|Robert Bosch Centre for Cyber-Physical Systems
Electrical Communication Engineering
|Yogesh Simmhan||Assistant Professor||Computational and Data Sciences|
|Rajesh Sundaresan||Professor||Robert Bosch Centre for Cyber-Physical Systems
Electrical Communication Engineering
|Malati Hegde||Principal Scientific Officer||Electrical Communication Engineering|
The Indian Government has identified Smart Cities as a priority area for development in the coming years. ICT will play a key role in supporting smart city solutions and more specifically, IoT technologies will be a key enabler for providing the “smarts”. The requirements from citizens in cities is diverse and cuts across many different verticals like transportation, water management, solid waste management, smart parking etc. Typically, each vertical will be addressed by a different vendor, who will provide an end-to-end solution. However it has been recognized of late, that a better approach might be to have a horizontal approach where in sensors, and other data are made available across different silos – in order to foster new, cost effective solutions to various city related problems and citizen needs. A simple example is that of a camera sensor which can aid in not only in surveillance but also in crowd management, smart parking, transit operations management etc. applications. Hence there is a need to develop smart city ICT/IoT framework as a generic platform that will support a diverse set of applications.
The main thrust of this work will be to develop an instrumented test bed in the IISc campus, which will consist of a collection of sensors, gateways, middleware and server side software, supporting a few candidate applications of energy monitoring, water monitoring and solid waste management. The testbed will be a hybrid of real and virtual components. Being able to support such a hybrid system will enable us to test concepts at scales of real cities with 1000s of gateways and 10s of thousands of sensors, while the real world portion of the testbed will still be small and low cost. The scaling limits will be determined via analysis of current and future requirements for Indian Smart Cities in terms of:
- Data base size, query latency, IO/second, availability
- Communication throughput, latency, loss tolerance
- Gateway performance (throughput), memory, storage,
- Sensor density per square kilometer and its data bandwidth
Finally the test bed will be validated via the physical portions instantiated in the IISc campus as well as the virtual portions instantiated to mimic a city of the scale of Bangalore. The concept of using light poles as smart city infrastructures will also be incorporated in the test bed via instrumenting about 10 light poles on the campus, with the gateways and sensors.
The aim of the test bed will be two-fold:
Allow for experimentation of different technologies at each level of the smart city solutions like:
- Different sensor hardware and firmware stacks, including crowd sourced sensing
- Different wireless front-haul schemes like WiFi, BLE, Sub-GHz
- Different gateway hardware and software stacks to support distributed analytics
- Different edge and cloud computing architectures
- Different data persistence platforms including in-memory DB, time-series DB, and NoSQL
- Different realtime and batch analytics platforms
Support for metrics at various levels from the sensors to the server, both in hardware and software to enable a quantitative assessment of the performance of various alternative approaches.
We believe that this activity will open up opportunities to explore new concepts like gateway virtualization, virtual agents, dynamic resource management, etc. in the second year onwards.