DocumentCode :
3733920
Title :
A 400-MHz wireless neural signal processing IC with 625? on-chip data reduction and reconfigurable BFSK/QPSK transmitter based on sequential injection locking
Author :
Kok-Hin Teng;Tong Wu;Zhi Yang;Chun-Huat Heng;Xiayun Liu
Author_Institution :
Department of Electrical and Computer Engineering, National University of Singapore, Singapore
fYear :
2015
Firstpage :
1
Lastpage :
4
Abstract :
An 8-channel wireless neural signal processing IC, which can perform real-time spike detection, alignment, feature extraction, and wireless data transmission is proposed. A reconfigurable BFSK/QPSK TX at MICS band is incorporated to support different data rate requirement. By using Exponential Component-Polynomial Component (EC-PC) spike processing unit with incremental principal component analysis (IPCA) engine, an overall 625× data reduction is achieved. The EC-PC unit is capable of detecting neural spikes of poor SNR. In TX, dual channels at 401 MHz and 403.8 MHz are supported by applying fixed injection and sequential injection locked techniques while attaining phase noise of -102 dBc/Hz@100 kHz offset. The measured EVM of 4.60%/9.55% with PA output power of -15dBm is achieved for QPSK@8 Mbps and BFSK@12.5 kbps. Fabricated in 65 nm CMOS with an area occupation of 1 mm2, the design consumes a total current of 5~5.6 mA with maximum energy efficiency of 0.7 nJ/b.
Keywords :
"Wireless communication","Feature extraction","Phase shift keying","Injection-locked oscillators","Payloads","Transmitters"
Publisher :
ieee
Conference_Titel :
Solid-State Circuits Conference (A-SSCC), 2015 IEEE Asian
Type :
conf
DOI :
10.1109/ASSCC.2015.7387454
Filename :
7387454
Link To Document :
بازگشت