Periodic Reporting for period 2 - NEMF21 (Noisy Electromagnetic Fields - A Technological Platform for Chip-to-Chip Communication in the 21st Century)
Berichtszeitraum: 2016-10-01 bis 2018-09-30
To be continuously connected via internet all day long has become a normal part of life. Because of this, data rates have increased enormously over the last decade. In the near future, this is expected to increase even more. For this reason, high performance integrated circuits (IC) with multi-functionality are being developed, and the number of functions nowadays integrated on IC’s (chips) smaller than a finger nail is staggering. The focus in the past was mainly on increasing the integration level of chips (increasing the number of functions) and not so much on the communication between chips. Although the data exchange on a chip can be extremely high, these data have at some time also to be exchanged with other components (chips). Chips are normally mounted on a Printed Circuit Board called PCB. These PCB’s contain an assembly of circuits, arranged in different layers, connecting the various chips together. All chips are hardwired together. Such configurations are not reconfigurable and cannot sustain dynamic scenarios like the introduction of new PCB’s or new chips into the system, or creating new data links between already mounted chips. Hence, such a configuration is not well prepared for the wireless world of today and tomorrow. A topology where chips can wirelessly communicate with each other like a mobile in a WLAN, would be far more suitable for our future needs and provide a complete agile and reconfigurable platform. This would greatly improve the capabilities of the communication platforms of the future telecommunication infrastructure and result in not only a better quality of connection (quality of service) but also higher data throughput at lower costs. This will in the end lead to a better quality and quicker access to the internet of things, for example, and similar services, and give people an increased information flexibility.
NEMF21 focuses on the specific problem of how to model from an ElectroMagnetic (EM) point of view, such an environment: chips mounted on a PCB communicating wirelessly (Chip-to-Chip: C2C). Until now, little investigations have been performed regarding this aspect. This does not only include the mathematical modelling of such complex EM scenarios but also the measurement technologies involved. In order to develop a correct model, one also has to measure the EM-fields propagating around/over a PCB. C2C-communication is complicated further by the fact that in the real world the communication is always jammed (polluted) by other unwanted EM-signals, also called noise. This kind of noise can be caused by PCB power lines, other communicating components on chips, reflections, and external wireless devices, just to list a few examples.
The following points are therefore addressed within NEMF21:
o Mathematical modelling of an EM noisy environment with respect to maximizing the data throughput.
o Measuring EM-fields propagating across a PCB (and between PCB’s).
o Designing & manufacturing antennas and antenna arrays for C2C-scenarios to optimise data throughput and minimising the influence of noise .
o Extending commercial EM field solving tools with noisy EM-signals supported by measurements tools.
These investigations aim at providing a new tool box and design guide lines for developing future state-of-the-art C2C PCB’s: the guidelines will give the initial blue print for i) where to mount the chips on the PCB and ii) with which antennas arrangement to work with in order to achieve maximum connectivity between chips. The tool box provides the means to design the PCB in detail accordingly.
NEMF21 focuses on the specific problem of how to model from an ElectroMagnetic (EM) point of view, such an environment: chips mounted on a PCB communicating wirelessly (Chip-to-Chip: C2C). Until now, little investigations have been performed regarding this aspect. This does not only include the mathematical modelling of such complex EM scenarios but also the measurement technologies involved. In order to develop a correct model, one also has to measure the EM-fields propagating around/over a PCB. C2C-communication is complicated further by the fact that in the real world the communication is always jammed (polluted) by other unwanted EM-signals, also called noise. This kind of noise can be caused by PCB power lines, other communicating components on chips, reflections, and external wireless devices, just to list a few examples.
The following points are therefore addressed within NEMF21:
o Mathematical modelling of an EM noisy environment with respect to maximizing the data throughput.
o Measuring EM-fields propagating across a PCB (and between PCB’s).
o Designing & manufacturing antennas and antenna arrays for C2C-scenarios to optimise data throughput and minimising the influence of noise .
o Extending commercial EM field solving tools with noisy EM-signals supported by measurements tools.
These investigations aim at providing a new tool box and design guide lines for developing future state-of-the-art C2C PCB’s: the guidelines will give the initial blue print for i) where to mount the chips on the PCB and ii) with which antennas arrangement to work with in order to achieve maximum connectivity between chips. The tool box provides the means to design the PCB in detail accordingly.
Over the first year of NEMF21, initial investigations have been conducted with respect to mathematical modelling of a noisy EM-environment, how to measure the performances of C2C-communication, and designing special purpose C2C-antennas. First scenarios of different communication setups - so-called benchmarks - have been defined. This includes the design of PCB’s and corresponding antenna topologies and arrangements. A special focus was put on studying wireles C2C communication in the near-field of antennas, an area where original research is necessary. We established an interesting connection with multiple-input multiple-output (MIMO) communications, typically studied in the far-field regime. MIMO performance metrics involve the calculation of channel capacity, related to the number of effective communication channels etsablished between transmitting and receiving arrays of antennas. Theoretical models supported by numerical simulations and measurement show the possibility of near-field communication channels scaling linearly with the number of antennas used. First tests on beam-steering have been carried out and the necessary experimental equipment has been set up. Measurements for obtainig data for measuring noisy EM-fields have been refined and further automised. Likewise, a series of techniques for numerically computing noisy EM-fields both in the near- and far-field are being developed and tested and will be vital for setting up C2C design plans. Particular emphasis is here on including multi-path interference from the environment - through first principles involving electromagnetic simulations - besides system noise generated within the PCBs. Mathematical models have been compiled and measurements at different frequency ranges of several antenna configurations have been performed. After the first twelve months, we have gathered a range of data and have gained enormous experience, experience which make us fit to deliver the breakthroughs necessary over the coming 24 months of the project.
The concepts formulated within the first year of NEMF21 are suggesting new paradigms for small scale wireless communications, with potential benefits
in high throughput wireless sensor networks as well as in any other wireless network architectures operating in the presence of a harsh medium.
Understanding how to configure and perform a multi antenna communication within complex environments will certainly pave the way for a better control of EM wave fields. The ability of allocating physical communication channels dynamically through coding schemes informed by the space-time physics of propagation will lead to potential implications in the reduction of unintended environmental interferences and thus mitigation of human exposure.
Presently, it is too early to speculate about potential commercial or socio-economic impact; we are looking forward to present this at a later stage of the project in more details.
in high throughput wireless sensor networks as well as in any other wireless network architectures operating in the presence of a harsh medium.
Understanding how to configure and perform a multi antenna communication within complex environments will certainly pave the way for a better control of EM wave fields. The ability of allocating physical communication channels dynamically through coding schemes informed by the space-time physics of propagation will lead to potential implications in the reduction of unintended environmental interferences and thus mitigation of human exposure.
Presently, it is too early to speculate about potential commercial or socio-economic impact; we are looking forward to present this at a later stage of the project in more details.