Mei Shunliang's research while affiliated with Tsinghua University and other places

In the wireless broadband network, most traffic are generated from indoor users. Meanwhile the indoor coverage is identified as the major bottleneck especially in dense urban environment with high-rising buildings. In this paper, the TD-LTE network downlink performance of indoor is studied with various network deployment options, including outdoor Macro cell, outdoor Micro cell, indoor distributed antenna system (DAS), indoor Femto cell, and their combinations in a realistic large-scale dense urban scenario. Since a DAS cell consists of several or even tens of access points (AP), the computation load is too high to be accepted when calculating the received signal power and the interference power by traditional widely-used methodology. Thus a new DAS model is proposed to decrease the computation complexity to a reasonable level in the simulation. To have more accurate channel modeling, the real three-dimensional (3D) city map of the investigated area is employed. A 3D ray-tracing tool is used to estimate the large scale propagation occurring between outdoor sites and buildings. The traffic offloading features of Micro, DAS, and Femto cell are evaluated through system simulations.

Sparse constraint Least Mean Square (LMS) is a recently proposed efficient adaptive algorithm for sparse system identification. However, its computational complexity is quite high especially when the filter length is long and convergence is slow for colored input signal. This paper extends the idea of sparse constraint into multidelay frequency adaptive filter (MDF) algorithm and proposes the sparse MDF algorithm for echo cancellation. The proposed algorithm perserves both the advantage of sparse LMS which has fast convergence performance for sparse system and MDF algorithm which has temporal decorrelation effect with lower implementation complexity. Two typical sparse constraints, l1-norm and an approximate l0-norm, are employed. And their performances of various aspects are simulated. Experiments show they have better performance than existing algorithms.