We were in Chongqing, China two weeks ago, overseeing the organization of CHINACOM 2016, 11th EAI International Conference on Communication and Networking in China, and took a moment to talk with Tay Wee Peng, Assistant Professor at the School of Electrical & Electronic Engineering at Nanyang Technological University. We were happy to welcome him as a keynote speaker at CHINACOM and pick his brains regarding research and innovation trends in China.
Could you summarize the scope of your current work and what you came to share with everyone at this event?
In an Internet of Things network, multiple sensors send information to a fusion center for it to infer about a public hypothesis of interest. However, the same sensor information may be used by the fusion center to make inferences of a private nature that the sensors wish to protect. To model this, we adopt a decentralized hypothesis testing framework with binary public and private hypotheses. Each sensor makes a private observation and utilizes a local sensor decision rule to summarize that observation before sending to the fusion center. We adopt a nonparametric approach to design local sensor decision rules that allow the fusion center to detect a public hypothesis with minimal regularized empirical risk, while keeping the empirical risk of detecting the private hypothesis above a privacy threshold. We develop iterative optimization algorithms to determine an appropriate privacy threshold and the best sensor local decision rules, and show that they converge. Numerical results on both synthetic and real data sets suggest that our proposed approach yields low error rates for inferring the public hypothesis, but high error rates for detecting the private hypothesis.
What do you see as the biggest challenge that communications and networking in China are currently facing?
China has done very well in recent years on the way to becoming a heavy weight in communications and networking research and innovations, as evidenced by the increasing number of journal and conference papers coming out of China. One challenge is that a lot of researchers tend to focus only on incremental improvements to existing ideas or systems. This could be due to publication or other types of pressure. To be on par with the US research community, more emphasis should be given to fundamental research that can generate new breakthroughs. To promote this, Chinacom may consider having special sessions on communication and signal processing theories. This will also encourage more PhD level students to submit papers, and alleviate the overall standard of the conference.
What would you say are the main trends in this area that are showing promise?
With emergence of new technology trends like social networks and IoT, there are many challenging research and engineering problems that need to be solved in order to bring such technologies to wide adoption. In particular, there is a lot of interest to enable “smart” living, learning and manufacturing, which relies heavily on IoT and social learning concepts.
To conclude, have you had an experience at this event that you found compelling, or which inspired you for future work?
I am impressed with the hardware and implementation facilities available to test and verify advanced theoretical concepts.