Conceitos essenciais
A two-stage communication protocol is proposed to estimate the number of active devices and dynamically allocate the preamble length for stable identity detection performance in massive machine-type communications.
Resumo
The paper addresses the problem of ensuring stable identity detection performance in the current grant-free protocol for massive machine-type communications (mMTC). In the existing grant-free protocol, the base station (BS) blindly allocates a fixed length of preamble for identity detection, regardless of the dynamic number of active devices (K). This can lead to degraded identity detection performance when K is large.
To solve this issue, the authors propose a two-stage communication protocol:
Phase I: Estimation of the number of active devices K
- Devices are allocated a small number of preamble symbols (LI) to estimate K at the BS.
- The estimated K is then used to dynamically allocate the preamble length (LII) for identity detection in Phase II through a table lookup approach.
Phase II: Detection of identities of active devices
- Devices are allocated the dynamically determined preamble length (LII) to report their activity to the BS.
- The authors also propose an efficient algorithm to reduce the computational complexity of the identity detector by exploiting the estimated K.
Numerical results demonstrate that the proposed two-stage protocol can achieve stable identity detection performance, even when the number of active devices K varies significantly. Compared to the existing grant-free protocol, the proposed approach provides better detection performance and lower computational complexity.
Estatísticas
The number of active devices K varies dynamically over time, and maintaining a fixed preamble length LII can significantly compromise identity detection performance when K is large.
The total preamble length L = LI + LII, where LI is used for estimating K in Phase I and LII is used for identity detection in Phase II.
Citações
"For stable identity detection performance, is it enough to permit active devices to transmit preambles without any handshaking with the base station (BS)?"
"Maintaining a fixed LII at the BS significantly compromises identity detection performance when K is large."