RINGS: Wideband NextG Tb/s mm-Wave Communication and Networking

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

The proposed sixth generation (6G) of wireless connectivity may involve much higher frequencies and bandwidths than current 5G systems, operating at approximately ten times higher frequency and offering up to ten times higher bandwidth, which will enable increased communication speed, sorely needed by the growing computational complexity of artificial intelligence and machine learning networks. But operating at higher frequencies has many shortcomings, namely higher loss in the channel, and more opportunity for blockages and other impairments to the connection. Additionally, the extremely high frequency of the data modulation requires novel signal processing techniques due to the infeasibility and energy expense of digitizing signals at such high speeds. To overcome these limits, a proposed analog approach that divide and conquer the frequency bands to save power and to make the systems feasible for mobile applications will be pursued. Taking advantage of the fact that transmission is accomplished with a large array of antennas, rather than a single high-power transmitter, can lead to energy savings and interference reduction by dividing the transmissions to various antennas to cause less inter-user interference. Finally, the resilience of such communication systems can be improved by incorporating more dense networks, which can form so-called mesh networks, which allows multiple paths from source to destination, which is needed when one path is lost due to blockages or other impairments. Such mesh networks also potentially increase interference and noise in the network, the directional nature of the transmissions can be utilized to realize a hierarchical mesh system. To test these ideas, a mm-wave testbed will be upgraded to support these experiments. The proposed goal is to increase the resilience of mm-wave networks both at the circuit and system level for 6G/NextG networks. At the circuit level, high throughput can be achieved by utilizing multiple carriers and efficient architectures for circuit building blocks that can simultaneously modulate/demodulate a large swath of bandwidth over an aggregation of non-contiguous sub-bands, similar to OFDM and OFDM-A, but using mixed-signal and analog techniques to reduce requirements on ADCs, DACs, and DSP. The main innovations are in increasing the bandwidth in the circuitry and exploiting system level innovations at the circuit architecture level to improve performance metrics, in particular noise, impact of phase noise, quantization noise, and output power and efficiency. The techniques proposed can help alleviate path loss, signal blockage, and beam tracking with minimal energy consumption. The benefits will be most pronounced when building large-scale NextG MIMO beamforming systems. Our systems research vector focuses on building robust and rapidly adaptable NextG networks by utilization of spatial mesh networking. By utilizing more degrees of freedom, in particular a wider channel bandwidth, and multiple carriers, the network can be further optimized for reliability and resilience. Frequency dependent beamforming, frequency selective fading, and many other channel impairments can be handled more easily in the frequency domain. However, OFDM modulation of a 30 GHz wide-band channel is not practical, instead mixed-signal techniques that can garner most of the benefits of a fully digital solution will be explored. Interference aware array transmitters will minimize inter-user interference by exploiting redundancies in the array. The proposed ideas will be tested both with integrated circuit prototypes and by demonstration using an upgraded mm-wave (E-band) MIMO testbed. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.