Real-time, high-accuracy frequency-phase estimation is the critical mission of Doppler tracking, which is a primary technique for deep space spacecraft navigation and planetary radio science experiments. Usually, the analog intermediate frequency signal is digitalized and converted to baseband by signal processing hardware platforms called digital back-ends (DBEs) and parameter estimation is performed by extra high performance computers. In this paper, a novel real-time, high-accuracy parameter estimator called a hardware-based integrated parameter estimator (HIPE) is proposed and implemented inside DBEs. An adaptive frequency tracker is proposed to make the initial signal detection, frequency tracking, and data reduction. Then a parameter estimation is sequentially obtained by a modified dechirp technique and a high-resolution spectral analysis technique called spec-zooming. Further, a folding architecture is designed to save hardware resources when realizing spec-zooming in a field programmable gate array (FPGA). An example design is deployed on a DBE with Xilinx Virtex-6 FPGA and an ARM processor. The performance is verified by X-band observations of Mars Express (MEX) and New Horizons (NH). Under an integration time of 1 s, HIPE only takes 2.2 ms to process single-channel baseband data and provides frequency accuracies of 7 mHz and 30 mHz for the tested MEX and NH data. HIPE is implemented inside DBE, so the extra computer is no longer required and the pressure of data transmission or storage is greatly relieved. It could easily be extended to parallel multi-channel, real-time processing and would be a powerful method for Doppler measurement in deep space exploration missions, such as the Chinese mission to Mars to be undertaken by 2020.

A modern pulsar survey generates a large number of pulsar candidates. Filtering these pulsar candidates in a large astronomical dataset is an important step towards discovering new pulsars. In this paper, a novel adaptive boosting algorithm based on deep self normalized neural network (Adaboost-DSNN) is proposed to accurately classify pulsar and non-pulsar signals. To train the proposed method on a highly-imbalanced dataset, the Synthetic Minority Oversampling Technique (SMOTE) was initially employed for balancing the dataset. Then, a deep ensemble network combined with a deep self-normalized neural network and adaptive boosting was developed to train and learn the processed pulsar data. The design of the proposed Adaboost-DSNN method significantly reduced the computational time when dealing with large astronomical datasets, while also improving the classification performance. The scaled exponential liner units (SELU) activation function was used to normalize the data. Considering their neighbor information and the special dropout technique (α-dropout), Adaboost-DSNN displayed good pulsar classification performance, while preserving the data properties across subsequent layers. The proposed Adaboost-DSNN method was tested on the High Time Resolution Universe Survey datasets (HTRU-1 and HTRU-2). According to experimental results, Adaboost-DSNN outperform other state-of-the-art methods with respect to training time and F1-score. The training time of the Adaboost-DSNN model is 10x times faster compared to other models of this kind.

In pulsar observation, dispersion occurs due to the interstellar medium. The dispersion significantly affects the detection of pulsar signals. To overcome the dispersion effect, incoherent dedispersion methods are often applied. The tranditional inchoherent dedispersion methods are computationally expensive and troublesome. To deal with this problem, in this paper, we developed a Real-Time, Pipelined Incoherent Dedispersion Method (RT-PIDM). RT-PIMD only caches the summed-up time series, instead of all the frequency spectra, so the memory consumption is determined by the number of DM trails, whereas the traditional method’s memory consumption is determined by the number of frequency channels. In most of the situations, the number of frequency channels is several times more than that of DM trails, which means the memory consumption of traditional methods is more than that of RT-PIDM. With RT-PIDM, we designed a 1.2 GHz bandwidth prototype digital backend, and we finished pulsar observation with the 40 m radio telescope at Yunnan Observatory. The results demonstrate that the RT-PIDM can be implemented inside a single FPGA chip with less Block RAM, and the proposed RT-PIDM dedisperses the pulsar signal in real time and achieves the same result as compared to traditional incoherent dedispersion.

The detection of pulsar signals is a highly intensive task. Numerous artificial intelligence (AI) and machine learning techniques (ML) have been proposed to classify pulsar and non-pulsar signals. While existing machine learning techniques improve classification efficiency, these methods are limited when it comes to dealing with large volumes of astronomical data, the extreme problem of class imbalance , and the polarization of high recall and precision. In this paper, to accurately classify pulsar and non-pulsar signals, extreme Gradient Boosting (XgBoost), and Light Gradient Boosting Machine (LightGBM) algorithms based on an asymmetric undersampling method are proposed. Firstly, the proposed model uses an asymmetric undersampling technique that divides the benchmark datasets into 75 subsets for the LOFAR Tied-Array All-Sky Survey (LOTAAS-1) and 9 subsets for the High Time Resolution Universe Pulsar Survey (HTRU2), eluding the class imbalance problem. Finally, XgBoost and LightGBM algorithms have been adopted for each new data subset, with a majority voting classifier being used to integrate the output of proposed models against each subset. Results of the proposed method were compared to state-of-the-art baseline models, which showed that our models significantly performed better than existing methods and gained almost 2% and 2.5% better F-score and precision results, respectively.

Channelization is one of the most important parts in a Digital Back-End(DBE) for radio astronomy. A DBE with wider bandwidth and higher resolution consumes larger amount of computing and memory resources, which results in much higher hardware cost. This paper presents an efficient channelization architecture, which consists of Bit-Inverted, Parallel Complex Fast Fourier Transform(BIPC-FFT) and In-place Forward-Backward Decomposition(IPFBD). The efficient architecture can assist with saving a lot of resources, so a wide-band and high-resolution DBE can be implemented on an resource restricted platform. Based on the efficient channelization architecture, we designed a Dual-Input, 64K-Channelized prototype DBE with 1.2 GHz bandwidth on a Xilinx Virtex-6 LX240T Field Programmable Gate Array(FPGA) chip. The test results in the lab and observation results at Yunnan Observatory demonstrate the DBE can be used for radio astronomy.