高效多通道生理參數電路和跌倒偵測系統

Design of Efficient Multichannel Physiological Parameters Circuit and a Fall Detection System

張元馨(Yuan-Shing Chang)

指導教授 ： 黃克穠

義守大學/電機資訊學院/電子工程學系/博士(2018年)

摘要

近年來生理信號偵測系統極為流行,現有的生理信號數據多通道記錄系統主要使用模擬前端，需要數量等於通道數，導致運動受限，可擴展性差。這些系統中的大多數使用多級串接的前置放大器和前端濾波來在傳輸到遠程數據記錄器之前提高信噪比(SNR)。但是目前基於模擬用的系列產品的規模較小，且由於多通道生物信號輸入相當複雜且不易測量,通道容易受到糸統內及外界的噪音干擾(Noise interference), 而影響到信號判別的準確性。由於現階段電子電路均以高速傳輸,在外界之雜訊源干擾下,極易產生高振輻的雜訊突波輸出 ,嚴重者甚至改變下一級電路之輸入邏輯準位而產生電路之誤動作,為了改善生理信號偵測器及信號可能接收到外界雜訊及干擾電路信號,本電路設計的目的即為針對此種雜訊干擾源所衍生的效應加以濾除。另外為了提高監護系統中的便攜性以及無線傳輸能力。對便攜式設備的尺寸，重量和構造的嚴格限制極大地限制了其可用的機載電池容量，而多通道生物醫學數據的無線傳輸加劇了這些固有的功率隔離的便攜式設備中的能量問題。大部分的功耗在無線傳輸過程中消耗，通過壓縮將數據量降到最低對於減少系統總能量消耗至關重要，從而延長設備的自主性和電池的使用時間。
基於對這些問題的改善,我們提出了一種以可穿戴式頭帶和耳塞形式的多模式生理信號監測系統，可以用於監測腦電圖（EEG），血腦屏障（HEG）和心率變異性（HRV），為了獲得更好的輸入無噪信號，進行準確的生物信號監測以及減少輸出數據的存儲和傳輸,我們首先設計了一種在監測系統前端嵌入式低功耗尖峰濾波器的改進系統。可以通過數字選擇信號動態地預定要去除的尖峰的邊界， 對於尖峰濾波器，我們設計了一個低功耗的CMOS濾波器，可以去除輸入信號上的雜訊突波。另外在提高傳輸數據效率方面，為了有效地壓縮動態生物信號，我們選擇MPEG-4 ALS標準作為生物信號的標準無損壓縮工具，利用信號通道間相關性的結合信號通道編碼工具的多聲道信號來提高多通道信號資訊流的壓縮率。並且使用了以FPGA電路來實現此改善的數據壓縮系統。另外為了完整的擷取人體動態的生物活動信號，進而增加監測系統的安全功能性,我們加上了跌倒偵測系統(FDA)，我們選擇了一個低功耗和高集成度的器件ADXL345實現這個系統，它可以滿足跌倒偵測和警報的需求。最後，我們可以通過行動裝置系統由手機藍芽和Wi-Fi網絡作為傳輸通道, 也可以在雲端或遠程觀察如EEG / HEG / HRV健康指數和FDA系統分析儀來監測多種生物生理醫學信號。
我們期望該改進的生物醫學信號量測系統被設計為可重構性晶片，為具有2通道EEG傳感器前端，2通道NIRS傳感器前端和12-b ADC。加上改進的數據傳輸系統，可以消除輸入信號的正負峰值，對輸出數據的壓縮比率約為15〜70％，同時還我們提出的FDA警報系統從而增加了老年人的安全,降低了事故或傷害的風險。

關鍵字

none

並列摘要

In recent years, biological signal detection systems have become extremely popular. Existing biological signal data multi-channel recording systems mainly use analog front-ends, and the number of channels required is equal to the number of channels, resulting in limited movement and poor scalability. Most of these systems use multi-stage serial preamplifiers and front-end filtering to increase the signal-to-noise ratio (SNR) before transmission to a remote data logger. However, the current series of products based on simulation are small in scale, and because the input of multi-channel biological signals is quite complicated and difficult to measure, the channels are susceptible to noise interference within the system and the outside world., And affect the accuracy of signal transmission. Due to the high-speed transmission of circuits atthe present stage, under the interference of external noise sources, it is easy to generate high-spike noise surge. In severe cases, even if the input logic level of the next-level circuit is changed, a circuit malfunction occurs. In order to improve the biological signal detector and the signal may receive external noise and interference circuit signal, the purpose of this circuit design is to filter out the effects derived from such noise interference sources. In addition, to improve the portability and wireless transmission capabilities in the monitoring system. About the portable device. The severe limitations of weight and construction greatly limit available on-board battery capacity, and the wireless transmission of multi-channel biomedical data exacerbates the energy problems in these inherently power-isolated portable devices. Most of the power consumption is consumed in the wireless transmission process, and minimizing the amount of data through compression is essential to reduce the total energy consumption of the system, thereby extending the autonomy of the device and the battery life.
Based on these improvements, we propose a multi-modal biosignal monitoring system in the form of wearable headbands and earbuds that can be used to monitor electroencephalogram (EEG), blood-brain barrier (HEG), and heart rate variability (HRV) In order to obtain better input noise-free signals, perform accurate biological signal monitoring, and reduce the storage and transmission of output data, we first designed an improved system that adds a low-power spike filter to the front end of the monitoring system. The boundaries of the spikes to be removed can be dynamically predetermined by the digital selection signal, for spike filters, we designed a low-power CMOS filter that removes noise spikes on the input signal. In addition, to improve the efficiency of transmission data and effectively compress dynamic biological signals, we chose the MPEG-4 ALS standard as a lossless compression tool for biological signals, and use signal channel correlation to combine multi-channel signals from signal channel coding tools. Improve the compression rate of multi-channel signal information flow. Which using the FPGA circuit to achieve this improved data compression system. In addition, to completely capture the dynamic human biological activity signals, the safety function of the monitoring system is further increased. We added the fall detection and alert system (FDA). We chose a low-power and highly integrated device, the ADXL345, to implement this system, which can meet the needs of fall detection and alarm. Finally, we use Bluetooth and Wi-Fi networks as the transmission channels through the mobile device system, and we also monitor various biomedical signals in the cloud or by remote observation such as the EEG/HEG/HRV Health Index and the FDA System Analyzer.
We implement this improved biomedical signal measurement system to be designed as a reconfigurable system with a 2-channel EEG sensor front end, a 2-channel NIRS sensor front end, and a 12-b ADC. With the improved data transmission system, the positive and negative peaks of the input signal can be eliminated, and the compression ratio of the output data is about 2.3. At the same time, our proposed FDA system increases the safety of the elderly and reduces accidents or injuries risk.

並列關鍵字

none

參考文獻

[1] H. U. Lee, Y. Kim, H. Roh, T. Bae, J. Kim and C. Yoo, “A Wearable EEG-HEG-HRV Multimodal System With Simultaneous Monitoring of tES for Mental Health Management”, IEEE Transactions on Biomedical Circuits and Systems, vol. 9, no. 6, pp. 758-766, Dec. 2015

[2] D. S. Kim, J. S. Kwon, “Development of integrated textile fabrics flexible platform monitoring system for safety and ease life”, wrote the dissertation. was supported by the IT R&D program of MKE/KCC/KEIT　　

[3] H. Abdullah, G. Holland, I. Cosic and D. Cvetkovic, “Correlation of sleep EEG frequency bands and heart rate variability,” Engineering in Medicine and Biology Society 2009 (EMBC 2009), Annual International Conference of the IEEE, pp.5014-5017, 3-6 Sept. 2009.

[4] C. Fidopiastis and C. Hughes, “Workshop 1: Use of psychophysiological measures in virtual rehabilitation”, Virtual Rehabilitation 2008, pp. 25-27 Aug. 2008.

[5] W. C. Chen, C. K. Chua, E. Fu, C. C. Tseng and S. Y. Kang, “A low power biomedical signal processing system-on-chip design for portable brain-heart monitoring systems”, International Conference on Green Circuits and Systems (ICGCS 2010) , pp.18-23, 21-23 June, 2010.