Mobile Communications SystemsMobile Communications Systems

5th Generation Non-Orthogonal Waveforms for Asynchronous Signalling

0

5GNOW: Sparse Message Delivery in PRACH

References

Pulse Shaped PRACH

Signalling in Physical Uplink Shared Channel (PUSCH) and the

Physical Random Access Channel (PRACH)

Orthogonal design principle to minimize the interference between the

channels, several subcarriers on both edges of PRACH are left zeros as

a guard band

Preamble generation: The

preamble is constructed by

Zadoff Chu sequence

[1] P. Jung and G. Wunder, “The WSSUS Pulse Design Problem in Multicarrier

Communications”, IEEE Transactions on Communications, vol. 55, no. 10, pp.

1918-1928

[2] P. Jung and G. Wunder, “On Time-Variant Distortions in Multicarrier

Transmission with Application to Frequency Offsets and Phase Noise”, IEEE

Transactions on Communications, September 2005, vol. 53, no. 9, pp. 1561-

1570

LTE/LTE-A PRACH and PUSCH

Sporadic Traffic

5GNOW adresses sporadic traffic as an enabler of the

‚Internet of Things‘.

Sources for sporadic traffic

Smartphone short message delivery

Machine-type communications (MTC)

Approach:

Waveform design, sparse signal processing

Goal:

Reduce the uplink overhead to 5%, improve the energy-efficiency of

MTC uplink data transmission by a factor of 20

Conclusion

Proof-of-concept: we have shown that short messages can be

signalled over PRACH

Goal in current setting only achievable with advanced signal

processing and tailored waveforms.

Some properties of optimal signalling derived

Simulation Results

5GNOW PRACH: asynchronous non-orthogonal multi-user

access under sparsity constraints.

Trinity of (1) link aquistion, (2) channel estimation, and (3)

demodulation must be achieved at the same time with good

performance

Approach

Biorthogonal Frequency Division Multiplexing (BFDM) scheme for

transmitting the preamble in PRACH

Set of shifted pulses on time-frequency lattice points, transmit/receive

pulses form biorthogonal Riesz bases

Joint acquisition, channel estimation, and demodulation by iterative

multi-user sparse detection

Waveform Design

Based upon 𝑳𝟏-norm optimality property

Time and frequency localization based on grid matching rule

Verification by simulation whether

utilizing the PRACH guard bands

with messages is really feasible

Parameters according to LTE

specification as given in Table I

Initial Waveform Results

Classical localization property is the uncertainty

principle: Gaussian signalling is optimal

In our approach waveform is designed to

minimize 𝑔 1 s.t. 𝑔 2 = 1, for a waveform

𝑔 ∈ 𝒫𝒲𝜋1 (Paley-Wiener space of integrable

bandlimited signals with normalized bandwidth

𝜋) which depends on bandwidth extension

factor 𝐿 ≥ 1

However: Gaussian

signalling is not

necessarily optimal

in a specific setting.

L 1 1.2 1.4 1.6

𝒈 𝟏 ∞ 2.15 1.75 1.6

Waveform design properties: