TOA Estimation With Dynamic-threshold For IR-UWB System Based On Quantized Compressed Sensing

Impulse Radio Ultra-wideband (IR-UWB) possesses numerous advantages such aslow cost, low power consumption, especially the capability in strong multi-pathresolving, it is suitable for indoor multipath environments ranging, can support theinferior of GPS signal in penetrability capability. Studies have shown that the rangingaccuracy of ultra-wideband signals can reach centimeter level. However, due toIR-UWB signal with high bandwidth, the sampling frequency of the ADC(Analog-to-Digital Converter) in digital receiver is hard to meet the samplingrequirements, if the ADC designed to meet the sampling requirements of ranging signal,it is difficult to meet the feature low cost for IR-UWB systems. The difficulty ofincreasing the sampling rate of ADC makes the IR-UWB ranging system difficult toapply to practice, and this problem becomes a bottleneck of IR-UWB ranging system.Compressive Sensing(CS) technology has become a hotspot of applied mathematicsresearch in recent years. Compressed Sensing theory indicate that for a signal which canbe compressed,we can use the sampling well below the Nyquist rate sampling tocompress the signal, and have the ability to accurately recover the signal, just providedan opportunity to solve the IR-UWB ranging system bottleneck. This article is studiedon how compressed sensing technology used in IR-UWB ranging system. The mainthesis work is as follows:(1) Proposed a mechanism that non-uniform quantization overload ranging basedon CS framework. Firstly, according to the characteristics of the ranging systemdesigned sparse representation of the received signal, making the ranging can be doneunder the framework of the CS, and then discuss the selection of the measurementmatrix and reconstruction algorithm selection, for the purpose of ranging estimation tobe prepared. A characteristic of the paper is that, before reconstructing the channelmodel, the full account of the receiver thermal noise and quantization noise impact forthe ranging process is done, and from a practical perspective on the real-time estimateof the thermal noise, combined with the design characteristics of CS theory anon-uniform quantization overload mechanism, ensuring ranging accuracy of thepremise, a greater degree of increase ranging efficiency and reduce the time complexityof ranging system.(2) Designed the dynamic threshold policy based on the mixed noise and peakvalues. After the reconstruction of the channel, we can use TOA (Time Of Arrive)estimation methods ranging from information obtained in the reconstructed signal.However, a fixed threshold can not meet TOA estimation complex noise environments,in order to improve ranging accuracy, we propose a dynamic threshold-based TOAestimation method. Affecting the threshold selected main factors size of the noise and channel individual characteristics, noise includes thermal noise and quantization noise,channel individual characteristics mainly refers multipath quantity and multi-pathdistribution, this article with kurtosis size to characterize the channel individualcharacteristics. From the end of the simulation results can be seen, based on mixednoise and kurtosis values obtained dynamic threshold strategies ranging accuracy ismuch better than a fixed threshold strategy, ranging accuracy can reach centimeter level.