بهره برداری از فضاهای خالی تلویزیون برای توزیع سیگنال های چند رسانه ای
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|20406||2012||7 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Signal Processing: Image Communication, Volume 27, Issue 8, September 2012, Pages 893–899
The new spectrum regulation policies for dynamic spectrum access, especially those concerning the use of the white spaces in the Digital Terrestrial Television (DTT) bands, arise the need for fast and reliable signal identification and classification methods. In this paper we present a two-stage identification method for signals in the white spaces, using combined energy detection and feature detection. The band of interest is divided by means of the Discrete Wavelet Packet Transformation (DWPT) in sub-bands where the signal power is calculated. Modulation classifiers taking into account the statistical parameters of the signal in the wavelet domain are used as features for identifying the modulation schemes, in this case specifically for the DVB-T broadcast standard. Finally, a signal transmission architecture based on Motion JPEG XR has been implemented in order to explore and evaluate a practical application of indoor signal distribution over white-spaces.
Within the cognitive radio paradigm, as a highly praised alternative for overcoming the inherent limitations of the RF spectrum, the current worldwide situation of the VHF and UHF TV channels is an excellent application scenario. In the US, the complete switchover to digital television in 2009 opened an entire new topic of the usability of the TV white spaces (TVWS) for short-range wireless consumer devices. Moreover, the gradual global passage to digital television poses new specific challenges to the white spaces detection. Within this framework, spectrum sensing for DTT broadcasting signals plays a crucial role, along with geolocation databases  (GL-DBs). In the US, the Federal Communications Commission (FCC) has already commissioned the creation of GL-DBs, free to access any cognitive radio (CR)  device. The database entries provide, for a certain location (geographical coordinates), the list of available channels and the allowable maximum effective isotropic radiated power (EIRP) useful to transmit without providing harmful interference . Even if the GL-DBs are up-to date, the values provided for a specific geographical point are still the results of applying signal propagation models and estimated power levels. Due to this static approach, the provided data might be inaccurate for different reasons such as variable atmospheric conditions or multipath and fading phenomena  and . Therefore, there is still the need of a validation in terms of frequency occupancy and maximum EIRP of the free frequency channels provided by the GL-DBs, using specific spectrum sensing methods. As it is known, spectrum sensing techniques mainly focus on primary transmitter detection and can be classified into three categories: matched filter, energy detection and signal feature detection . Combinations of these methods are used for achieving good results in terms of sensitivity, computational time and signal classification, in the so-called two-stage spectrum sensing schemes proposed initially in  and then refined in  and especially in . The mentioned two-stage schemes perform coarse sensing based on energy detection, followed by a feature detection performed on the signals in the sub-bands declared free by the previous stage. This work presents a different spectrum sensing approach in a two-stage scheme using the Discrete Wavelet Packet Transformation (DWPT) for dividing the analyzed frequency band and calculating the signal power in the resulting sub-bands (channels). The sub-bands identified as free can be directly used for transmission. The remaining sub-bands with a signal power higher than a pre-defined threshold are subsequently analyzed by the feature detector, for distinguishing between primary users (PU) and possible secondary users (SU). The feature detection method used in the second stage of the spectrum sensing exploits the statistical properties of the DWPT's coefficients. The performance of the proposed system for distributing high-definition video content is tested and evaluated. In order to satisfy requirements of high-quality image compression and low computational complexity a recent standard by ISO/IEC named Motion JPEG XR has been chosen over other standard video coding tools. The remainder of the paper is organized as follows: in Section 2, we first present the use of the DWPT for sub-band division and energy detection and then we analyze the proposed feature detection method. Section 3 presents the initial software simulation, while Section 4 shows the hardware set-up and the test results using real recorded signals. Section 5 presents a signal transmission architecture based on Motion JPEG XR, its hardware implementation and the experimental results. Finally, in Section 6 we draw the conclusions and present the future work.
نتیجه گیری انگلیسی
In this paper we propose a complete implementation of a signal distribution system using the TV White Spaces. The system is based on a two-stage spectrum sensing approach for white spaces identification in the 470–790 MHz band, combining energy and feature detection methods in the wavelet domain. The current work is specifically adapted for PU signals compliant to the European DVB-T standard. Different from other approaches we are looking at the energy detection and signal classification in a combined way, both at system level and at computational level. First we implemented the proposed sensing scheme in a fully software simulator and calculated with these simulated signals the thresholds for both energy detection and the signal classification stages. The simulations validated our design for the next step, i.e. the tests with real DVB-T signals. Our spectrum sensing approach was tested in a functional system consisting of an RF hardware front-end (USRP2 SDR platform) connected to a computer running a Matlab/Simulink model. Real signals were acquired and fed offline to the computational model in order to test the system's behavior in real conditions. The receiver operation characteristics were slightly inferior to those obtained with the simulated signals. The initial tests results and the obtained ROC curves showed the need of improving the reliability of the sensing method for much lower SNR. The first step will be the calculation of new, more accurate threshold values using a large set of real DVB-T signals with various characteristics. Further steps will also include tests with different wavelet filters. The signal distribution was implemented using Motion JPEG XR over the previously identified white-space and tested over the functional system at three different bitrates (image quality). The signal distribution was implemented using two USRP2 SDR platforms in a master–slave configuration working on the previously identified white spaces. From the signal distribution point of view, we will extend the current architecture with the aim to implement a full cognitive signal distribution system to be deployed in the 470–790 MHz bands. The first tests to be performed will be coverage tests in an indoor environment, to assess the overall feasibility of the proposed system.