Search

Objectives: Based on the electronic medical record data, this paper analyzes the law of TCM treatment of low back pain, and provides reference for TCM treatment of low back pain. Technology or Method: The whole process records of 360 patients with low back pain from January 1, 2015 to August 27, 2019 were obtained from the highly structured electronic medical record system of Mianyang Hospital of Traditional Chinese Medicine, and 337 prescription records were selected from them. Data mining technology was used to explore and analyze the law of prescription medication using the Apriori algorithm. RESULTS: Among the 337 prescriptions, a total of 257 herbs were involved, and the top 5 herbs were licorice root, Ligusticum Chuanxiong, twotoothed achyranthes root, eucommia bark, and Poria Cocos. In the four nature of drugs, the main warm, calm; In the five flavors, to sweet, pungent, bitter; In terms of channel tropism, liver, spleen, lung, kidney, stomach and heart are the main factors. Clinical or Biological Impact: In the treatment of lumbago, TCM focuses on dispelling wind and dampness, invigorating spleen and replenishing qi, promoting blood circulation for removing blood stasis, and always aims to “strengthen the kidney and strengthen the spleen”. The combination of drugs with strong correlation provides a reliable basis for the use of drugs for clinical syndrome differentiation. The conclusions obtained still need clinical testing, but it can provide some reference value for clinical practitioners to further explore.

When an electricity meter is calibrated under the condition of non-reference temperature, it is generally necessary to correct the verification result according to the actual temperature of the verification environment, and then judge whether the electricity meter is qualified. Under the mode of large-scale automatic verification, the space of workplace can reach thousands of square meters, and the temperature distribution is uneven. In order to reduce the cost of controlling verification environment temperature, the effort of trying to broaden the range of verification environment temperature has been always made by researchers. The problem is that it is difficult to make an effective judgment on the batch verification results based on the verification data. Under this circumstance, a data analysis method which helps to ensure weather the verification results are effective is needed. In this paper, the batch electricity meters verification experiment is designed and carried out firstly, then the distribution characteristics of verification data are studied, and the conclusion is obtained that the distribution type of verification data is normal distribution. On this basis, the method based on parameter test is used to analyze the data consistency, which shows that this data analysis method can make auxiliary evaluation on the validity of verification data produced in non-reference temperature environment. The data analysis method described in this paper can be applied to the operation state evaluation of large-scale automatic verification system and can also provide data analysis means for broadening the range of environment temperature for large-scale automatic verification.

With the rise of artificial intelligence, automatic driving, 5G, Internet of things and other emerging industries, the demand of 3D integrated packaging technology is increasing strongly to meet the ever-increasing market demands of high performance, small size, high reliability and ultra-low power consumption. This paper presents a new 3D system integrated packaging technology named embedded system in chip (eSinC®) technology. This technology is a combination of TSV technology and eSiFO technology which has been reported in 2016 [1]-[4]. After finishing the standard eSiFO process, backside RDLs and via last TSV process were fabricated by using laser temporary bonding technology. A totally three stacking package including five chips with two layer frontside and backside RDLs was fabricated successfully. Individual package was connected by micro bumps and TSVs. The packaging size is 5×5mm with an overall 0.78mm packaging thickness, while the individual eSinC packaging thickness is 0.28mm. Several key technologies were developed to achieve eSinC package, including high aspect ratio TSVs, wafer thin and handling, low temperature PECVD process, as well as temporary bonding. Good electrical yield was achieved after process optimization.

Data Framing Attacks (DFA) have been proposed in recent years, and have attracted much interest in the research community on smart grid security. DFA compromises Bad Data Identification and Removal (BDIR, leading BDIR to remove secure data, which will result in incorrect state estimation. Therefore, successful detection of DFA is significantly important for the control and operation of the power grid. This paper presents a study on the utilization of machine learning to detect DFA. Since the detection problem can be formulated as a classification problem between secure measurements and attacked measurements, a mature machine learning technology, Support Vector Machine (SVM) is chosen in this work. The proposed method is examined on the 118-bus IEEE test system. The experimental analyses indicate that SVM can detect DFA with good performance.

The demand for miniaturized package size, higher performance and integration density, lower power consumption, and lower manufacturing cost drives the development of various new packaging technologies, such as WLP, 2.5D interposer, 3D WLCSP etc. Recently, Fan-Out Wafer Level Package (FOWLP) has aroused more and more interest since it emerged as a successful technology in providing the solution to meet advanced system packaging and integration requirements. FOWLP has also become a key-enabling technology for multi-chip and 3D system integration. Embedded silicon fan-out (eSiFO®) technology developed by Huatian group was a low cost FOWLP technology, which eliminates molding, temporary bonding and de-bonding in the process flow. The eSiFO® technology for single chip with a high yield of 99.5% was reported. In this paper, the development of eSiFO® technology for multi-chip and 3D system integration was reported. One package in the size of 5.2×4.0 mm2 and containing 5 chips has been successfully produced. Small warpage of ~1mm for eSiFO® 300mm wafer was achieved even with 2 layer thick RDLs. Another 3D-Integration package based on the eSiFO® technology has also been successfully demonstrated, where two dies were stacked on the top of the eSiFO® wafer by die placement and mass reflow processes. For the 3D eSiFO® SiP package, the size and thickness was 4.1×4.1mm, and 0.375 mm respectively. One RDL was formed on both front and backside of the eSiFO®. The interconnection between the embedded chip and the stacked chips was made through micro bumps and through-silicon vias (TSVs) located in the fan-out area of the carrier wafer. The manufactured packages have shown good electrical yield and passed standard reliability tests.

The accurate knowledge of relative position and attitude is essential to reduce control error and avoid collision in small satellite formation flying. However, the traditional relative navigation systems composed of many sensors have usually large volume and high cost. Considering the miniaturization and highly integrated functional modules of small satellites, cost-effective relative position and attitude estimation method is required. In this paper, we propose a novel approach, called Relative States Estimation based on Double Antenna arrays (RSEDA). The approach utilizes the existing communication systems on flying vehicles. In RSEDA, the chief vehicle is equipped with two planar antenna arrays to receive the probing signal transmitted by the planar antenna array installed on the deputy vehicle. Thus two sets of angle-of-arrival (AoA) and angle-of-departure (AoD) of the line-of-sight (LOS) signal propagation path from transmitting antenna to the two receiving antennas can respectively be estimated using the array signal processing. The distance between the two receiving antenna in the frame of the chief body frame is to be known. Then based on this distance and the measured AoAs and AoDs of the LOS paths from the transmitting antenna to the two receiving antennas, the relative distance between the chief and deputy vehicles can be deduced. Then the observation equation of the EKF can be derived. Based on which, the relative position and attitude between the CV and DVs can be estimated iteratively. RSEDA is theoretically proved and also validated by the simulation results in this paper. The signal transceiver architecture for inter-vehicle communications and the RSEDA method are also presented for implementation in a real vehicle.