is described. PDS is used to non-uniformly control the Analog-to-Digital Converter (ADC) performing IF sub-sampling in proposed GSM/UMTS/WiFi multistandard receiver architecture. PDS architecture is based on using modified Direct Digital Synthesizer (DDS) including pseudorandom behavior. A 90-nm CMOS FPGA based prototype of PDS reveals an internal clocking up to 350 MHz and a power consumption lower than 4 mW.

Time-domain characterization of a wireless ECG system event driven A/D converter

Heart strokes are one of the leading cause of global deaths. In order to prevent such strokes, a variety of wearable devices have been proposed recently. These are based on the wireless Electrocardiogram (ECG) acquisition, analysis and transmission. An efficient acquisition, analysis and transmission of such signals can be achieved by adapting the system activity according to the input signal local characteristics. In this context, an event driven analog-to-digital converter (ADC) based on level-crossing sampling scheme (LCSS) is designed for the ECG acquisition. The analog-to-digital (A/D) conversion error is a vital parameter to characterize its performance. It implicitly provides information about the ADC effective resolution. In this context, a novel method is proposed to measure the conversion error in time domain of the designed level-crossing ADC (LC-ADC). In the aim of acquiring the desired conversion precision, a criterion for properly choosing the system parameters is also discussed.

Mathematical modeling of clean and noisy ECG signals in a level-crossing sampling context

In this paper, a state-of-the-art review of electro cardiographic (ECG) signal modelling techniques is presented. The mathematical Gaussian modelling is selected as the best technique to generate realastic ECG signals regarding its ability to preserve relevant information of ECG databases signals. The generated Gaussian ECG signal model is applied to a level-crossing converter to measure the accuracy and the efficiency of sampling and to evaluate the effects of the signal noise on the sampling paradigm. The performances of the conversion were evaluated in terms of quality of the reconstructed signal. A percentage root mean square difference (PRD) around 2 % is obtained. It is shown that the LC-ADC is robust against respiratory and movement noise. However, for low signal distortions the electromyogram (EMG) noise should not exceed 20 μν which is less than IEC 60601-2-47 specifications.